![2 1 Data and Energy Integrated Communication Networks: An Overview
sumer electronic devices are becoming shirt-pocket-sized and mobile. These devices
are normally powered by embedded batteries. However, as their functions become
ever more sophisticated, their thirst for abundant energy is not matched by the slow
progress of the batteries’ capacity. The situation in the communication industry is
even more daunting. Since the roll-out of the fifth-generation (5G) cellular system and
of the Internet of Things (IoT) is just around the corner, people’s appetite for super-
high data transmission rates, for high density of connectivity and for high mobilities
will indeed be satisfied to a large extent. A major portion of the future mobile data
traffic will be constituted by novel types of services, including high-definition stero-
scopic video streams, augmented/virtual reality, holographic tele-presence, cloud
desktops, as well as online games, etc. All these services require the user terminals
to be implemented with high computing capabilities for real-time signal process-
ing, which may quickly drain the embedded batteries. Furthermore, sensors will
be deployed in every corner of the future smart cities [1]. These sensors monitor
the environment and upload sensing results to central servers [2]. The life-span of
sensors and of sensing networks largely depends on the sensors’ battery capacity.
Regularly replacing the batteries may be an unrealistic or tedious task. Accordingly,
new sources of energy have to be explored to prolong the depletion period of con-
ventional batteries in order to relieve the energy concerns of various communication
devices.
1.2 Near-Field Wireless Energy Transfer
Nowadays, resonant inductive coupling [3] and magnetic resonance coupling [4]
have emerged for remotely charging electronic devices in the near-field. Resonant
inductive coupling based wireless charging relies on the magnetic coupling that
delivers electrical energy between two coils tuned to resonate at the same frequency.
Thistechniquehasalreadybeencommercialisedforsomehomeelectronicappliances
[5], such as mobile phones, electric toothbrushes and smart watches etc. However,
the coupling coils only support near-field wireless energy transfer (WET) over a
distance spanning from a few millimetres to a few centimetres [6], while achieving
a WET efficiency as high as 56.7%, when operating at a frequency of 508 kHz [7].
Furthermore, resonant inductive coupling requires strict alignment of the coupling
coils. Even a small misalignment may result in dramatic reduction of the WET
efficiency [8]. As a result, during the charging process, the electronic appliances
cannot be freely moved.
By contrast, magnetic resonance coupling [9] delivers electrical energy between
two resonators by exploiting evanescent-wave coupling. This technique has already
been adopted for charging the electric vehicles due to its high WET efficiency [10].
For example, magnetic resonance coupling is capable of achieving a WET efficiency
of 90% over a distance of 0.75 m [11]. Both its WET efficiency and its charging dis-
tance are much higher than that of the resonant inductive coupling. However, mag-
netic resonance coupling still belongs to the category of near-field wireless charging,
since its power transfer efficiency dramatically reduces to 30%, when the distance is
increased to 2.25 m [11]. Nonetheless, magnetic resonance coupling does not require](https://image.slidesharecdn.com/35569-250423081821-a8a4d60d/85/Data-and-Energy-Integrated-Communication-Networks-Jie-Hu-16-320.jpg)
![1.2 Near-Field Wireless Energy Transfer 3
strict alignment between the rechargeable device and the energy source. Hence, dur-
ing the charging process, the electronic appliances may be moved within the charging
area [12]. Furthermore, a multiple-input-multiple-output (MIMO) system, which has
already been widely adopted for improving the performance of the wireless commu-
nication, can also be introduced into the magnetic resonance coupling based WET
system for the sake of further enhancing the WET efficiency [13, 14].
1.3 RF Signal Based Wireless Energy Transfer
In contrast to the above-mentioned near-field WET techniques, the propagation of
the RF signals is capable of supporting far-field WET [15]. The history of the RF
signal based WET dates back to 1960, when the first long-distance WET system
was established by Brown [16, 17]. Brown jointly designed rectifiers and antennas
for energy receivers, which is now widely known as rectennas. They are capable of
efficiently converting the Alternating-Current (AC) energy carried by the RF signals
to Direct Current (DC) energy. This RF signal based WET system was validated
by remotely powering a model helicopter from the ground in 1964 [16, 17]. In the
1970s and 1980s, intense efforts were invested into the research of RF signal based
WET, which was largely motivated by the intention of developing a solar-powered
satellite [18, 19]. In this system, a satellite may harvest energy from sunlight in the
outer space and beam the energy back to ground stations via the propagation of RF
signals. Furthermore, the Jet Propulsion Laboratory of the National Aeronautics and
Space Administration (NASA) led a project from 1969 to 1975, in which 30 kW of
power was beamed over a distance of 1 mile at a 84% RF-DC efficiency [20].
There are three main technical challenges in the RF based WET. Firstly, the long-
distance propagation and adverse multipath fading may substantially attenuate the
RF signals before they arrive at the receivers, which inevitably results in energy
loss. Secondly, the energy carried by RF signals is of AC nature, which cannot be
directly invoked for driving an electronic load. As a result, the AC energy carried
by RF signals have to be converted to DC energy for any further use. However,
some portion of energy is inevitably lost during this conversion process. Last but
not the least, the diffraction of the RF signals’ waveform may expand the beam
size. As a result, the receive antenna having limited size is not capable of capturing
all the energy carried by the RF signals. For counteracting the signal attenuation
of wireless channels, the transmit beams have to be accurately aimed at the energy
receivers [21], which requires the joint design of the transmit and receive antennas.
For improving the AC-DC conversion efficiency, the receive antennas have to be
designed together with the rectifiers in order to achieve the impedance match for the
sake of high-efficiency AC-DC conversion [22]. For alleviating the adverse effect
of beam diffraction, the non-diffracted Bessel-Gaussian beam [23] can be invoked,
which is capable of efficiently reducing the energy loss during the propagation and
hence improve the WET efficiency over wireless channels.
In general, the RF signal based WET has the following advantages over its near-
field counterparts:](https://image.slidesharecdn.com/35569-250423081821-a8a4d60d/85/Data-and-Energy-Integrated-Communication-Networks-Jie-Hu-17-320.jpg)
![1.4 WET Versus WIT in the RF Spectral Band 5
1.4 WET Versus WIT in the RF Spectral Band
Since RF signal based WET techniques require highly fexible beam directivity in
order to satisfy diverse charging requests, the best spectral band for steering energy
beams is in the range of 10MHz to 100 GHz, which almost covers all the bands
allocated for wireless communication services. For example, TV/Radio broadcasting
services operate in the band spanning from 40MHz to 220 MHz [24], the mobile
cellular communication system operates in the spectral band spanning from 800MHz
to 3.7 GHz [25], while the WiFi communication system operates in the spectral
band spanning from 2.4GHz to 6 GHz [26]. Furthermore, as a key technique in
the upcoming 5G era, millimetre wave (mmW) [27] may significantly increase the
achievable throughput of the air interface, which operates in the spectral band ranging
from 10GHz to 100 GHz.
Although they both operate in the same RF band, WET and WIT still have the
following distinctive characteristics:
• They have different functional circuits. RF signals in the pass-band cannot be
directly invoked for both the information decoding and the energy harvesting.
For the information decoding, the RF signals in the pass-band have to be firstly
converted to the base-band, since all the signal processing has to be accomplished
in the base-band. By contrast, for the energy harvesting, the AC energy carried by
the RF signals has to be converted to the DC energy first, since only DC energy
can be stored in batteries or drive electronic loads. Specifically, during the AC-DC
conversion, the phase information carried by the RF signals is filtered.
• They require different absolute energy at receivers. The activation of the energy
harvesting circuits requires a relatively high energy carried by the received RF
signals, which is approximately on the order of −20 dBm. If the energy carried
by the received RF signal does not achieve the required activation threshold, none
of this energy can be harvested. By contrast, the successful information recovery
relies on the energy ratio between the received RF signal and the noise plus inter-
ference, not on the absolute energy carried by the received RF signal. As a result,
even a small amount of energy is capable of activating the information receiver,
which is approximately on the order of −80 dBm.
• They have different coverage. The RF signals are attenuated by hostile wireless
channels, such as the path loss, shadowing and multipath fading. Since the energy
harvesting requires a much higher absolute energy at the receivers than the infor-
mation decoding, the range of WET is accordingly much shorter than that of WIT.
Therefore, given the same set of transmitters and receivers, the resultant WET
network has a different topology with the WIT network.
• They treat noise and interference differently. The interference and noise ubiqui-
tously exist in any WIT system, which seriously impair the WIT performance.
Mitigating the performance degradation induced by the interference and noise is a
major challenge in the WIT system design. By contrast, WET systems may actu-
ally benefit from the interference and noise, since both of them are RF signals and
they both carry useful energy. The interference and noise can be jointly harvested](https://image.slidesharecdn.com/35569-250423081821-a8a4d60d/85/Data-and-Energy-Integrated-Communication-Networks-Jie-Hu-19-320.jpg)
![6 1 Data and Energy Integrated Communication Networks: An Overview
by the energy harvesting circuits, which may provide additional energy harvesting
gains for the energy requesters.
• They have different definitions in energy efficiency. The energy efficiency of WET
can be defined as the ratio of energy harvested by the receiver to the energy emitted
by the transmitter, which can be formulated as
ηW ET =
1
Pt
· ρ (Pr + PI + PN ) (Watt/Watt), (1.1)
where ρ is the conversion rate from the received RF energy to the DC energy by
considering a linear RF-DC converter. By contrast, In the community of green
communications, the energy efficiency of WIT is defined as the ratio of spectral
efficiency to energy consumption, which is evaluated in the unit of bps/Hz/Watt or
bps/Hz/Joule. By exploiting the classic Shannon-Hartley theorem in an Additive-
White-Gaussian-Noise (AWGN) channel, the energy efficiency of WIT can be
expressed as
ηW I T =
1
Pt
· log2
1 +
Pr
PI + PN
(bps/Hz/Watt), (1.2)
where Pt is the transmit power of the RF signal, Pr is the power received after
the signal being attenuated by the hostile wireless channel, PI is the aggregate
interference power and PN is the noise power at the receiver.
In Fig.1.1, we exemplify the energy efficiency of WET and that of WIT, which
can be calculated by (1.2) and (1.1), respectively. Observe from Fig.1.1a that in
our setting, the energy efficiency of WET reduces from 1.1% but converges to
1%, which is due to the channel attenuation incurred by the path loss between the
transmitter and receiver pair. Observe from Fig.1.1b that the energy efficiency of
WIT gradually reduces from 35 bps/Hz/mW to 0 as the transmit power of the RF
signal increases. By contrast, WET and WIT operating in the same RF spectral
band may compete for the precious resources in the air interface and they may
thus impair each other’s performance to some extent. For example, WET requires
that the RF signals carry a high power to the receivers for the efficient energy
harvesting. However, the high-power RF signals of the WET system may impose
excessive interference on the WIT receivers, which may thus significantly degrade
the WIT performance attained. As a result, coordinating WET and WIT in the same
RF band imposes critical challenges on the RF circuit design, on the transceiver
design of the physical layer, on the resource scheduling/allocation schemes and
on the corresponding protocol design of the medium-access-control (MAC) layer.
Furthermore, integrated data and energy transfer in the RF band also requires
a joint networking concept for heterogeneous data and energy transceivers. All
these challenging issues require novel Data and Energy Integrated Communication
Networks (DEINs) [33].](https://image.slidesharecdn.com/35569-250423081821-a8a4d60d/85/Data-and-Energy-Integrated-Communication-Networks-Jie-Hu-20-320.jpg)
![1.5 Ubiquitous Architecture of the DEIN 7
10 15 20 25 30 35 40 45 50
Transmit power (dBm)
(b)
0
5
10
15
20
25
30
35
40
Energy
effciency
of
WIT
(bps/Hz/mW)
10 15 20 25 30 35 40 45 50
Transmit power (dBm)
(a)
0
0.15
0.3
0.45
0.6
0.75
0.9
1.05
1.2
Energy
efficiency
of
WET
(mW/mW
%)
1.0
Channel
Attenuation
Fig. 1.1 Energy efficiency of WET (a) and that of WIT (b) against transmit power of RF signals.
The noise power is PN = −94 dBm, which is calculated by the power spectrum density of the
thermal noise −174 dBm/Hz and 100 MHz of the RF signals’ bandwidth. The aggregate interference
power at the receiver is set to be PI = −20 dBm, which appears in a heterogeneous cellular network
with the highest probability [28]. The distance between a transmitter and receiver pair is 10 m. The
path loss is calculated by the model invoked in [29–32], where the path loss exponent is 2. No fading
is assumed. The antenna gain in this example is set to be 40 dBi in order to counteract the path loss
1.5 Ubiquitous Architecture of the DEIN
DEINs are naturally heterogeneous in terms of all their technical aspects. We will
investigate the heterogeneity of the DEINs and synthesise a diverse range of WET
and WIT scenarios into its ubiquitous architecture, which is exemplified in Fig.1.2.
1.5.1 Heterogeneous Infrastructure
First of all, there are various types of infrastructure elements in heterogeneous DEIN.
As portrayed in Fig.1.2, we have generally three basic type of infrastructure in
DEINs, namely DEIN stations, WET stations and WIT stations/relays. DEIN stations
[34]arecapableofoperatingbothasinformationtransmitterandasenergytransmitter
for satisfying both of the user equipments’ (UEs’) data and energy requests. Thanks to
their powerful functionalities, DEIN stations are also capable of realising integrated
data and energy transfer for the sake of increasing the spectrum efficiency of the
congested RF band. Therefore, DEIN stations have to be connected to the core
communication network and they also have to be powered by stable energy sources,
such as large solar energy harvesters and the power grid. As illustrated in Fig.1.2,](https://image.slidesharecdn.com/35569-250423081821-a8a4d60d/85/Data-and-Energy-Integrated-Communication-Networks-Jie-Hu-21-320.jpg)
![8 1 Data and Energy Integrated Communication Networks: An Overview
WET Range
(Isotropic Antenna)
WIT Range
(Isotropic Antenna)
WIT-UE-1
WIT
WIT
WIT WET
WET
WIT-Relay-1
WIT-Relay-2
WET-Station-3
WET-Station-2
WET
WIT-UE-2
DEIN-
Station-1
WET-UE-1
WIT
WIT
DEIN-UE-1
WIT
WET
DEIN Devices
for IoT
Wide Beam for Integrated
Data and Energy Multicast
(Directional Antenna)
WET-UE-2
Narrow beam for point-to-
point WET
(Directional Antenna)
WIT
DEIN-UE-2
WIT
WIT
DEIN-
Station-2
WET-Station-1
WET
Fig. 1.2 Ubiquitous architecture of heterogeneous DEIN
DEIN-Station-1 may satisfy the integrated data and energy requests from the IoT
devices and those from DEIN-UE-1.
However, as we have discussed in Sect.1.4, the reliable WET range is far shorter
than the reliable WIT range, as exemplified in Fig.1.2. As a result, some blind
areas cannot be adequately covered by WET of DEIN stations. Furthermore, some
dedicated WET [35] stations are also deployed in order to supply energy to the
devices roaming in these blind areas. These WET stations are only connected to
energy sources, but they do not have to be connected to the core communication
network. As a result, they are dedicated for satisfying the UEs’ charging requests.
For instance, as shown in Fig.1.2, three WET stations are deployed in order to supply
energy to the UEs beyond the WET range of the DEIN stations.
