QoS-aware scheduling in LTE-A networks with SDN control

QoS-aware scheduling in LTE-A networks with
SDN control
Emmanouil Skondras1, Angelos Michalas2, Aggeliki Sgora1, Dimitrios D. Vergados1
1
Department of Informatics, University of Piraeus, Piraeus, Greece, Email: {skondras, asgora, vergados}@unipi.gr
2
Department of Informatics Engineering, Technological Education Institute of Western Macedonia,
Kastoria, Greece, Email: amichalas@kastoria.teiwm.gr
Abstract—The 3GPP Long Term Evolution Advanced (LTE-A)
standard specifies a set of pioneer features such as relay nodes
and carrier aggregation. At the same time, the Software Defined
Networks (SDN) have become an emerging technology which
provides centralized control and programmability to modern
networks. In the current communication environment, cloud
computing could combine the advantages of both technologies
in order to create a novel cloud assisted Software Defined LTE-
A architecture with relay nodes. Moreover, due to the increased
requirements of modern services, the optimal resource allocation
is a necessity. In such a context, this paper describes a QoS
aware cross carrier scheduler for downlink flows, aiming at
the optimization of system resources allocation. The proposed
scheduler is evaluated against the PF, MLWDF, EXP/PF, EXP
RULE, LOG RULE, FLS and FLSA schedulers in a cloud assisted
Software Defined LTE-A topology with relay nodes. Simulation
results show that the proposed scheduler improves the real
time services performance while at the same time maintains an
acceptable performance for best effort flows.
I. INTRODUCTION
The LTE network technology uses an all-IP, purely packet-
switched architecture to provide high communication speeds
and satisfy the increased QoS requirements of modern ser-
vices. The first LTE release (rel. 8) [1] provides data rates up
to 100 Mbps for downlink and up to 50Mbps for uplink. In
such an architecture, information is transmitted in frames of
10ms of length and every frame is split into 10 sub-frames
of 1 Time Transmission Interval (TTI) of length each. The
minimum resource which can be allocated for transmission is
called Resource Block (RB). The scheduler assigns RBs to
users in each TTI and the number of available RBs per TTI
depends on the system bandwidth.
The LTE-A (rels. from 11 to 14) [2] significantly enhances
the rel. 8 specifications and improves the access network
capacity using evolved characteristics, such as Relay Nodes
(RNs) and Carrier Aggregation (CA) providing increased cov-
erage with rates up to 1Gbps for downlink and up to 500Mbps
for uplink. Relaying is one of the vital elements of LTE-A
to fulfill the users’ requirements on high data rate coverage.
Wireless small-range cells improve connectivity of mobile
users experiencing bad channel conditions within the coverage
area of an LTE base station (eNodeB). Relays are connected
to the backbone network by maintaining wireless connections
with the base station. Furthermore, using CA mode in LTE-
A, the system bandwidth is increased by aggregating up to
5 component carriers and thus having more available RBs for
scheduling in each TTI. The first component carrier is referred
as Primary Component Carrier (PCC) and the rest component
carriers are referred as Secondary Component Carriers (SCCs).
According to the Cross Carrier Scheduling (CCS) method-
ology, the PCC includes both a control region and a data
region, while each SCC includes only a data region. Regarding
the non-Cross Carrier Scheduling (n-CCS) operating principle
each PCC and SCC use their own control regions to allocate
RBs. Both principles provide flexible component carrier band-
widths from 1.4 up to 20MHz while 100 RBs are available
for allocation per component carrier in the case of 20MHz
bandwidth. However, the CCS methodology manipulates the
individual component carriers as a single entity and acquires a
total supervision of the channel conditions optimizing system
capacity.
On the other hand, Software Defined Networks (SDN) is
an emerging technology that centralizes the network intelli-
gence enhancing the flexibility of resource manipulation. More
specifically, the network control functions are decoupled from
the data forwarding procedures while the network architecture
becomes easily programmable. Thus, resource optimization is
achieved by dynamically adjusting the traffic forwarding to
meet the changing needs that exist in the modern wireless
communication environment.
In this context, cloud computing could combine both LTE
and SDN technologies advantages to create a novel cloud
assisted Software Defined LTE-A network architecture with
RNs. In this architecture, network control operations are
centrally organized having a wide view of the entire system.
This paper presents the FLS Advanced - Cross Carrier
(FLSA-CC) downlink scheduler which is an enhanced version
of the FLS Advanced (FLSA) presented in [3]. FLSA-CC
operates in a relay assisted LTE-A network which supports
CA. The proposed scheduler aims at a QoS aware resource
allocation, satisfying the requirements of strict real times
services while maintaining an acceptable level of performance
for the best effort service. The performance of FLSA-CC is
studied on a software defined cloud architecture where