Apart from DEIN stations and WET stations, there are still many conventional
communication stations in heterogeneous DEINs, namely the classic femto-cellular
stations, pico-cellular stations and macro-cellular stations [36]. These communica-
tion stations have different levels of transmit power and coverage, which results
in obvious heterogeneity in DEINs. Sometimes, low-cost relay stations are also
deployed for forwarding the data packets to cell-edge UEs, as illustrated in Fig.1.2.
However, small cellular stations and relay stations [37] are only capable of emitting
RF signals at a limited power. They are not suitable for carrying out sophisticated
WET tasks. Therefore, they are regarded as a dedicated communication infrastruc-
ture.
1.5.2 Heterogeneous User Equipment
Apart from the heterogeneous infrastructure, our DEINs have to accommodate both
charging and communication requests from diverse types of UEs. We generally have](https://image.slidesharecdn.com/35569-250423081821-a8a4d60d/85/Data-and-Energy-Integrated-Communication-Networks-Jie-Hu-22-320.jpg)
![1.5 Ubiquitous Architecture of the DEIN 9
three types of UEs in DEINs, namely the WIT UEs, the WET UEs and the DEIN
UEs [38], as exemplified in Fig.1.2. WIT UEs only require downlink and uplink
data transmission in DEINs. Since these UEs are always powered by stable energy
sources, they do not request any wireless charging from the DEIN stations. Laptops
and tablets are typical WIT UEs, which are either powered by high-capacity batteries
or are connected to the power grid. For example, as illustrated in the left part of
Fig.1.2, WIT-UE-1 receives its requested data from DEIN-Station-1 with the aid of
two WIT relay stations, while WIT-UE-2 may consume its own energy for powering
its uplink information transmission.
By contrast, since WET UEs are not powered by stable energy sources, they
have to request additional energy supply either from the DEIN stations or from the
WET stations in order to support their basic functionalities, such as uplink informa-
tion transmissions and energy-consuming computations [39]. For instance, although
WET-UE-1 is beyond the WIT range of DEIN-Station-1, it may still establish reli-
able uplink transmissions with DEIN-Station-1 by exploiting the additional energy
received from WET-Station-1, as exemplified in Fig.1.2. Similarly, the uplink trans-
mission of WET-UE-2 towards DEIN-Station-2 is powered by DEIN-Station-2 itself.
Miniature-sized IoT devices are typical WET UEs, since their functionalities are lim-
ited by the amount of energy stored in their batteries.
Furthermore, some UEs simultaneously request data and energy transmissions,
which are regarded as DEIN UEs [40]. For instance, in the right cell of Fig.1.2, DEIN-
UE-1 simultaneously receives its requested data and energy from DEIN-Station-2,
while DEIN-UE-2 also simultaneously requests both downlink data transmission and
wireless charging. However, since DEIN-UE-2 is beyond the WET range of DEIN-
Station-1, it can only receive the requested data from DEIN-Station-2, but it can
receive energy from WET-Station-1. This energy may be exploited for supporting
DEIN-UE-2’s uplink data transmission to its associated DEIN-Station-2.
Sometimes, the functionalities of WIT relay stations are also limited by their
energy supply, especially when the WIT relay stations rely on energy gleaned from
batteries or harvested from renewable sources. As a result, they also need wireless
charging from DEIN stations or WET stations for powering their data packet for-
warding actions [41]. As a result, WIT relay stations can also be regarded as special
“DEIN UEs”. As portrayed in the left cell of Fig.1.2, both data and energy are simul-
taneously transferred from DEIN-Station-1 to WIT-Relay-1. The energy harvested
by WIT-Relay-1 may be further exploited for forwarding the data packets to the
next hop. Since WIT-Relay-2 is beyond the WET range of DEIN-Station-2, it has to
request WET from the nearby WET-Station-2 and WET-Station-3. After receiving
the data packets from WIT-Relay-1 and gleaning sufficient energy from the WET
stations, the data packets are finally forwarded to their destination WIT-UE-1 by
WIT-Relay-2.](https://image.slidesharecdn.com/35569-250423081821-a8a4d60d/85/Data-and-Energy-Integrated-Communication-Networks-Jie-Hu-23-320.jpg)
![10 1 Data and Energy Integrated Communication Networks: An Overview
1.5.3 Heterogeneous Techniques for WIT and WET
Our DEIN architecture has to accommodate both the WET and WIT in the same
RF spectral band. Although the WET and WIT both rely on the RF signal, they still
have distinctive features, as summarised in Sect.1.4. Therefore, in order to satisfy
the UEs’ information and energy requests, the coexistence of WET and WIT in the
DEIN results in natural heterogeneity.
In order to guarantee the seamless WIT coverage, different techniques have to be
invoked. As exemplified in Fig.1.2, when the omnidirectional antennas are adopted,
the boundary of a DEIN cell is determined by the WIT range of a DEIN sta-
tion. As a result, the UEs residing within the WIT range of a DEIN station may
receive their requested information via a single-hop cellular link. Furthermore, these
UEs are also capable of uploading information to their associated DEIN stations.
Observe from Fig.1.2 that WIT-UE-2, DEIN-UE-1 and DEIN-UE-2 all receive their
requested information from the downlink WIT of their associated DEIN stations,
while WET-UE-2 and DEIN-UE-2 both upload their information to their associated
DEIN-Station-2. By contrast, in order to satisfy the information request of a UE
sitting beyond the WIT range of a DEIN station, multiple relay stations have to be
relied upon for forwarding the information from the DEIN station to the requester
or in a reverse direction via the multi-hop transmissions, such as the downlink trans-
mission from DEIN-Station-1 to WIT-UE-1 of Fig.1.2. In addition, by exploiting
the extra energy supplied by the WET stations, a UE beyond the WIT range of a
DEIN station is also capable of uploading data to the DEIN station [39], such as the
uplink transmission of WET-UE-1 to DEIN-Station-1 in Fig.1.2, which is powered
by WET-Station-1.
If we further look into the wireless charging actions in DEINs, various WET tech-
niques have to be invoked for satisfying diverse charging requirements. As illustrated
in Fig.1.2, a DEIN station’s WET range is much shorter than its WIT range, when
the omnidirectional antenna is adopted. The reason is that for the successful WET,
the energy harvesting circuit of the receiver can only be activated by a high received
energy. As a result, the WET is more sensitive to the wireless channel attenuation,
which is dominated by the path loss. If a WET UE resides within the WET range of
a DEIN station, it may successfully harvest energy from the RF signal emitted by
this DEIN station. By contrast, when a WET UE is beyond the WET range, it has
to request energy from its nearby WET station. Directional antennas may enable a
DEIN station to focus its energy in the main-lobe, which substantially increases the
long-range WET efficiency in the direction of the main-lobe. However, the resultant
energy loss in the side-lobes may significantly reduce the WET efficiency in other
directions. If directional antennas are adopted by the DEIN stations, they may form
a narrow energy beam [42] for charging the WET UE beyond the WET range, which
is characterised by the omnidirectional antennas. As exemplified in the right DEIN
cell of Fig.1.2, WET-UE-2 is still capable of receiving energy from the dedicated
narrow energy beam forming by DEIN-Station-2.
Furthermore, IoT devices will be pervasively deployed in the near future. Our
heterogeneous DEINs are also responsible for satisfying both of their communication](https://image.slidesharecdn.com/35569-250423081821-a8a4d60d/85/Data-and-Energy-Integrated-Communication-Networks-Jie-Hu-24-320.jpg)
![1.5 Ubiquitous Architecture of the DEIN 11
and energy demands. IoT devices normally are clustered in a specific area in order to
jointly carry out their tasks. As a result, for the sake of satisfying the charging requests
of the multiple IoT devices, DEIN stations may form wide-angle energy beams for
covering the cluster of requesters [43]. This technique may be regarded as energy
multicast. Moreover, this wide beam is also capable of transferring information and
energy together to the multiple requesters.
References
1. A. Zanella, N. Bui, A. Castellani, L. Vangelista, M. Zorzi, Internet of things for smart cities.
IEEE Internet Things J. 1(1), 22–32 (2014)
2. A. Alrawais, A. Alhothaily, C. Hu, X. Cheng, Fog computing for the internet of things: security
and privacy issues. IEEE Internet Comput. 21(2), 34–42 (2017)
3. B.H. Choi, V.X. Thai, E.S. Lee, J.H. Kim, C.T. Rim, Dipole-coil-based wide-range inductive
power transfer systems for wireless sensors. IEEE Trans. Ind. Electron. 63(5), 3158–3167
(2016)
4. V. Jiwariyavej, T. Imura, Y. Hori, Coupling coefficients estimation of wireless power transfer
system via magnetic resonance coupling using information from either side of the system.
IEEE J. Emerg. Sel. Top. Power Electron. 3(1), 191–200 (2015)
5. P.S. Riehl, A. Satyamoorthy, H. Akram, Y.C. Yen, J.C. Yang, B. Juan, C.M. Lee, F.C. Lin,
V. Muratov, W. Plumb, P.F. Tustin, Wireless power systems for mobile devices supporting
inductive and resonant operating modes. IEEE Trans. Microw. Theory Tech. 63(3), 780–790
(2015)
6. M. Pinuela, D.C. Yates, S. Lucyszyn, P.D. Mitcheson, Maximizing DC-to-load efficiency for
inductive power transfer. IEEE Trans. Power Electron. 28(5), 2437–2447 (2013)
7. H. Liu, Maximizing efficiency of wireless power transfer with resonant inductive coupling
(2011)
8. C. Zheng, H. Ma, J.S. Lai, L. Zhang, Design considerations to reduce gap variation and mis-
alignment effects for the inductive power transfer system. IEEE Trans. Power Electron. 30(11),
6108–6119 (2015)
9. Z. Yan, Y. Li, C. Zhang, Q. Yang, Influence factors analysis and improvement method on
efficiency of wireless power transfer via coupled magnetic resonance. IEEE Trans. Magn.
50(4), 1–4 (2014)
10. J.M. Miller, O.C. Onar, M. Chinthavali, Primary-side power flow control of wireless power
transfer for electric vehicle charging. IEEE J. Emerg. Sel. Top. Power Electron. 3(1), 147–162
(2015)
11. A. Kurs, A. Karalis, R. Moffatt, J.D. Joannopoulos, P. Fisher, M. Soljačić, Wireless power
transfer via strongly coupled magnetic resonances. Science 317(5834), 83–86 (2007)
12. H. Hwang, J. Moon, B. Lee, C.H. Jeong, S.W. Kim, An analysis of magnetic resonance coupling
effects on wireless power transfer by coil inductance and placement. IEEE Trans. Consum.
Electron. 60(2), 203–209 (2014)
13. M.Q. Nguyen, D. Plesa, S. Rao, J.C. Chiao, A multi-input and multi-output wireless energy
transfer system, in 2014 IEEE MTT-S International Microwave Symposium (IMS2014) (2014),
pp. 1–3
14. M.Q. Nguyen, Y. Chou, D. Plesa, S. Rao, J.C. Chiao, Multiple-inputs and multiple-outputs
wireless power combining and delivering systems. IEEE Trans. Power Electron. 30(11), 6254–
6263 (2015)
15. J.H. Kim, H.Y. Yu, C. Cha, Efficiency enhancement using beam forming array antenna for
microwave-based wireless energy transfer, in 2014 IEEE Wireless Power Transfer Conference
(2014), pp. 288–291](https://image.slidesharecdn.com/35569-250423081821-a8a4d60d/85/Data-and-Energy-Integrated-Communication-Networks-Jie-Hu-25-320.jpg)
![16 2 Fundamental of Integrated WET and WIT
2.1 Information Theoretical Essence
As previously discussed in Sect. 1.4, WET and WIT entail several conflicting speci-
fications, when they are coordinated in the same radio frequency (RF) spectral band.
As a result, theoretical investigations have to be carried out in order to reveal the
underlying relationship between the WET and WIT in data and energy integrated
communication networks (DEINs), which may provide researchers and engineers
with further valuable insights on improving the system-level performance of DEINs.
In this section, we will explore the information theoretical essence for DEIN and
reveal the natural contradiction between WET and WIT from an information theo-
retical perspective, which requires further efforts for jointly designing energy and
information transfer.
Note that the information theoretical exploration remain valid not only for the
integratedWETandWIToperatingintheRFspectralband,butforallotherintegrated
data and power transfer scenarios, such as power line communication [1] and power
over Ethernet technique [2]. Classic information theoretical channel capacity analysis
has been dedicated to maximising the mutual information under the constraint of
specific input signals. By contrast, the pioneering work of Gastpar [3] has attempted
to maximise the mutual information under the constraint of specific output signals,
which aims for controlling the interference imposed by a communicating pair on
other peers. This piece of work may provide us with valuable hints for finding the
performance limits of integrated data and energy transfer.
2.1.1 Discrete-Input-Discrete-Output Memoryless Channel
We first consider the classic Discrete-Input-Discrete-Output Memoryless Channel
(DIDO-MC) of Fig.2.1. If a DIDO-MC has a single input symbol x and a single
output symbol y, the transition probability of this DMC channel may be expressed
by the probability function of pY|X (y|x). All the legitimate values of the input
symbol x constitute the input codebook X, while all the legitimate values of the
output symbol y constitute the output codebook Y. Furthermore, given a specific
output symbol y, the energy carried by it can be represented by the non-negative
function g(y).
Let us now move on to the n-dimensional random input, which is expressed
by the vector Xn
= (X1, X2, . . . , Xn). All the random symbols in the vector Xn
DIDO-MC
Single input
symbol
Single output
symbol
Fig. 2.1 Discrete-input-discrete-output memoryless channel](https://image.slidesharecdn.com/35569-250423081821-a8a4d60d/85/Data-and-Energy-Integrated-Communication-Networks-Jie-Hu-29-320.jpg)
![2.1 Information Theoretical Essence 17
are independent of one another. If a sample of the n dimensional random input is
xn
= (x1, x2, . . . , xn), its corresponding occurrence probability can be expressed as
pXn (xn
) =
n
i=1 pX (xi ), where Xn
represents the n-dimensional codebook con-
taining all the possible values of the random output Xn
and pX (xi ) represents the
probability of the symbol xi being generated. When the input sample xn
is trans-
mitted by the information source, the corresponding output at the information des-
tination is denoted by the vector yn
= (y1, y2, . . . , yn). This sequence of symbols
occurs with a probability of pYn (yn
) =
n
i=1 pY (yi ), where Yn
represents the n-
dimensional codebook containing all the possible values of the random output Yn
and p†(yi ) =
xi ∈X pX (xi )pY|X (yi |xi ) represents the probability of the symbol yi
being received by the information destination. The energy carried by the sequence
yn
can be calculated by the non-negative function g(yn
). Assuming a random output
sequence Yn
= (Y1, Y2, . . . , Yn), the average energy carried by this n-dimensional
sequence can be formulated as
E[g(Yn
)] =
yn∈Yn
g(yn
) · pYn (yn
). (2.1)
As a result, the WIT performance limit can be formulated as the following optimi-
sation problem:
Objective: max
pXn (xn)
I (Xn
; Yn
), (2.2)
Subject to:
1
n
· E[g(Yn
)] ≥ β, (2.2a)
where I (Xn
; Yn
) is the average mutual information between the n-dimensional
input symbol sequence and its output counterpart. Given the transition probabili-
ties pY|X (Y|X) of the DIDO-MC, the optimisation problem (2.2) aims for finding
the optimal n-dimensional information source, which is represented by the probabil-
ities pXn (xn
) of the symbol sequences being generated by this information source.
However, this optimisation problem is subject to the condition (2.2a), suggesting
that the average energy carried by a single output symbol has to be higher than a
threshold β in order to satisfy the basic WET requirement. Substituting the optimal
pXn (xn
) into the objective (2.2), we may derive the channel capacity Cn(β), when
the input is an n-dimensional symbol sequence. Note that Cn(β) is a function of
the energy constraint β. Furthermore, the normalised channel capacity subject to the
energy constraint β can be further formulated as
C(β) = sup
n
1
n
· Cn(β), [bit/symbol] (2.3)
which may be regarded as the rate-energy function. Observe from (2.3) that the rate-
energy function is a natural extension of the classic channel capacity concept. This](https://image.slidesharecdn.com/35569-250423081821-a8a4d60d/85/Data-and-Energy-Integrated-Communication-Networks-Jie-Hu-30-320.jpg)
![18 2 Fundamental of Integrated WET and WIT
function only depends on a channel’s statistical property and on the requirement of
the energy harvested, but it does not rely on the information source.
The rate-energy functions of some simple binary channels will now be studied for
illustrating the above information theoretical methodology. By adopting the classic
ON-OFF-Keying (OOK) as our modulation scheme, for a binary random input, we
may assume that the symbol ‘0’ does not carry any energy as g(0) = 0, while the
symbol ‘1’ carries a single unit of energy as g(1) = 1. The probability distribution
of the input binary symbols is denoted as pX = {pX (0) = q, pX (1) =
q = 1 − q}.
As illustrated in Fig.2.2a, in a noiseless channel, an input binary symbol, either ‘0’
or ‘1’, can be correctly output. Therefore, neither energy loss nor energy gain exist in
the noiseless channel. The mutual information of the noiseless channel is IN (X; Y) =
−q log2 q −
q log2
q. Without any energy constraint, we can maximise IN (X; Y) by
letting q = 1/2. Accordingly, the maximum mutual information is IN,max(X; Y) = 1
bit/symbol.Moreover,theaverageenergycarriedbythecorrespondingoutputsymbol
is βth = 1/2 unit. Furthermore, the maximum energy carried by the output symbol
at the information destination Y is βmax = 1 unit, when the source never send the
symbol ‘0’, namely q = 0. In a nutshell, given a specific energy transfer requirement
β, the maximum achievable information rate of the noiseless channel is formulated
as [4]
CN (β) =
1, 0 ≤ β ≤ 1
2
,
H2(β), 1
2
β ≤ 1,
[bit/symbol], (2.4)
where H2(β) = −β log2 β − (1 − β) log2(1 − β)isthebinaryentropyfunctionwith
respect to β.
As shown in Fig.2.2b, in a Z channel, the input binary symbol ‘0’ can be correctly
output for certain. By contrast, the input binary symbol ‘1’ can be erroneously output
as the symbol ‘0’ with a probability ω, while it can be correctly output with a
probability
ω = 1 − ω. In this channel, the energy carried by the symbol ‘1’ may
be lost during the transmission, due to the channel attenuation. By contrast, since
no additional energy is supplied to the system, the input symbol ‘0’ does not have
a chance to become the energy carrier symbol ‘1’. The mutual information of the Z
channel is expressed as
Input Output
0
1
0
1
(a) Noiseless Channel
0
1
0
1
(c) Symmetric Channel
Input Output
0
1
0
1
(b) Z Channel
Input Output
Fig. 2.2 Three typical binary channels](https://image.slidesharecdn.com/35569-250423081821-a8a4d60d/85/Data-and-Energy-Integrated-Communication-Networks-Jie-Hu-31-320.jpg)
![2.1 Information Theoretical Essence 19
IZ (X; Y) = −(q +
qω) log2(q +
qω) −
q
ω log2
q +
qω log2 ω. (2.5)
Without any energy constraint, we can maximise IZ (X; Y) by letting
q =
1 − ω
1
1−ω
1 + (1 − ω)ω
ω
1−ω
. (2.6)
Accordingly, the maximum mutual information is IZ,max = log2(1 +
ωω
ω
ω ). More-
over, the average energy carried by the corresponding output symbol is formulated
as
βth =
ω
1
ω
1 +
ωω
ω
ω
[unit]. (2.7)
When the source only sends the energy carrier symbol ‘1’, namely q = 0, the maxi-
mum energy carried by the output symbol at the information destination is βmax = ω
unit. In a nutshell, given a specific energy transfer requirement β, the maximum
achievable information rate of the Z channel can be formulated as [4]
CZ (β) =
log
1 − ω
1
1−ω + ω
ω
1−ω , 0 ≤ β ≤ (1 − ω)π∗
H2(β) − β
1−ω
H2(ω), (1 − ω)π∗
β ≤ 1 − ω,
[bit/symbol],
(2.8)
where the variable π∗
is given by [4]
π∗
=
ω
ω
1−ω
1 + (1 − ω)ω
ω
1−ω
. (2.9)
As portrayed in Fig.2.2c, in a symmetric channel, both the input binary symbols
‘0’ and ‘1’ can be erroneously delivered to the output end with a probability of ω,
while they can be correctly delivered with a probability of
ω = 1 − ω. Apart from
the energy loss incurred by the transition from the input symbol ‘1’ to the output
symbol ‘0’, the energy of the interference may change the input symbol ‘0’ to the
output symbol ‘1’, which results in the energy gain at the information destination.
The mutual information of the symmetric channel is expressed as
IS(X; Y) = − (q
ω +
qω) log2(q
ω +
qω) − (qω +
q
ω) log2(qω +
q
ω)
+
ω log2
ω + ω log2 ω. (2.10)
Without any energy constraint, we can maximise IS(X; Y) by letting q = 1/2.
Accordingly, the maximum mutual information is IS,max = 1 +
ω log2
ω + ω log2 ω.
Moreover, the average energy carried by the corresponding output symbol is βth =
1/2 unit. When the source only sends the energy carrier symbol ‘1’, namely q = 0,](https://image.slidesharecdn.com/35569-250423081821-a8a4d60d/85/Data-and-Energy-Integrated-Communication-Networks-Jie-Hu-32-320.jpg)
![Mr. Ball. I'm going to ask you what you learned during the
course of your investigation.
Mr. Boyd. All right.
Mr. Ball. Now, can you tell me something about yourself, where
you were born and where you went to school and what you have
done most of your life?
Mr. Boyd. Well, yes, sir. I can tell you I was born in Navarro
County—the particular place was Blooming Grove, Tex., and it's
about 15 miles west of Corsicana, and I was raised up about 7 miles
north of there. I attended school, well, I started at a little country
school—it was Pecan, was the name of the school. I went there 2
years and then they sent me to Blooming Grove and I started to
school in my second grade. The reason I was in the second grade—I
had to go through a primer before I got in the first grade—I didn't
fail—I just had to go through this primer before I got in the first
grade, and I graduated from high school at Blooming Grove in 1946
and I went into the Navy and served for 2 years, I believe I served
about 22 months in the Navy—I joined and I went through boot
training at San Diego, went from there to Newport, R. I., and caught
my first ship, the USS Kenneth D. Bailey. I don't recall just how many
months I spent on that—somewhere around 15 or 16 months, I've
forgotten, and then they sent me to—I transferred from that ship
and went on the USS Cone, that's another destroyer [spelling] C-o-n-
e, and along about the first part of January, I believe, in 1948, they
transferred me to Pensacola where I caught my third destroyer, the
USS Forrest Royal, and we operated in and out of there until I got
out of the Navy, and I believe it was about the first day of April
1948, when I was discharged, and I came to Dallas and I have been
here in Dallas ever since.
I went to work on the police department May 19, 1952. Prior to
that I worked, I believe, about 3 years for the gas company and I
started out reading gas meters, and then I went into collecting, and
I was a collector for the gas company when I came on the police
department. I think I worked a couple of more places before then—](https://image.slidesharecdn.com/35569-250423081821-a8a4d60d/85/Data-and-Energy-Integrated-Communication-Networks-Jie-Hu-38-320.jpg)
Data and Energy Integrated Communication Networks Jie Hu
- 1. Data and Energy Integrated Communication Networks Jie Hu download https://textbookfull.com/product/data-and-energy-integrated- communication-networks-jie-hu/ Download more ebook from https://textbookfull.com
- 2. We believe these products will be a great fit for you. Click the link to download now, or visit textbookfull.com to discover even more! Mobile Networks and Management Jiankun Hu https://textbookfull.com/product/mobile-networks-and-management- jiankun-hu/ Data and Communication Networks: Proceedings of GUCON 2018 Lakhmi C. Jain https://textbookfull.com/product/data-and-communication-networks- proceedings-of-gucon-2018-lakhmi-c-jain/ Fundamentals of Data Communication Networks 1st Edition Oliver C. Ibe https://textbookfull.com/product/fundamentals-of-data- communication-networks-1st-edition-oliver-c-ibe/ Smart Sensors Networks Communication Technologies and Intelligent Applications A volume in Intelligent Data Centric Systems Fatos Xhafa https://textbookfull.com/product/smart-sensors-networks- communication-technologies-and-intelligent-applications-a-volume- in-intelligent-data-centric-systems-fatos-xhafa/
- 3. An Integrated Approach To Communication Theory And Research Don W. Stacks https://textbookfull.com/product/an-integrated-approach-to- communication-theory-and-research-don-w-stacks/ Large Scale Integrated Energy Systems Planning and Operation Qing-Hua Wu https://textbookfull.com/product/large-scale-integrated-energy- systems-planning-and-operation-qing-hua-wu/ 5G Green Mobile Communication Networks Xiaohu Ge https://textbookfull.com/product/5g-green-mobile-communication- networks-xiaohu-ge/ Big Data and Networks Technologies Yousef Farhaoui https://textbookfull.com/product/big-data-and-networks- technologies-yousef-farhaoui/ Data Driven Remaining Useful Life Prognosis Techniques Stochastic Models Methods and Applications Hu https://textbookfull.com/product/data-driven-remaining-useful- life-prognosis-techniques-stochastic-models-methods-and- applications-hu/
- 4. SPRINGER BRIEFS IN COMPUTER SCIENCE Jie Hu · KunYang Data and Energy Integrated Communication Networks A Brief Introduction
- 5. SpringerBriefs in Computer Science Series editors Stan Zdonik, Brown University, Providence, Rhode Island, USA Shashi Shekhar, University of Minnesota, Minneapolis, Minnesota, USA Xindong Wu, University of Vermont, Burlington, Vermont, USA Lakhmi C. Jain, University of South Australia, Adelaide, South Australia, Australia David Padua, University of Illinois Urbana-Champaign, Urbana, Illinois, USA Xuemin Sherman Shen, University of Waterloo, Waterloo, Ontario, Canada Borko Furht, Florida Atlantic University, Boca Raton, Florida, USA V. S. Subrahmanian, University of Maryland, College Park, Maryland, USA Martial Hebert, Carnegie Mellon University, Pittsburgh, Pennsylvania, USA Katsushi Ikeuchi, University of Tokyo, Tokyo, Japan Bruno Siciliano, Università di Napoli Federico II, Napoli, Italy Sushil Jajodia, George Mason University, Fairfax, Virginia, USA Newton Lee, Newton Lee Laboratories, LLC, Tujunga, California, USA
- 6. SpringerBriefs present concise summaries of cutting-edge research and practical applications across a wide spectrum of fields. Featuring compact volumes of 50 to 125 pages, the series covers a range of content from professional to academic. Typical topics might include: • A timely report of state-of-the art analytical techniques • A bridge between new research results, as published in journal articles, and a contextual literature review • A snapshot of a hot or emerging topic • An in-depth case study or clinical example • A presentation of core concepts that students must understand in order to make independent contributions Briefs allow authors to present their ideas and readers to absorb them with minimal time investment. Briefs will be published as part of Springer’s eBook collection, with millions of users worldwide. In addition, Briefs will be available for individual print and electronic purchase. Briefs are characterized by fast, global electronic dissemination, standard publishing contracts, easy-to-use manuscript preparation and formatting guidelines, and expedited production schedules. We aim for publication 8–12 weeks after acceptance. Both solicited and unsolicited manuscripts are considered for publication in this series. More information about this series at http://www.springer.com/series/10028
- 7. Jie Hu • Kun Yang Data and Energy Integrated Communication Networks A Brief Introduction 123
- 8. Jie Hu School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu, Sichuan China Kun Yang School of Computer Science and Electronic Engineering University of Essex Colchester UK ISSN 2191-5768 ISSN 2191-5776 (electronic) SpringerBriefs in Computer Science ISBN 978-981-13-0115-5 ISBN 978-981-13-0116-2 (eBook) https://doi.org/10.1007/978-981-13-0116-2 Library of Congress Control Number: 2018943388 © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd., part of Springer Nature 2018 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Printed on acid-free paper This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore
- 9. Preface In order to satisfy the power thirsty of communication devices in the imminent fifth-generation (5G) and Internet of Things (IoT) era, wireless charging techniques have attracted much attention both from the academic and industrial communities. Although the inductive coupling and magnetic resonance based charging techniques are indeed capable of supplying energy in a wireless manner, they tend to restrict the freedom of movement. By contrast, RF signals are capable of supplying energy over distances, which are gradually inclining closer to our ultimate goal—charging anytime and anywhere. Furthermore, transmitters capable of emitting RF signals have been widely deployed, in TV towers, cellular base stations and WiFi access points. This communication infrastructure may indeed be employed also for wireless energy transfer (WET). Therefore, no extra investment in a dedicated WET infrastructure is required. However, allowing radio frequency (RF) signal based wireless energy transfer (WET) may impair the wireless information transfer (WIT) operating in the same spectrum. Hence, it is crucial to coordinate and balance WET and WIT for simultaneous wireless information and power transfer (SWIPT), which evolves to data and energy integrated communication networks (DEINs). This brief aims for providing a landscape picture of DEINs, while including latest research contributions in this promising topic. To this end, we first provide an overview of DEIN in Chap. 1. We will look into the energy shortage of the electronic devices, compare the popular wireless charging techniques with one another and highlight the RF signal based WET and its distinctive features against the conventional wireless communication in the same spectral bands. Then, we will describe the ubiquitous architecture of DEINs. In Chap. 2, we will focus on the fundamental of the physical layer for imple- menting the integrated WET and WIT of the point-to-point link. Key enabling modules of the generic transceiver architecture for the integrated WET and WIT will be introduced. Then, we will introduce several popular receivers equipped with multiple antennas for simultaneously information and energy reception, namely the ideal receiver, the spatial splitting based receiver, the power splitting based receiver and the time switching based receiver. v
- 10. In Chap. 3, we consider a typical DEIN system consisting of a single H-BS and multiple DEIN users, who are eager to receive both information and energy simultaneously during the downlink transmission of the H-BS. The DEIN users then exploit the energy harvested from the downlink for powering their own uplink transmission. Both the downlink and uplink transmissions are time slotted in order to reduce the potential interference and transmission collision among the multiple users. Optimal time slot allocation schemes in the MAC layer are proposed for maximising the sum-throughput and the fair-throughput of the DEIN users’ uplink transmissions, respectively. In Chap. 4, a full-duplex aided H-BS is conceived in a multi-user DEIN for the sake of simultaneously transferring energy during its downlink transmission, while receiving the information uploaded by the multiple users. The uplink transmissions of the multiple users are powered by the energy harvested from the H-BS’s downlink energy broadcast. In this full-duplex aided DEIN, a joint time allocation and user scheduling algorithm is proposed for the sake of maximising the sum-throughput of the users’ uplink transmissions by further considering the users’ actual information uploading requirements. Finally, we conclude this brief in Chap. 5 by providing some emerging research topics in the DEIN. This brief aims for boosting the joint effort from both the academia and industry so as to push the DEIN a step closer to the practical implementation. It is also suitable for the undergraduate/postgraduate students to be familiar with this cutting-edge technique. We would like to thank Mr. Yizhe Zhao and Mr. Kesi Lv for their tremendous contribution to Chaps. 2–5. We would also like to thank Prof. Xuemin (Sherman) Shen, University of Waterloo, for his outstanding editorial organisation of this influential series in the computer science. The financial support of the National Natural Science Foundation of China (NSFC), Grant No. 61601097, U1705263, and 61620106011, as well as that of Fundamental Research Funds for the Central Universities, Grant No. ZYGX2016Z011 are gratefully acknowledged. This work is also sponsored by Huawei Innovative Research Program (HIRP). Chengdu, China Jie Hu Colchester, UK Kun Yang vi Preface
- 11. Contents 1 Data and Energy Integrated Communication Networks: An Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Energy Dilemma for Electronic Devices. . . . . . . . . . . . . . . . . . . . 1 1.2 Near-Field Wireless Energy Transfer . . . . . . . . . . . . . . . . . . . . . . 2 1.3 RF Signal Based Wireless Energy Transfer . . . . . . . . . . . . . . . . . 3 1.4 WET Versus WIT in the RF Spectral Band . . . . . . . . . . . . . . . . . 5 1.5 Ubiquitous Architecture of the DEIN . . . . . . . . . . . . . . . . . . . . . . 7 1.5.1 Heterogeneous Infrastructure . . . . . . . . . . . . . . . . . . . . . . 7 1.5.2 Heterogeneous User Equipment . . . . . . . . . . . . . . . . . . . . 8 1.5.3 Heterogeneous Techniques for WIT and WET. . . . . . . . . . 10 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2 Fundamental of Integrated WET and WIT . . . . . . . . . . . . . . . . . . . 15 2.1 Information Theoretical Essence . . . . . . . . . . . . . . . . . . . . . . . . . 16 2.1.1 Discrete-Input-Discrete-Output Memoryless Channel . . . . . 16 2.1.2 Continuous-Input-Continuous-Output Memoryless Channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2.2 Transceiver Architecture of DEIN Devices . . . . . . . . . . . . . . . . . . 23 2.3 Signal Splitter Based Receiver Architecture . . . . . . . . . . . . . . . . . 26 2.3.1 Ideal Receiver Architecture . . . . . . . . . . . . . . . . . . . . . . . 26 2.3.2 Spatial Splitting Based Receiver . . . . . . . . . . . . . . . . . . . . 26 2.3.3 Power Splitting Based Receiver . . . . . . . . . . . . . . . . . . . . 28 2.3.4 Time Switching Based Receiver . . . . . . . . . . . . . . . . . . . . 30 2.3.5 Rate-Energy Region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 vii
- 12. 3 Throughput Maximization and Fairness Assurance in Data and Energy Integrated Communication Networks . . . . . . . . . . . . . . 35 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 3.2 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 3.2.1 Structure of the TDMA Aided Operating Cycle . . . . . . . . . 38 3.2.2 Channel Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 3.2.3 Throughput of the Downlink Transmission . . . . . . . . . . . . 41 3.2.4 Throughput of the Uplink Transmission . . . . . . . . . . . . . . 41 3.3 Sum-Throughput Maximisation . . . . . . . . . . . . . . . . . . . . . . . . . . 42 3.4 Fair-Throughput Maximisation . . . . . . . . . . . . . . . . . . . . . . . . . . 46 3.5 Numerical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 3.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 4 Joint Time Allocation and User Scheduling in a Full-Duplex Aided Multi-user DEIN. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 4.2 Preliminary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 4.2.1 Wireless Powered Communication Network . . . . . . . . . . . 57 4.2.2 Full-Duplex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 4.3 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 4.3.1 Network Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 4.3.2 Frame Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 4.3.3 The Downlink WET and Uplink WIT . . . . . . . . . . . . . . . . 61 4.4 Problem Formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 4.4.1 Sum-Throughput Maximisation. . . . . . . . . . . . . . . . . . . . . 63 4.4.2 Iterative Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 4.5 Performance Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 4.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 5 Conclusions and Open Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . 71 5.1 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 5.2 Open Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 5.2.1 Efficiency Enhancement of WET . . . . . . . . . . . . . . . . . . . 73 5.2.2 Efficient Energy Storage. . . . . . . . . . . . . . . . . . . . . . . . . . 73 5.2.3 Heterogeneous Internet of Energy . . . . . . . . . . . . . . . . . . . 74 5.2.4 Information Theoretic WET Capacity . . . . . . . . . . . . . . . . 75 5.2.5 Interference Cancellation and Signal Decoupling . . . . . . . . 75 5.2.6 Socially Aware Placement of DEIN Stations . . . . . . . . . . . 76 5.2.7 DEIN Aided Mobile Cloud Computing . . . . . . . . . . . . . . . 76 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 viii Contents
- 13. Acronyms AC Alternative Current AWGN Additive White Gaussian Noise CDF Cumulative Distribution Function CDMA Code Division Multiple Access CICO-MC Continuous Input Continuous Output Memoryless Channel DC Direct Current DEIN Data and Energy Integrated communication Network DIDO-MC Discrete Input Discrete Output Memoryless Channel FDD Frequency Division Duplex H-BS Hybrid Base Station IoT Internet of Things KKT Karush–Kuhn–Tucker LPF Low-Pass Filter MAC Medium Access Control MIMO Multiple Input Multiple Output MISO Multiple Input Single Output mmW millimetre Wave NASA National Aeronautics and Space Administration NOMA Non-Orthogonal Multiple Access NSFC National Natural Science Foundation of China OFDMA Orthogonal Frequency Division Multiple Access PAPR Peak to Average Power Ratio PS Power Splitting PSK Phase Shift Keying QAM Quadrature Amplitude Modulation QoS Quality of Service RF Radio Frequency SCMA Sparse Code Multiple Access SER Symbol Error Ratio SIMO Single-Input-Multiple-Output ix
- 14. SISO Signle-Input-Single-Output SS Spatial Splitting SVD Singular Value Decomposition SWIPT Simultaneous Wireless Information and Power Transfer TDD Time Division Duplex TDMA Time Division Multiple Access TS Time Switching UE User Equipment UESTC University of Electronic Science and Technology of China WET Wireless Energy Transfer WIT Wireless Information Transfer WPCN Wireless Powered Communication Network 5G Fifth Generation x Acronyms
- 15. Chapter 1 Data and Energy Integrated Communication Networks: An Overview Abstract In order to address the energy supply issue of communication devices in the imminent 5G and IoT era, wireless charging techniques have attracted much attention both from the academic and industrial communities. Thankfully, RF signals are capable of delivering energy over distances. However, allowing RF signal based wireless energy transfer (WET) may impair the wireless information transfer (WIT) operating in the same spectral band. Hence, it is crucial to coordinate and balance WET and WIT for simultaneous wireless information and power transfer (SWIPT), which evolves to Data and Energy Integrated communication Networks (DEINs). To this end, a ubiquitous IDEN architecture is characterised by summarising its natural heterogeneity and by synthesising a diverse range of integrated WET and WIT scenarios. Keywords Data and Energy Integrated Communication Network (DEIN) Energy Efficiency · RF Signal based Wireless Charging · Simultaneous Wireless Information and Energy Transfer (SWIPT) · Ubiquitous Architecture of DEIN Wireless Energy Transfer (WET) · Wireless Information Transfer (WIT) Wireless Powered Communication Network (WPCN) We provide an overview of Data and Energy Integrated communication Network (DEIN) in this chapter. We will look into the energy shortage of the electronic devices, compare the popular wireless charging techniques with one another and highlight the RF signal based Wireless Energy Transfer (WET) and its distinctive features against the conventional wireless Information Transfer (WIT) in the same spectral bands. Then we will describe the ubiquitous architecture of DEINs by introducing its natural heterogeneity and by synthesising a diverse range of WET and WIT scenarios. 1.1 Energy Dilemma for Electronic Devices According to the prediction of the classic Moore’s Law, the density of transistors in an integrated circuit doubles approximately every two years, which have been fuelling the spectacular proliferation of electronic devices since the 1960s. Furthermore, con- © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd., part of Springer Nature 2018 J. Hu and K. Yang, Data and Energy Integrated Communication Networks, SpringerBriefs in Computer Science, https://doi.org/10.1007/978-981-13-0116-2_1 1
- 16. 2 1 Data and Energy Integrated Communication Networks: An Overview sumer electronic devices are becoming shirt-pocket-sized and mobile. These devices are normally powered by embedded batteries. However, as their functions become ever more sophisticated, their thirst for abundant energy is not matched by the slow progress of the batteries’ capacity. The situation in the communication industry is even more daunting. Since the roll-out of the fifth-generation (5G) cellular system and of the Internet of Things (IoT) is just around the corner, people’s appetite for super- high data transmission rates, for high density of connectivity and for high mobilities will indeed be satisfied to a large extent. A major portion of the future mobile data traffic will be constituted by novel types of services, including high-definition stero- scopic video streams, augmented/virtual reality, holographic tele-presence, cloud desktops, as well as online games, etc. All these services require the user terminals to be implemented with high computing capabilities for real-time signal process- ing, which may quickly drain the embedded batteries. Furthermore, sensors will be deployed in every corner of the future smart cities [1]. These sensors monitor the environment and upload sensing results to central servers [2]. The life-span of sensors and of sensing networks largely depends on the sensors’ battery capacity. Regularly replacing the batteries may be an unrealistic or tedious task. Accordingly, new sources of energy have to be explored to prolong the depletion period of con- ventional batteries in order to relieve the energy concerns of various communication devices. 1.2 Near-Field Wireless Energy Transfer Nowadays, resonant inductive coupling [3] and magnetic resonance coupling [4] have emerged for remotely charging electronic devices in the near-field. Resonant inductive coupling based wireless charging relies on the magnetic coupling that delivers electrical energy between two coils tuned to resonate at the same frequency. Thistechniquehasalreadybeencommercialisedforsomehomeelectronicappliances [5], such as mobile phones, electric toothbrushes and smart watches etc. However, the coupling coils only support near-field wireless energy transfer (WET) over a distance spanning from a few millimetres to a few centimetres [6], while achieving a WET efficiency as high as 56.7%, when operating at a frequency of 508 kHz [7]. Furthermore, resonant inductive coupling requires strict alignment of the coupling coils. Even a small misalignment may result in dramatic reduction of the WET efficiency [8]. As a result, during the charging process, the electronic appliances cannot be freely moved. By contrast, magnetic resonance coupling [9] delivers electrical energy between two resonators by exploiting evanescent-wave coupling. This technique has already been adopted for charging the electric vehicles due to its high WET efficiency [10]. For example, magnetic resonance coupling is capable of achieving a WET efficiency of 90% over a distance of 0.75 m [11]. Both its WET efficiency and its charging dis- tance are much higher than that of the resonant inductive coupling. However, mag- netic resonance coupling still belongs to the category of near-field wireless charging, since its power transfer efficiency dramatically reduces to 30%, when the distance is increased to 2.25 m [11]. Nonetheless, magnetic resonance coupling does not require
- 17. 1.2 Near-Field Wireless Energy Transfer 3 strict alignment between the rechargeable device and the energy source. Hence, dur- ing the charging process, the electronic appliances may be moved within the charging area [12]. Furthermore, a multiple-input-multiple-output (MIMO) system, which has already been widely adopted for improving the performance of the wireless commu- nication, can also be introduced into the magnetic resonance coupling based WET system for the sake of further enhancing the WET efficiency [13, 14]. 1.3 RF Signal Based Wireless Energy Transfer In contrast to the above-mentioned near-field WET techniques, the propagation of the RF signals is capable of supporting far-field WET [15]. The history of the RF signal based WET dates back to 1960, when the first long-distance WET system was established by Brown [16, 17]. Brown jointly designed rectifiers and antennas for energy receivers, which is now widely known as rectennas. They are capable of efficiently converting the Alternating-Current (AC) energy carried by the RF signals to Direct Current (DC) energy. This RF signal based WET system was validated by remotely powering a model helicopter from the ground in 1964 [16, 17]. In the 1970s and 1980s, intense efforts were invested into the research of RF signal based WET, which was largely motivated by the intention of developing a solar-powered satellite [18, 19]. In this system, a satellite may harvest energy from sunlight in the outer space and beam the energy back to ground stations via the propagation of RF signals. Furthermore, the Jet Propulsion Laboratory of the National Aeronautics and Space Administration (NASA) led a project from 1969 to 1975, in which 30 kW of power was beamed over a distance of 1 mile at a 84% RF-DC efficiency [20]. There are three main technical challenges in the RF based WET. Firstly, the long- distance propagation and adverse multipath fading may substantially attenuate the RF signals before they arrive at the receivers, which inevitably results in energy loss. Secondly, the energy carried by RF signals is of AC nature, which cannot be directly invoked for driving an electronic load. As a result, the AC energy carried by RF signals have to be converted to DC energy for any further use. However, some portion of energy is inevitably lost during this conversion process. Last but not the least, the diffraction of the RF signals’ waveform may expand the beam size. As a result, the receive antenna having limited size is not capable of capturing all the energy carried by the RF signals. For counteracting the signal attenuation of wireless channels, the transmit beams have to be accurately aimed at the energy receivers [21], which requires the joint design of the transmit and receive antennas. For improving the AC-DC conversion efficiency, the receive antennas have to be designed together with the rectifiers in order to achieve the impedance match for the sake of high-efficiency AC-DC conversion [22]. For alleviating the adverse effect of beam diffraction, the non-diffracted Bessel-Gaussian beam [23] can be invoked, which is capable of efficiently reducing the energy loss during the propagation and hence improve the WET efficiency over wireless channels. In general, the RF signal based WET has the following advantages over its near- field counterparts:
- 18. 4 1 Data and Energy Integrated Communication Networks: An Overview • Large coverage. Relying on the RF signals, energy can be transferred to receivers miles away. • High flexibility. The angular selectivity transmit beam can be intelligently adjusted according to various WET requirements. For instance, a narrow beam can be invoked for realising accurate and high-efficiency point-to-point WET, while a wide beam can be used for simultaneously charging multiple devices. • More applications. RF signals can be leveraged for supplying a large amount of energy to energy-hungry appliances, such as solar-powered satellite system. It can also be exploited for supplying energy to low-power devices, such as sensors and biomedical implants. • Low investment. The transmitters of the RF signals have been deployed at every corner of the globe, such as radio broadcast stations, TV towers, cellular base sta- tions and WiFi access points, etc. The legacy of the communication infrastructure can all be exploited for radiating energy to electronic devices. Only limited extra investment is required for deploying energy transmitters in order to cover some blind spots. The main features of different WET techniques are summarised in Table1.1. Table 1.1 Main features of different WET techniques Technique Range Direc. Frequency Antenna Application RF signals Long High MHz-GHz Parabolic dishes, rectennas, phased arrays Solar-powered satellite, drone aircraft, IoT devices, portable devices, RFID, smart cards and etc. Magnetic resonant coupling Middle Low kHz-GHz Tuned wire coils, lumped element resonators Portable devices, biomedical implants, electric vehicles, RFID, smartcard and etc. Inductive coupling Short Low Hz-MHz Wire coils Stovetops, industrial heaters and small electric appliances, such as electric toothbrush, razor and etc.
- 19. 1.4 WET Versus WIT in the RF Spectral Band 5 1.4 WET Versus WIT in the RF Spectral Band Since RF signal based WET techniques require highly fexible beam directivity in order to satisfy diverse charging requests, the best spectral band for steering energy beams is in the range of 10MHz to 100 GHz, which almost covers all the bands allocated for wireless communication services. For example, TV/Radio broadcasting services operate in the band spanning from 40MHz to 220 MHz [24], the mobile cellular communication system operates in the spectral band spanning from 800MHz to 3.7 GHz [25], while the WiFi communication system operates in the spectral band spanning from 2.4GHz to 6 GHz [26]. Furthermore, as a key technique in the upcoming 5G era, millimetre wave (mmW) [27] may significantly increase the achievable throughput of the air interface, which operates in the spectral band ranging from 10GHz to 100 GHz. Although they both operate in the same RF band, WET and WIT still have the following distinctive characteristics: • They have different functional circuits. RF signals in the pass-band cannot be directly invoked for both the information decoding and the energy harvesting. For the information decoding, the RF signals in the pass-band have to be firstly converted to the base-band, since all the signal processing has to be accomplished in the base-band. By contrast, for the energy harvesting, the AC energy carried by the RF signals has to be converted to the DC energy first, since only DC energy can be stored in batteries or drive electronic loads. Specifically, during the AC-DC conversion, the phase information carried by the RF signals is filtered. • They require different absolute energy at receivers. The activation of the energy harvesting circuits requires a relatively high energy carried by the received RF signals, which is approximately on the order of −20 dBm. If the energy carried by the received RF signal does not achieve the required activation threshold, none of this energy can be harvested. By contrast, the successful information recovery relies on the energy ratio between the received RF signal and the noise plus inter- ference, not on the absolute energy carried by the received RF signal. As a result, even a small amount of energy is capable of activating the information receiver, which is approximately on the order of −80 dBm. • They have different coverage. The RF signals are attenuated by hostile wireless channels, such as the path loss, shadowing and multipath fading. Since the energy harvesting requires a much higher absolute energy at the receivers than the infor- mation decoding, the range of WET is accordingly much shorter than that of WIT. Therefore, given the same set of transmitters and receivers, the resultant WET network has a different topology with the WIT network. • They treat noise and interference differently. The interference and noise ubiqui- tously exist in any WIT system, which seriously impair the WIT performance. Mitigating the performance degradation induced by the interference and noise is a major challenge in the WIT system design. By contrast, WET systems may actu- ally benefit from the interference and noise, since both of them are RF signals and they both carry useful energy. The interference and noise can be jointly harvested
- 20. 6 1 Data and Energy Integrated Communication Networks: An Overview by the energy harvesting circuits, which may provide additional energy harvesting gains for the energy requesters. • They have different definitions in energy efficiency. The energy efficiency of WET can be defined as the ratio of energy harvested by the receiver to the energy emitted by the transmitter, which can be formulated as ηW ET = 1 Pt · ρ (Pr + PI + PN ) (Watt/Watt), (1.1) where ρ is the conversion rate from the received RF energy to the DC energy by considering a linear RF-DC converter. By contrast, In the community of green communications, the energy efficiency of WIT is defined as the ratio of spectral efficiency to energy consumption, which is evaluated in the unit of bps/Hz/Watt or bps/Hz/Joule. By exploiting the classic Shannon-Hartley theorem in an Additive- White-Gaussian-Noise (AWGN) channel, the energy efficiency of WIT can be expressed as ηW I T = 1 Pt · log2 1 + Pr PI + PN (bps/Hz/Watt), (1.2) where Pt is the transmit power of the RF signal, Pr is the power received after the signal being attenuated by the hostile wireless channel, PI is the aggregate interference power and PN is the noise power at the receiver. In Fig.1.1, we exemplify the energy efficiency of WET and that of WIT, which can be calculated by (1.2) and (1.1), respectively. Observe from Fig.1.1a that in our setting, the energy efficiency of WET reduces from 1.1% but converges to 1%, which is due to the channel attenuation incurred by the path loss between the transmitter and receiver pair. Observe from Fig.1.1b that the energy efficiency of WIT gradually reduces from 35 bps/Hz/mW to 0 as the transmit power of the RF signal increases. By contrast, WET and WIT operating in the same RF spectral band may compete for the precious resources in the air interface and they may thus impair each other’s performance to some extent. For example, WET requires that the RF signals carry a high power to the receivers for the efficient energy harvesting. However, the high-power RF signals of the WET system may impose excessive interference on the WIT receivers, which may thus significantly degrade the WIT performance attained. As a result, coordinating WET and WIT in the same RF band imposes critical challenges on the RF circuit design, on the transceiver design of the physical layer, on the resource scheduling/allocation schemes and on the corresponding protocol design of the medium-access-control (MAC) layer. Furthermore, integrated data and energy transfer in the RF band also requires a joint networking concept for heterogeneous data and energy transceivers. All these challenging issues require novel Data and Energy Integrated Communication Networks (DEINs) [33].
- 21. 1.5 Ubiquitous Architecture of the DEIN 7 10 15 20 25 30 35 40 45 50 Transmit power (dBm) (b) 0 5 10 15 20 25 30 35 40 Energy effciency of WIT (bps/Hz/mW) 10 15 20 25 30 35 40 45 50 Transmit power (dBm) (a) 0 0.15 0.3 0.45 0.6 0.75 0.9 1.05 1.2 Energy efficiency of WET (mW/mW %) 1.0 Channel Attenuation Fig. 1.1 Energy efficiency of WET (a) and that of WIT (b) against transmit power of RF signals. The noise power is PN = −94 dBm, which is calculated by the power spectrum density of the thermal noise −174 dBm/Hz and 100 MHz of the RF signals’ bandwidth. The aggregate interference power at the receiver is set to be PI = −20 dBm, which appears in a heterogeneous cellular network with the highest probability [28]. The distance between a transmitter and receiver pair is 10 m. The path loss is calculated by the model invoked in [29–32], where the path loss exponent is 2. No fading is assumed. The antenna gain in this example is set to be 40 dBi in order to counteract the path loss 1.5 Ubiquitous Architecture of the DEIN DEINs are naturally heterogeneous in terms of all their technical aspects. We will investigate the heterogeneity of the DEINs and synthesise a diverse range of WET and WIT scenarios into its ubiquitous architecture, which is exemplified in Fig.1.2. 1.5.1 Heterogeneous Infrastructure First of all, there are various types of infrastructure elements in heterogeneous DEIN. As portrayed in Fig.1.2, we have generally three basic type of infrastructure in DEINs, namely DEIN stations, WET stations and WIT stations/relays. DEIN stations [34]arecapableofoperatingbothasinformationtransmitterandasenergytransmitter for satisfying both of the user equipments’ (UEs’) data and energy requests. Thanks to their powerful functionalities, DEIN stations are also capable of realising integrated data and energy transfer for the sake of increasing the spectrum efficiency of the congested RF band. Therefore, DEIN stations have to be connected to the core communication network and they also have to be powered by stable energy sources, such as large solar energy harvesters and the power grid. As illustrated in Fig.1.2,
- 22. 8 1 Data and Energy Integrated Communication Networks: An Overview WET Range (Isotropic Antenna) WIT Range (Isotropic Antenna) WIT-UE-1 WIT WIT WIT WET WET WIT-Relay-1 WIT-Relay-2 WET-Station-3 WET-Station-2 WET WIT-UE-2 DEIN- Station-1 WET-UE-1 WIT WIT DEIN-UE-1 WIT WET DEIN Devices for IoT Wide Beam for Integrated Data and Energy Multicast (Directional Antenna) WET-UE-2 Narrow beam for point-to- point WET (Directional Antenna) WIT DEIN-UE-2 WIT WIT DEIN- Station-2 WET-Station-1 WET Fig. 1.2 Ubiquitous architecture of heterogeneous DEIN DEIN-Station-1 may satisfy the integrated data and energy requests from the IoT devices and those from DEIN-UE-1. However, as we have discussed in Sect.1.4, the reliable WET range is far shorter than the reliable WIT range, as exemplified in Fig.1.2. As a result, some blind areas cannot be adequately covered by WET of DEIN stations. Furthermore, some dedicated WET [35] stations are also deployed in order to supply energy to the devices roaming in these blind areas. These WET stations are only connected to energy sources, but they do not have to be connected to the core communication network. As a result, they are dedicated for satisfying the UEs’ charging requests. For instance, as shown in Fig.1.2, three WET stations are deployed in order to supply energy to the UEs beyond the WET range of the DEIN stations. Apart from DEIN stations and WET stations, there are still many conventional communication stations in heterogeneous DEINs, namely the classic femto-cellular stations, pico-cellular stations and macro-cellular stations [36]. These communica- tion stations have different levels of transmit power and coverage, which results in obvious heterogeneity in DEINs. Sometimes, low-cost relay stations are also deployed for forwarding the data packets to cell-edge UEs, as illustrated in Fig.1.2. However, small cellular stations and relay stations [37] are only capable of emitting RF signals at a limited power. They are not suitable for carrying out sophisticated WET tasks. Therefore, they are regarded as a dedicated communication infrastruc- ture. 1.5.2 Heterogeneous User Equipment Apart from the heterogeneous infrastructure, our DEINs have to accommodate both charging and communication requests from diverse types of UEs. We generally have
- 23. 1.5 Ubiquitous Architecture of the DEIN 9 three types of UEs in DEINs, namely the WIT UEs, the WET UEs and the DEIN UEs [38], as exemplified in Fig.1.2. WIT UEs only require downlink and uplink data transmission in DEINs. Since these UEs are always powered by stable energy sources, they do not request any wireless charging from the DEIN stations. Laptops and tablets are typical WIT UEs, which are either powered by high-capacity batteries or are connected to the power grid. For example, as illustrated in the left part of Fig.1.2, WIT-UE-1 receives its requested data from DEIN-Station-1 with the aid of two WIT relay stations, while WIT-UE-2 may consume its own energy for powering its uplink information transmission. By contrast, since WET UEs are not powered by stable energy sources, they have to request additional energy supply either from the DEIN stations or from the WET stations in order to support their basic functionalities, such as uplink informa- tion transmissions and energy-consuming computations [39]. For instance, although WET-UE-1 is beyond the WIT range of DEIN-Station-1, it may still establish reli- able uplink transmissions with DEIN-Station-1 by exploiting the additional energy received from WET-Station-1, as exemplified in Fig.1.2. Similarly, the uplink trans- mission of WET-UE-2 towards DEIN-Station-2 is powered by DEIN-Station-2 itself. Miniature-sized IoT devices are typical WET UEs, since their functionalities are lim- ited by the amount of energy stored in their batteries. Furthermore, some UEs simultaneously request data and energy transmissions, which are regarded as DEIN UEs [40]. For instance, in the right cell of Fig.1.2, DEIN- UE-1 simultaneously receives its requested data and energy from DEIN-Station-2, while DEIN-UE-2 also simultaneously requests both downlink data transmission and wireless charging. However, since DEIN-UE-2 is beyond the WET range of DEIN- Station-1, it can only receive the requested data from DEIN-Station-2, but it can receive energy from WET-Station-1. This energy may be exploited for supporting DEIN-UE-2’s uplink data transmission to its associated DEIN-Station-2. Sometimes, the functionalities of WIT relay stations are also limited by their energy supply, especially when the WIT relay stations rely on energy gleaned from batteries or harvested from renewable sources. As a result, they also need wireless charging from DEIN stations or WET stations for powering their data packet for- warding actions [41]. As a result, WIT relay stations can also be regarded as special “DEIN UEs”. As portrayed in the left cell of Fig.1.2, both data and energy are simul- taneously transferred from DEIN-Station-1 to WIT-Relay-1. The energy harvested by WIT-Relay-1 may be further exploited for forwarding the data packets to the next hop. Since WIT-Relay-2 is beyond the WET range of DEIN-Station-2, it has to request WET from the nearby WET-Station-2 and WET-Station-3. After receiving the data packets from WIT-Relay-1 and gleaning sufficient energy from the WET stations, the data packets are finally forwarded to their destination WIT-UE-1 by WIT-Relay-2.
- 24. 10 1 Data and Energy Integrated Communication Networks: An Overview 1.5.3 Heterogeneous Techniques for WIT and WET Our DEIN architecture has to accommodate both the WET and WIT in the same RF spectral band. Although the WET and WIT both rely on the RF signal, they still have distinctive features, as summarised in Sect.1.4. Therefore, in order to satisfy the UEs’ information and energy requests, the coexistence of WET and WIT in the DEIN results in natural heterogeneity. In order to guarantee the seamless WIT coverage, different techniques have to be invoked. As exemplified in Fig.1.2, when the omnidirectional antennas are adopted, the boundary of a DEIN cell is determined by the WIT range of a DEIN sta- tion. As a result, the UEs residing within the WIT range of a DEIN station may receive their requested information via a single-hop cellular link. Furthermore, these UEs are also capable of uploading information to their associated DEIN stations. Observe from Fig.1.2 that WIT-UE-2, DEIN-UE-1 and DEIN-UE-2 all receive their requested information from the downlink WIT of their associated DEIN stations, while WET-UE-2 and DEIN-UE-2 both upload their information to their associated DEIN-Station-2. By contrast, in order to satisfy the information request of a UE sitting beyond the WIT range of a DEIN station, multiple relay stations have to be relied upon for forwarding the information from the DEIN station to the requester or in a reverse direction via the multi-hop transmissions, such as the downlink trans- mission from DEIN-Station-1 to WIT-UE-1 of Fig.1.2. In addition, by exploiting the extra energy supplied by the WET stations, a UE beyond the WIT range of a DEIN station is also capable of uploading data to the DEIN station [39], such as the uplink transmission of WET-UE-1 to DEIN-Station-1 in Fig.1.2, which is powered by WET-Station-1. If we further look into the wireless charging actions in DEINs, various WET tech- niques have to be invoked for satisfying diverse charging requirements. As illustrated in Fig.1.2, a DEIN station’s WET range is much shorter than its WIT range, when the omnidirectional antenna is adopted. The reason is that for the successful WET, the energy harvesting circuit of the receiver can only be activated by a high received energy. As a result, the WET is more sensitive to the wireless channel attenuation, which is dominated by the path loss. If a WET UE resides within the WET range of a DEIN station, it may successfully harvest energy from the RF signal emitted by this DEIN station. By contrast, when a WET UE is beyond the WET range, it has to request energy from its nearby WET station. Directional antennas may enable a DEIN station to focus its energy in the main-lobe, which substantially increases the long-range WET efficiency in the direction of the main-lobe. However, the resultant energy loss in the side-lobes may significantly reduce the WET efficiency in other directions. If directional antennas are adopted by the DEIN stations, they may form a narrow energy beam [42] for charging the WET UE beyond the WET range, which is characterised by the omnidirectional antennas. As exemplified in the right DEIN cell of Fig.1.2, WET-UE-2 is still capable of receiving energy from the dedicated narrow energy beam forming by DEIN-Station-2. Furthermore, IoT devices will be pervasively deployed in the near future. Our heterogeneous DEINs are also responsible for satisfying both of their communication
- 25. 1.5 Ubiquitous Architecture of the DEIN 11 and energy demands. IoT devices normally are clustered in a specific area in order to jointly carry out their tasks. As a result, for the sake of satisfying the charging requests of the multiple IoT devices, DEIN stations may form wide-angle energy beams for covering the cluster of requesters [43]. This technique may be regarded as energy multicast. Moreover, this wide beam is also capable of transferring information and energy together to the multiple requesters. References 1. A. Zanella, N. Bui, A. Castellani, L. Vangelista, M. Zorzi, Internet of things for smart cities. IEEE Internet Things J. 1(1), 22–32 (2014) 2. A. Alrawais, A. Alhothaily, C. Hu, X. Cheng, Fog computing for the internet of things: security and privacy issues. IEEE Internet Comput. 21(2), 34–42 (2017) 3. B.H. Choi, V.X. Thai, E.S. Lee, J.H. Kim, C.T. Rim, Dipole-coil-based wide-range inductive power transfer systems for wireless sensors. IEEE Trans. Ind. Electron. 63(5), 3158–3167 (2016) 4. V. Jiwariyavej, T. Imura, Y. Hori, Coupling coefficients estimation of wireless power transfer system via magnetic resonance coupling using information from either side of the system. IEEE J. Emerg. Sel. Top. Power Electron. 3(1), 191–200 (2015) 5. P.S. Riehl, A. Satyamoorthy, H. Akram, Y.C. Yen, J.C. Yang, B. Juan, C.M. Lee, F.C. Lin, V. Muratov, W. Plumb, P.F. Tustin, Wireless power systems for mobile devices supporting inductive and resonant operating modes. IEEE Trans. Microw. Theory Tech. 63(3), 780–790 (2015) 6. M. Pinuela, D.C. Yates, S. Lucyszyn, P.D. Mitcheson, Maximizing DC-to-load efficiency for inductive power transfer. IEEE Trans. Power Electron. 28(5), 2437–2447 (2013) 7. H. Liu, Maximizing efficiency of wireless power transfer with resonant inductive coupling (2011) 8. C. Zheng, H. Ma, J.S. Lai, L. Zhang, Design considerations to reduce gap variation and mis- alignment effects for the inductive power transfer system. IEEE Trans. Power Electron. 30(11), 6108–6119 (2015) 9. Z. Yan, Y. Li, C. Zhang, Q. Yang, Influence factors analysis and improvement method on efficiency of wireless power transfer via coupled magnetic resonance. IEEE Trans. Magn. 50(4), 1–4 (2014) 10. J.M. Miller, O.C. Onar, M. Chinthavali, Primary-side power flow control of wireless power transfer for electric vehicle charging. IEEE J. Emerg. Sel. Top. Power Electron. 3(1), 147–162 (2015) 11. A. Kurs, A. Karalis, R. Moffatt, J.D. Joannopoulos, P. Fisher, M. Soljačić, Wireless power transfer via strongly coupled magnetic resonances. Science 317(5834), 83–86 (2007) 12. H. Hwang, J. Moon, B. Lee, C.H. Jeong, S.W. Kim, An analysis of magnetic resonance coupling effects on wireless power transfer by coil inductance and placement. IEEE Trans. Consum. Electron. 60(2), 203–209 (2014) 13. M.Q. Nguyen, D. Plesa, S. Rao, J.C. Chiao, A multi-input and multi-output wireless energy transfer system, in 2014 IEEE MTT-S International Microwave Symposium (IMS2014) (2014), pp. 1–3 14. M.Q. Nguyen, Y. Chou, D. Plesa, S. Rao, J.C. Chiao, Multiple-inputs and multiple-outputs wireless power combining and delivering systems. IEEE Trans. Power Electron. 30(11), 6254– 6263 (2015) 15. J.H. Kim, H.Y. Yu, C. Cha, Efficiency enhancement using beam forming array antenna for microwave-based wireless energy transfer, in 2014 IEEE Wireless Power Transfer Conference (2014), pp. 288–291
- 26. 12 1 Data and Energy Integrated Communication Networks: An Overview 16. W.C. Brown, R.H. George, Rectification of microwave power. IEEE Spectr. 1(10), 92–97 (1964) 17. W.C. Brown, The history of power transmission by radio waves. IEEE Trans. Microw. Theory Tech. 32(9), 1230–1242 (1984) 18. W.C. Brown, Status of the microwave power transmission components for the solar power satellite. IEEE Trans. Microw. Theory Techn. 29(12), 1319–1327 (1981) 19. W. Brown, Recent advances in key microwave components that impact the design and deploy- ment of the solar power satellite system, in 1984 Antennas and Propagation Society Interna- tional Symposium, vol. 22 (1984), pp. 339–340 20. R.M. Dickinson, Performance of a high-power, 2.388-GHz receiving array in wireless power transmission over 1.54 km, in 1976 IEEE-MTT-S International Microwave Symposium (1976), pp. 139–141 21. P.S. Yedavalli, T. Riihonen, X. Wang, J.M. Rabaey, Far-field RF wireless power transfer with blind adaptive beamforming for internet of things devices. IEEE Access 5, 1743–1752 (2017) 22. T. Mitani, S. Kawashima, T. Nishimura, Analysis of voltage doubler behavior of 2.45-GHz voltage doubler-type rectenna. IEEE Trans. Microw. Theory Tech. 65(4), 1051–1057 (2017) 23. X. Chu, Q. Sun, J. Wang, P. Lu, W.X.X. Xu, Generating a Bessel-Gaussian beam for the application in optical engineering. Scientific Reports 5(6), 18665 (2015) 24. N. Moraitis, P.N. Vasileiou, C.G. Kakoyiannis, A. Marousis, A.G. Kanatas, P. Constantinou, Radio planning of single-frequency networks for broadcasting digital TV in mixed-terrain regions. IEEE Antennas Propag. Mag. 56(6), 123–141 (2014) 25. M. Morelli, M. Moretti, A maximum likelihood approach for SSS detection in LTE systems. IEEE Trans. Wirel. Commun. 16(4), 2423–2433 (2017) 26. L. Sanabria-Russo, J. Barcelo, B. Bellalta, F. Gringoli, A high efficiency MAC protocol for WLANs:providingfairnessindensescenarios.IEEE/ACMTrans.Netw.25(1),492–505(2017) 27. R. Ford, M. Zhang, M. Mezzavilla, S. Dutta, S. Rangan, M. Zorzi, Achieving ultra-low latency in 5G millimeter wave cellular networks. IEEE Commun. Mag. 55(3), 196–203 (2017) 28. T. Zhang, L. An, Y. Chen, K.K. Chai, Aggregate interference statistical modeling and user outage analysis of heterogeneous cellular networks, in 2014 IEEE International Conference on Communications (ICC) (2014), pp. 1260–1265 29. J. Hu, L.L. Yang, L. Hanzo, Distributed multistage cooperative-social-multicast-aided content dissemination in random mobile networks. IEEE Trans. Veh. Technol. 64(7), 3075–3089 (2015) 30. J. Hu, L.L. Yang, H.V. Poor, L. Hanzo, Bridging the social and wireless networking divide: information dissemination in integrated cellular and opportunistic networks. IEEE Access 3, 1809–1848 (2015) 31. J. Hu, L.L. Yang, L. Hanzo, Delay analysis of social group multicast-aided content dissemina- tion in cellular system. IEEE Trans. Commun. 64(4), 1660–1673 (2016) 32. J. Hu, L.L. Yang, L. Hanzo, Energy-efficient cross-layer design of wireless mesh networks for content sharing in online social networks. IEEE Trans. Veh. Technol. PP(99), 1–1 (2017) 33. K. Yang, Q. Yu, S. Leng, B. Fan, F. Wu, Data and energy integrated communication networks for wireless big data. IEEE Access 4, 713–723 (2016) 34. M.D. Renzo, W. Lu, System-Level analysis and optimization of cellular networks with simul- taneous wireless information and power transfer: Stochastic geometry modelling. IEEE Trans. Veh. Technol. 66(3), 2251–2275 (2017) 35. D.H. Chen, Y.C. He, Full-Duplex secure communications in cellular networks with downlink wireless power transfer. IEEE Trans. Commun. 66(1), 265–277 (2018) 36. J. An, K. Yang, J. Wu, N. Ye, S. Guo, Z. Liao, Achieving sustainable ultra-dense heterogeneous networks for 5g. IEEE Commun. Mag. 55(12), 84–90 (2017) 37. X. Liu, Z. Li, C. Wang, Secure decode-and-forward relay SWIPT systems with power splitting scheme. IEEE Trans. Veh. Technol. pp. 1–1 (2018) 38. Y. Zhao, D. Wang, J. Hu, K. Yang, H-AP Deployment for joint wireless information and energy transfer in smart cities. IEEE Trans. Veh. Technol. pp. 1–1 (2018) 39. J. Hu, Y. Xue, Q. Yu, K. Yang, A joint time allocation and UE scheduling algorithm for full- duplex wireless powered communication networks. in 2017 IEEE 86th Vehicular Technology Conference (VTC-Fall) (2017), pp. 1–5
- 27. References 13 40. K. Lv, J. Hu, Q. Yu, K. Yang, Throughput maximization and fairness assurance in data and energy integrated communication networks. IEEE Internet of Things Journal, 5(2), 636–644 (2018) 41. Y. Ye, Y. Li, D. Wang, F. Zhou, R.Q. Hu, H. Zhang, Optimal transmission schemes for DF relaying networks using SWIPT. IEEE Trans. Veh. Technol. pp. 1–1 (2018) 42. Y. Zeng, B. Clerckx, R. Zhang, Communications and signals design for wireless power trans- mission. IEEE Trans. Commun. 65(5), 2264–2290 (2017) 43. Q. Li, Q. Zhang, J. Qin, Robust tomlinson-harashima precoding with gaussian uncertainties for SWIPT in MIMO broadcast channels. IEEE Trans. Signal Process. 65(6), 1399–1411 (2017)
- 28. Chapter 2 Fundamental of Integrated WET and WIT Abstract In order to realise integrated wireless energy transfer (WET) and wireless information transfer (WIT), we have to revisit the information theory for finding its performance limits, while redesigning the transceiver architecture in the physi- cal layer for practical implementation. As a result, in this chapter, we impose the energy delivery requirement on the channel output sequence, when maximising the mutual information. The rate-energy tradeoff is studied from the information the- oretical perspective for both the discrete-input-discrete-output channel and for the continuous-input-continuous-output channel. Then we provide an overview on the transceiver architecture in the physical layer by considering diverse signal splitters, namely the spatial splitter, the power splitter and the time switcher. The resultant integrated WET and WIT performance is then evaluated for different transceiver architectures. Keywords Continuous-Input-Continuous-Output-Channel Discrete-Input-Discrete-Output Channel · Information Theory · Integrated WET and WIT · Multiple-Input-Multiple-Output (MIMO) system · Mutual Information · Power Splitting · Rate-Energy Tradeoff · RF based Wireless Charging · Simultaneous Wireless Information and Power Transfer (SWIPT) Spatial Splitting · Time Switching · Transceiver Architecture · Wireless Energy Transfer (WET) · Wireless Information Transfer (WIT) In this chapter, we will focus on the fundamental of the physical layer for implement- ing the integrated wireless energy transfer (WET) and wireless information transfer (WIT) of the point-to-point link. First of all, the information theoretical essence of the integrated WET and WIT will be introduced. Key enabling modules of the generic transceiver architecture for the integrated WET and WIT will also be included. Then, we will cover the architectures of several popular receivers equipped with multi- ple antennas for simultaneous information and energy reception, namely the ideal receiver, the spatial splitting based receiver, the power splitting based receiver and the time switching based receiver. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd., part of Springer Nature 2018 J. Hu and K. Yang, Data and Energy Integrated Communication Networks, SpringerBriefs in Computer Science, https://doi.org/10.1007/978-981-13-0116-2_2 15
- 29. 16 2 Fundamental of Integrated WET and WIT 2.1 Information Theoretical Essence As previously discussed in Sect. 1.4, WET and WIT entail several conflicting speci- fications, when they are coordinated in the same radio frequency (RF) spectral band. As a result, theoretical investigations have to be carried out in order to reveal the underlying relationship between the WET and WIT in data and energy integrated communication networks (DEINs), which may provide researchers and engineers with further valuable insights on improving the system-level performance of DEINs. In this section, we will explore the information theoretical essence for DEIN and reveal the natural contradiction between WET and WIT from an information theo- retical perspective, which requires further efforts for jointly designing energy and information transfer. Note that the information theoretical exploration remain valid not only for the integratedWETandWIToperatingintheRFspectralband,butforallotherintegrated data and power transfer scenarios, such as power line communication [1] and power over Ethernet technique [2]. Classic information theoretical channel capacity analysis has been dedicated to maximising the mutual information under the constraint of specific input signals. By contrast, the pioneering work of Gastpar [3] has attempted to maximise the mutual information under the constraint of specific output signals, which aims for controlling the interference imposed by a communicating pair on other peers. This piece of work may provide us with valuable hints for finding the performance limits of integrated data and energy transfer. 2.1.1 Discrete-Input-Discrete-Output Memoryless Channel We first consider the classic Discrete-Input-Discrete-Output Memoryless Channel (DIDO-MC) of Fig.2.1. If a DIDO-MC has a single input symbol x and a single output symbol y, the transition probability of this DMC channel may be expressed by the probability function of pY|X (y|x). All the legitimate values of the input symbol x constitute the input codebook X, while all the legitimate values of the output symbol y constitute the output codebook Y. Furthermore, given a specific output symbol y, the energy carried by it can be represented by the non-negative function g(y). Let us now move on to the n-dimensional random input, which is expressed by the vector Xn = (X1, X2, . . . , Xn). All the random symbols in the vector Xn DIDO-MC Single input symbol Single output symbol Fig. 2.1 Discrete-input-discrete-output memoryless channel
- 30. 2.1 Information Theoretical Essence 17 are independent of one another. If a sample of the n dimensional random input is xn = (x1, x2, . . . , xn), its corresponding occurrence probability can be expressed as pXn (xn ) = n i=1 pX (xi ), where Xn represents the n-dimensional codebook con- taining all the possible values of the random output Xn and pX (xi ) represents the probability of the symbol xi being generated. When the input sample xn is trans- mitted by the information source, the corresponding output at the information des- tination is denoted by the vector yn = (y1, y2, . . . , yn). This sequence of symbols occurs with a probability of pYn (yn ) = n i=1 pY (yi ), where Yn represents the n- dimensional codebook containing all the possible values of the random output Yn and p†(yi ) = xi ∈X pX (xi )pY|X (yi |xi ) represents the probability of the symbol yi being received by the information destination. The energy carried by the sequence yn can be calculated by the non-negative function g(yn ). Assuming a random output sequence Yn = (Y1, Y2, . . . , Yn), the average energy carried by this n-dimensional sequence can be formulated as E[g(Yn )] = yn∈Yn g(yn ) · pYn (yn ). (2.1) As a result, the WIT performance limit can be formulated as the following optimi- sation problem: Objective: max pXn (xn) I (Xn ; Yn ), (2.2) Subject to: 1 n · E[g(Yn )] ≥ β, (2.2a) where I (Xn ; Yn ) is the average mutual information between the n-dimensional input symbol sequence and its output counterpart. Given the transition probabili- ties pY|X (Y|X) of the DIDO-MC, the optimisation problem (2.2) aims for finding the optimal n-dimensional information source, which is represented by the probabil- ities pXn (xn ) of the symbol sequences being generated by this information source. However, this optimisation problem is subject to the condition (2.2a), suggesting that the average energy carried by a single output symbol has to be higher than a threshold β in order to satisfy the basic WET requirement. Substituting the optimal pXn (xn ) into the objective (2.2), we may derive the channel capacity Cn(β), when the input is an n-dimensional symbol sequence. Note that Cn(β) is a function of the energy constraint β. Furthermore, the normalised channel capacity subject to the energy constraint β can be further formulated as C(β) = sup n 1 n · Cn(β), [bit/symbol] (2.3) which may be regarded as the rate-energy function. Observe from (2.3) that the rate- energy function is a natural extension of the classic channel capacity concept. This
- 31. 18 2 Fundamental of Integrated WET and WIT function only depends on a channel’s statistical property and on the requirement of the energy harvested, but it does not rely on the information source. The rate-energy functions of some simple binary channels will now be studied for illustrating the above information theoretical methodology. By adopting the classic ON-OFF-Keying (OOK) as our modulation scheme, for a binary random input, we may assume that the symbol ‘0’ does not carry any energy as g(0) = 0, while the symbol ‘1’ carries a single unit of energy as g(1) = 1. The probability distribution of the input binary symbols is denoted as pX = {pX (0) = q, pX (1) = q = 1 − q}. As illustrated in Fig.2.2a, in a noiseless channel, an input binary symbol, either ‘0’ or ‘1’, can be correctly output. Therefore, neither energy loss nor energy gain exist in the noiseless channel. The mutual information of the noiseless channel is IN (X; Y) = −q log2 q − q log2 q. Without any energy constraint, we can maximise IN (X; Y) by letting q = 1/2. Accordingly, the maximum mutual information is IN,max(X; Y) = 1 bit/symbol.Moreover,theaverageenergycarriedbythecorrespondingoutputsymbol is βth = 1/2 unit. Furthermore, the maximum energy carried by the output symbol at the information destination Y is βmax = 1 unit, when the source never send the symbol ‘0’, namely q = 0. In a nutshell, given a specific energy transfer requirement β, the maximum achievable information rate of the noiseless channel is formulated as [4] CN (β) = 1, 0 ≤ β ≤ 1 2 , H2(β), 1 2 β ≤ 1, [bit/symbol], (2.4) where H2(β) = −β log2 β − (1 − β) log2(1 − β)isthebinaryentropyfunctionwith respect to β. As shown in Fig.2.2b, in a Z channel, the input binary symbol ‘0’ can be correctly output for certain. By contrast, the input binary symbol ‘1’ can be erroneously output as the symbol ‘0’ with a probability ω, while it can be correctly output with a probability ω = 1 − ω. In this channel, the energy carried by the symbol ‘1’ may be lost during the transmission, due to the channel attenuation. By contrast, since no additional energy is supplied to the system, the input symbol ‘0’ does not have a chance to become the energy carrier symbol ‘1’. The mutual information of the Z channel is expressed as Input Output 0 1 0 1 (a) Noiseless Channel 0 1 0 1 (c) Symmetric Channel Input Output 0 1 0 1 (b) Z Channel Input Output Fig. 2.2 Three typical binary channels
- 32. 2.1 Information Theoretical Essence 19 IZ (X; Y) = −(q + qω) log2(q + qω) − q ω log2 q + qω log2 ω. (2.5) Without any energy constraint, we can maximise IZ (X; Y) by letting q = 1 − ω 1 1−ω 1 + (1 − ω)ω ω 1−ω . (2.6) Accordingly, the maximum mutual information is IZ,max = log2(1 + ωω ω ω ). More- over, the average energy carried by the corresponding output symbol is formulated as βth = ω 1 ω 1 + ωω ω ω [unit]. (2.7) When the source only sends the energy carrier symbol ‘1’, namely q = 0, the maxi- mum energy carried by the output symbol at the information destination is βmax = ω unit. In a nutshell, given a specific energy transfer requirement β, the maximum achievable information rate of the Z channel can be formulated as [4] CZ (β) = log 1 − ω 1 1−ω + ω ω 1−ω , 0 ≤ β ≤ (1 − ω)π∗ H2(β) − β 1−ω H2(ω), (1 − ω)π∗ β ≤ 1 − ω, [bit/symbol], (2.8) where the variable π∗ is given by [4] π∗ = ω ω 1−ω 1 + (1 − ω)ω ω 1−ω . (2.9) As portrayed in Fig.2.2c, in a symmetric channel, both the input binary symbols ‘0’ and ‘1’ can be erroneously delivered to the output end with a probability of ω, while they can be correctly delivered with a probability of ω = 1 − ω. Apart from the energy loss incurred by the transition from the input symbol ‘1’ to the output symbol ‘0’, the energy of the interference may change the input symbol ‘0’ to the output symbol ‘1’, which results in the energy gain at the information destination. The mutual information of the symmetric channel is expressed as IS(X; Y) = − (q ω + qω) log2(q ω + qω) − (qω + q ω) log2(qω + q ω) + ω log2 ω + ω log2 ω. (2.10) Without any energy constraint, we can maximise IS(X; Y) by letting q = 1/2. Accordingly, the maximum mutual information is IS,max = 1 + ω log2 ω + ω log2 ω. Moreover, the average energy carried by the corresponding output symbol is βth = 1/2 unit. When the source only sends the energy carrier symbol ‘1’, namely q = 0,
- 33. Other documents randomly have different content
- 34. Mr. Westbrook. It don't seem to me—I can't remember for sure. Mr. Ball. I offer this exhibit, Westbrook No. D. Mr. Westbrook. Now, I did, when I left this scene, I turned this jacket over to one of the officers and I went by that church, I think, and I think that would be on 10th Street. Mr. Ball. I show you Commission Exhibit 162, do you recognize that? Mr. Westbrook. That is exactly the jacket we found. Mr. Ball. That is the jacket you found? Mr. Westbrook. Yes, sir. Mr. Ball. And you turned it over to whom? Mr. Westbrook. Now, it was to this officer—that got the name. Mr. Ball. Does your report show the name of the officer? Mr. Westbrook. No, sir; it doesn't. When things like this happen— it was happening so fast you don't remember those things. Mr. Ball. Then, it was after that you went over to 10th and Patton? Mr. Westbrook. To 10th and Patton—yes, sir. Mr. Ball. And from there you went to the theatre? Mr. Westbrook. Yes; from there we went to the theatre, and I can't remember exactly how that I got back with Bob Barrett and Stringer, but anyway, we got together again—probably at 10th and Patton. Mr. Ball. Were you in the personnel office at a time that a gun was brought in? Mr. Westbrook. Yes, sir; it was brought to my office when it shouldn't have been. Mr. Ball. But it was brought to your office?
- 35. Mr. Westbrook. Yes; it was. Mr. Ball. And it was marked by some officer? Mr. Westbrook. It was marked by Officer Jerry Hill and a couple or three more, and when they come in with the gun, I just went on down and told Captain Fritz that the gun was in my office and he sent a man up after it. I didn't take it down. Mr. Ball. Did you see McDonald mark it? Mr. Westbrook. He possibly could have—he was in there. Mr. Ball. Did you see the gun unloaded? Mr. Westbrook. No, sir; I didn't see it unloaded. When I saw it, the gun was laying on Mr. McGee's desk and the shells were out of it. Mr. Ball. Did you look at any of the shells? Mr. Westbrook. No, sir. Mr. Ball. Did you look the gun over? Mr. Westbrook. No, sir. Mr. Ball. Do you have any questions? Mr. Ely. Yes; I have one. Captain, you mentioned that you had left orders for somebody to take the names of everybody in the theatre, and you also stated you did not have this list; do you know who has it? Mr. Westbrook. No; possibly Lieutenant Cunningham will know, but I don't know who has the list. Mr. Ely. That's all. Mr. Westbrook. And I'm sorry that I'm so vague on names, but it's just—the only reason that I knew Sergeant Stringer, I think, that day he worked with me. Mr. Ball. Do you have any questions?
- 36. Mr. Stern. No, sir. Mr. Ball. I think that's all. Thank you very much, captain. Mr. Westbrook. Thank you, sir, Mr. Ball, it has been a pleasure.
- 37. TESTIMONY OF ELMER L. BOYD The testimony of Elmer L. Boyd was taken at 11 a.m., on April 6, 1964, in the office of the U. S. attorney, 301 Post Office Building, Bryan and Ervay Streets, Dallas, Tex., by Messrs. Joseph A. Ball, John Hart Ely and Samuel A. Stern, assistant counsel of the President's Commission. Dr. Alfred Goldberg, historian, was present. Mr. Ball. Mr. Boyd, do you swear that the testimony you are about to give before this Commission shall be the truth, the whole truth, and nothing but the truth, so help you God? Mr. Boyd. I do. Mr. Ball. Will you state your name, please? Mr. Boyd. Elmer L. Boyd. Mr. Ball. And what is your occupation? Mr. Boyd. I am a detective in the homicide and robbery bureau for the Dallas Police Department. Mr. Ball. You received a letter asking you to appear here today, didn't you? Mr. Boyd. I think they received one over at the office and they notified me. Mr. Ball. And you have been told the purpose of this investigation is to inquire into the facts and circumstances surrounding the assassination of President Kennedy? Mr. Boyd. Yes, sir.
- 38. Mr. Ball. I'm going to ask you what you learned during the course of your investigation. Mr. Boyd. All right. Mr. Ball. Now, can you tell me something about yourself, where you were born and where you went to school and what you have done most of your life? Mr. Boyd. Well, yes, sir. I can tell you I was born in Navarro County—the particular place was Blooming Grove, Tex., and it's about 15 miles west of Corsicana, and I was raised up about 7 miles north of there. I attended school, well, I started at a little country school—it was Pecan, was the name of the school. I went there 2 years and then they sent me to Blooming Grove and I started to school in my second grade. The reason I was in the second grade—I had to go through a primer before I got in the first grade—I didn't fail—I just had to go through this primer before I got in the first grade, and I graduated from high school at Blooming Grove in 1946 and I went into the Navy and served for 2 years, I believe I served about 22 months in the Navy—I joined and I went through boot training at San Diego, went from there to Newport, R. I., and caught my first ship, the USS Kenneth D. Bailey. I don't recall just how many months I spent on that—somewhere around 15 or 16 months, I've forgotten, and then they sent me to—I transferred from that ship and went on the USS Cone, that's another destroyer [spelling] C-o-n- e, and along about the first part of January, I believe, in 1948, they transferred me to Pensacola where I caught my third destroyer, the USS Forrest Royal, and we operated in and out of there until I got out of the Navy, and I believe it was about the first day of April 1948, when I was discharged, and I came to Dallas and I have been here in Dallas ever since. I went to work on the police department May 19, 1952. Prior to that I worked, I believe, about 3 years for the gas company and I started out reading gas meters, and then I went into collecting, and I was a collector for the gas company when I came on the police department. I think I worked a couple of more places before then—
- 39. one for a printing company down here on Cockrell, down here by Sears Roebuck for a while, but I didn't stay there long. Mr. Ball. How long have you been in homicide? Mr. Boyd. I came in there on October 15, I believe, in 1957. Mr. Ball. November 22, 1963, what were your hours of duty? Mr. Boyd. Well, my hours of duty on November 22, 1963, I believe, was 4 to midnight. Mr. Ball. So, on that day you went to work earlier? Mr. Boyd. Yes, sir; I did. Mr. Ball. What time? Mr. Boyd. I came to work at 9 o'clock. Is it all right for me to go by this? Mr. Ball. I see you have there a report that is entitled Report on Officer's Duty in Regard to the President's Murder, R. M. Sims, No. 629, and E. L. Boyd, No. 840. Mr. Boyd. Yes; we are partners. Mr. Ball. Did you prepare that report yourself? Mr. Boyd. He and I together prepared it. Mr. Ball. When did you prepare it? Mr. Boyd. Let me see—the last part of November—I'm not sure of the date. Mr. Ball. Was it within a week after the events took place that are recorded there? Mr. Boyd. I would say so; yes. Mr. Ball. You dictated it to a secretary? Mr. Boyd. Well, I wrote it out in longhand and carried it to the secretary and she typed it up.
- 40. Mr. Ball. It was written out in your longhand? Mr. Boyd. Yes, sir. Mr. Ball. Do you have those longhand notes? Mr. Boyd. No, sir; I do not. Mr. Ball. This report has already been attached to Officer Sims' deposition as Exhibit A, so we have read it. Mr. Boyd. Yes, sir. Mr. Ball. During the course of your work, did you make notes of what you were doing in a notebook? Mr. Boyd. Well, I made notes, and I believe I had a notebook. Mr. Ball. Did you make it a habit of carrying a notebook with you? Mr. Boyd. Yes, sir. Mr. Ball. When you work? Mr. Boyd. Yes. Mr. Ball. And you just jot things down as they occur? Mr. Boyd. Yes, sir. Mr. Ball. Do you have that notebook with you? Mr. Boyd. No; I do not. Mr. Ball. Do you know where it is? Mr. Boyd. No, sir; right offhand, I don't know where it is. Part of the time, you know, I just took a sheet of paper and put down the particular times, you know, and after I fixed this—I don't recall what I did with it. I may have torn it up. Mr. Ball. You didn't have a regular notebook that you kept with you at all times?
- 41. Mr. Boyd. I had a regular notebook, but I didn't put everything in it, I'm sure. Mr. Ball. This notebook that you had on November 22, 1963, have anything in it with respect to what you did on the 22d and the 23d of November? Mr. Boyd. Of 1963—I don't recall if I have these showups in there or not—it seems like I did. Mr. Ball. Do you have it with you? Mr. Boyd. No; I do not. Mr. Ball. Can you get it for me? Mr. Boyd. I probably could if I have it. Mr. Ball. Will you look it up? Mr. Boyd. I will look for it. Mr. Ball. I'll be down to the police department tomorrow morning at 10 o'clock and will you look it up between now and then and then let me see it if you still have it? Mr. Boyd. All right. Mr. Ball. I'll be up there in your department—near Captain Fritz' office. Mr. Boyd. What time—at 10 o'clock? Mr. Ball. At 10 o'clock in the morning. Mr. Boyd. I'll be there—I come on at 10. Mr. Ball. You come on at 10? Mr. Boyd. Yes. Mr. Ball. Then, I'll see you in the morning. Mr. Boyd. All right.
- 42. Mr. Ball. On this morning of November 22, you had been ordered to work early; why was that? Mr. Boyd. Well, President Kennedy was coming into Dallas and I was assigned to work with Captain Fritz and Detective Sims out at the Trade Mart. Mr. Ball. Where did you hear that the President had been shot? Mr. Boyd. Yes; I heard that. Mr. Ball. You heard that over the radio, didn't you? Mr. Boyd. Well, I believe it was around 12:40 when Chief Stevenson called and he talked to Captain Fritz out at the Trade Mart and he told him that—Captain Fritz told me that Chief Stevenson told him that the President had been involved in an accident down at the triple underpass and was on his way to Parkland. Mr. Ball. Did you go over there? Mr. Boyd. When we got out of the car, we checked, I believe, with—Mr. Sims called in on the radio and they told us he had been shot and we went to Parkland Hospital and pulled up to the emergency and saw there were a lot of people out there, but we saw Chief Curry out in front of the emergency there and he advised us to go back down to the scene of where we thought the shooting had occurred, down at the Texas Book Depository, and Mr. Sims and Captain Fritz and Sheriff Decker was also out there, and he rode back down with us. Mr. Ball. And you went to the School Depository Building, did you? Mr. Boyd. Yes, sir. Mr. Ball. And you were told by Chief Curry to go to the School Depository Building at that time? Mr. Boyd. Yes; down at the scene and that's where we had heard that they thought that the shot came from—from the Texas Book Store.
- 43. Mr. Ball. Where were you when you first heard that? Mr. Boyd. We were at the Trade Mart when we heard that— pulling out—we were on our way to Parkland Hospital from the Trade Mart, pulling out in the car. Mr. Ball. Now, when you arrived down here at the building, what did you do? Mr. Boyd. Well, we went outside the building and we made two or three stops going up, you know, at different floors, and when we got up to the top floor—I believe it was the top one—I think it's the seventh floor, and someone called us and said they had found some hulls, rifle hulls, down on the sixth floor, I believe it was the sixth floor. Mr. Ball. And you were with whom at that time? Mr. Boyd. I was with Captain Fritz and Detective Sims. Mr. Ball. Did you go down to the sixth floor? Mr. Boyd. We stopped at the sixth floor—you say, did we go down to the sixth floor? Mr. Ball. When you heard that they found some hulls, just tell us what you did. Mr. Boyd. We went down to the sixth floor and found the hulls over on the southeast corner of the building and they had some books, I suppose it was books—boxes of books stacked up back over there that way. Mr. Ball. Did you see the hulls on the floor? Mr. Boyd. Yes. Mr. Ball. Did you see anything else around there where the hulls were on the floor? Mr. Boyd. Well, over to the west there was some paper sacks, and I think some chicken bones up on top of some boxes. Mr. Ball. That was west?
- 44. Mr. Boyd. Right; yes, sir. Mr. Ball. Near the windows? Mr. Boyd. Yes, sir; they were near the windows. Mr. Ball. How far west from where the hulls were located? Mr. Boyd. Oh, I would say roughly between 30 and 40 feet, probably. Mr. Ball. Where, with reference to the rows of windows—there are pairs of windows—how many pairs of windows away from where the hulls were located did you see the paper sack and chicken bones? Mr. Boyd. Let me see—I don't recall just how many rows of windows from there it was. They are in rows of two, now, I'm not sure, I think it was in front of the third or fourth window over from the southeast corner. Mr. Ball. Third or fourth? Mr. Boyd. Yes. Mr. Ball. Pair of windows? Mr. Boyd. Yes, sir; now—pair of windows—let's see. Mr. Ball. The windows are in pairs on that side, on the Elm Street side—now, what sort of sack was it? Mr. Boyd. The best I remember it was just a brown paper sack—it looked like a lunch sack. Mr. Ball. About the size of a lunch sack? Mr. Boyd. Yes. Mr. Ball. Did you see any other paper sack around there? Mr. Boyd. I don't recall any if I did. Mr. Ball. Did you see any brown wrapping paper near the window where the hulls were found, near the windows alongside
- 45. which the hulls were found? Mr. Boyd. I don't believe I did. Mr. Ball. What else did you see? Mr. Boyd. I just saw those stacks of books up there, and after we had been up there a while, I saw a rifle back over toward the southwest corner over there. Mr. Ball. Where was that located? Mr. Boyd. It was down between some boxes. Mr. Ball. Now, did you see any pictures taken of the hulls, photographs taken of the hulls? Mr. Boyd. Well, let's see, Detective Studebaker and Lieutenant Day, I believe, came up there and they were taking pictures over there at the scene of the hulls. Mr. Ball. And what about where the rifle was found, did you see pictures taken there? Mr. Boyd. Yes; I saw pictures taken over there. Mr. Ball. By whom? Mr. Boyd. Lieutenant Day. Mr. Ball. Did you see anything else on the sixth floor there? Mr. Boyd. I saw a lot of officers. Mr. Ball. Did you find anything yourself? Mr. Boyd. Not on the sixth floor—I don't believe so. Mr. Ball. What time did you leave there? Mr. Boyd. Well, I think I've got it down here somewhere—near 2 o'clock—I believe, but let me check to make sure. It would have been between 1:30 and 2 o'clock. Mr. Ball. Where were you when you heard the rifle had been found?
- 46. Mr. Boyd. I was over near the scene of where the shells had been found. Mr. Ball. Did you see Captain Fritz handle the rifle after it had been found? Mr. Boyd. I don't believe so. Mr. Ball. Did you see him eject anything from it? Mr. Boyd. Let me see, now, I believe they did get a shell out of it after Lieutenant Day came over there. Mr. Ball. Did you see it, or are you just telling us what you heard? Mr. Boyd. Well, I don't believe I saw him get it out. Mr. Ball. You heard about it? Mr. Boyd. Yes, sir. Mr. Ball. You left there and went up to the police department, didn't you? Mr. Boyd. Well, when we left there, we started to go to Irving, but someone—when we got downstairs—someone told Captain Fritz that Sheriff Decker wanted to see him over in his office. Mr. Ball. You say you started to go where? Mr. Boyd. Irving, Tex. Mr. Ball. Where did you get the address in Irving, Tex., or the place to go to in Irving, Tex.? Mr. Boyd. Captain Fritz got it from some man there on the sixth floor. He came up and talked to him a minute and then he told Mr. Sims and I that we should check this Lee Harvey Oswald out, and that was the address they gave us—it was in Irving, Tex. Mr. Ball. And what did you do then? Mr. Boyd. We started to go over there and when we got downstairs, like I said, someone told Captain Fritz that Sheriff
- 47. Decker wanted to see him a minute before he left, and we went in there and while we were in there we learned that the man that had shot Officer Tippit, we thought was the man, was on his way up to our office and Captain Fritz wanted to go by there and we carried him there. Mr. Ball. You were in Decker's office when you heard that a man had been arrested for the murder of Tippit? Mr. Boyd. Yes; we heard about Tippit getting shot when we were up on the sixth floor. Mr. Ball. Then, Fritz told you to go to Irving, didn't he? Mr. Boyd. Yes, sir; we started to Irving. Mr. Ball. Where were you when you heard the man had been arrested, the suspect for the murder of Tippit? Mr. Boyd. Well, I think we was still in the Texas Book Depository when we heard about him being arrested over there. Mr. Ball. Did you go to Decker's office with Fritz? Mr. Boyd. Yes sir. Mr. Ball. And then you went with Fritz up to your office? Mr. Boyd. Yes, sir. Mr. Ball. And did Fritz send somebody else out to Irving, or do you remember? Mr. Boyd. I think later on, I believe, he sent someone else out there. Mr. Ball. He told you to stay there at the police department, did he? Mr. Boyd. Yes, sir. Mr. Ball. What did you do when you got there? Mr. Boyd. Well, we went in and there was a good many people there—I don't recall who all was there—I know we talked to
- 48. Lieutenant Baker, and he told us that the man that shot Tippit was in the interrogation room and about 5 minutes or so after we were in the office, we took Lee Harvey Oswald out of there and brought him into Captain Fritz' office and he talked to him in there. Mr. Ball. Tell us about what time of day that was? Mr. Boyd. I believe it was around 2:20 when we took him out in there; yes, sir. Mr. Ball. And who was there in the room with Oswald at that time? Mr. Boyd. With Oswald at that time—? Mr. Ball. You took Oswald into Fritz' office about 2:20? Mr. Boyd. Yes, sir. Mr. Ball. Who was there besides Oswald? Mr. Boyd. Well, Captain Fritz, and let me see, there was some FBI agents. Mr. Ball. Do you remember their names? Mr. Boyd. I know one came in just shortly thereafter and I remember Mr. Bookhout and Mr. Hosty came in right after we got in there. Mr. Ball. And who else was there? Mr. Boyd. Mr. Hall and Mr. Sims; M. G. Hall is our other partner. Mr. Ball. He's your other partner? Mr. Boyd. Yes, sir. Mr. Ball. And Sims was there, and was there a Secret Service man in there? Mr. Boyd. Let me see—I think there was a Secret Service man there, but I don't recall—I don't know what his name was. Mr. Ball. Do you remember what was said?
- 49. Mr. Boyd. Well, I don't remember exactly what was said. Mr. Ball. Well, in general, what was the substance of what was said? Mr. Boyd. Well—— Mr. Ball. Give me the substance. Mr. Boyd. Well, I knew Captain Fritz asked him his name. Mr. Ball. What did he say? Mr. Boyd. I think he told us his name. I think when he asked him —I'm sure he told him his name because he would talk for a while and then he would quit. Mr. Ball. Did he ask him where he lived? Mr. Boyd. Yes, sir; I think he asked him where he lived. Mr. Ball. What did he say? Mr. Boyd. He said he lived over on Beckley. Mr. Ball. Did he give the address? Mr. Boyd. I believe that he said, well, I know he gave an address —I know he gave an address but he didn't say if it was north or south—I remember that—he didn't say if it was North Beckley or South Beckley and I remember another thing—Mr. Hosty came in and identified him himself, you know, as he came in. Mr. Ball. What do you mean identified him? Mr. Boyd. He took his identification out of his pocket and put it down there in front of him and told him who he was with. Mr. Ball. He told Oswald his name and who he was with? Mr. Boyd. Yes, sir. Mr. Ball. What else happened? Mr. Boyd. Well, they participated in the interrogation—Mr. Hosty asked him some questions and he was pretty upset with Mr. Hosty.
- 50. Mr. Ball. What do you mean by that, what gave you that impression—what happened? Mr. Boyd. Well, just by Oswald's actions, he said he had been to his house two or three times talking to his wife and he didn't appreciate him coming out there when he wasn't there. Mr. Ball. Is that what he said to Hosty? Mr. Boyd. Yes, sir. Mr. Ball. Anything else? Mr. Boyd. I don't recall—I know Mr. Hosty asked him several questions and finally he jumped up and hit the desk, Oswald did, and sat down, and like I say, he was pretty upset. Mr. Ball. Was he handcuffed at that time? Mr. Boyd. Yes; I believe he was handcuffed. Mr. Ball. Was he handcuffed with his hands behind him? Mr. Boyd. No, sir. Mr. Ball. Had his hands been handcuffed behind him before he came into the room? Mr. Boyd. I couldn't say if they had or not—they could have been. Mr. Ball. Do you know whether the handcuffs were changed after he got in the room? Mr. Boyd. They could have been changed after he got in the room—I'm not certain. Mr. Ball. Who changed them? Mr. Boyd. I don't recall. Mr. Ball. Now, when Oswald jumped up and struck the desk, he struck the desk with what? With his hand? Mr. Boyd. With his hands. Mr. Ball. What did Hosty ask him before that?
- 51. Mr. Boyd. He had asked him about a trip to Mexico City? Mr. Ball. Who did? Mr. Boyd. Mr. Hosty. Mr. Ball. What did Oswald say? Mr. Boyd. He told him he hadn't been to Mexico City. Mr. Ball. What else? Mr. Boyd. I don't recall just exactly—I think that the words that he used when he was talking to Mr. Hosty was that he had been out there and accosted his wife, I believe that's the words that he used and like I said, after he talked to him, he said he didn't appreciate him coming out there to his house. Mr. Ball. What was it that Hosty said before Oswald got up and struck the desk with his hand—what question did he ask? Mr. Boyd. I don't remember what the question was. I know it had something to do with—let me see—I'm not sure if he was still talking to him about his wife or the trip to Mexico City. Mr. Ball. You remember he did ask him if he took a trip to Mexico? Mr. Boyd. Yes, sir. Mr. Ball. Oswald said he had not? Mr. Boyd. He said he had not been to Mexico. Mr. Ball. And what did Hosty say to that? Mr. Boyd. He asked him if he denied being to Mexico City—I've just forgotten—it wasn't too awful long before that—I don't recall just exactly what time that he said—I know it was something recent. Mr. Ball. What did Oswald say? Mr. Boyd. He said he had not been there. Mr. Ball. Do you remember anything else that was said?
- 52. Mr. Boyd. No, sir; right offhand—I don't. Mr. Stern. Did he ask him anything about Russia? Mr. Boyd. Yes, sir; something was asked him—I don't recall who asked him about that, and he told us about going over to Russia, I believe he was there in 1959, or something like that—about 1959. I'll tell you, I didn't keep notes in there because of the fact I was sitting right beside Oswald—right in front of him—more or less. Mr. Ball. Did anybody keep notes? Mr. Boyd. I saw the FBI man writing—they had a little book— across the table over there. Mr. Ball. Did you have any microphones in there to record the conversation? Mr. Boyd. No, sir. Mr. Ball. Do you as a practice record the interrogations of your prisoners? Mr. Boyd. No, sir; we don't. Mr. Ball. How long did this take—how long was he questioned at this time? Mr. Boyd. Let me see—we took him down to the first showup right after 4 o'clock, I think I have the exact time here—4:05 is when we left. Mr. Ball. Was he in Captain Fritz' office from the time you took him in there—what time was that? Mr. Boyd. At 2:15–2:20. Mr. Ball. From 2:20 until 4 o'clock? Mr. Boyd. Yes, sir. Mr. Ball. Now, you took him into the first showup, did you? Mr. Boyd. Yes, we left Captain Fritz' office at 4:05.
- 53. Mr. Ball. Who picked the men to go in the showup with him? Mr. Boyd. Who picked the men? Mr. Ball. Yes. Mr. Boyd. I don't recall who picked those men. Mr. Ball. Did you? Mr. Boyd. No, sir; I didn't. Mr. Ball. Did Sims? Mr. Boyd. I don't recall if he did—I don't recall who picked those men. Mr. Ball. Who were the men in this showup? Mr. Boyd. Well, one of them's names was—we call him Bill Perry, his name is William E. Perry, he's a police officer and he was No. 1; and we had Lee Oswald, was No. 2; and R. L. Clark was No. 3; and Don Ables was No. 4. Mr. Ball. The No. 4 man was a clerk there in the jail, was he? Mr. Boyd. I believe he was a clerk down in the jail office. Mr. Ball. Is it usual to have police officers show up with prisoners? Mr. Boyd. Well, I have seen them in there before—I mean—it isn't done real often. Mr. Ball. It's unusual to use officers to showup with prisoners? Mr. Boyd. Well, I would say so, but I know that there has been officers. Mr. Ball. Is that usual to use Don Ables, the clerk, in a showup? Mr. Boyd. No, sir. Mr. Ball. It is unusual? Mr. Boyd. Yes.
- 54. Welcome to our website – the ideal destination for book lovers and knowledge seekers. With a mission to inspire endlessly, we offer a vast collection of books, ranging from classic literary works to specialized publications, self-development books, and children's literature. Each book is a new journey of discovery, expanding knowledge and enriching the soul of the reade Our website is not just a platform for buying books, but a bridge connecting readers to the timeless values of culture and wisdom. With an elegant, user-friendly interface and an intelligent search system, we are committed to providing a quick and convenient shopping experience. Additionally, our special promotions and home delivery services ensure that you save time and fully enjoy the joy of reading. Let us accompany you on the journey of exploring knowledge and personal growth! textbookfull.com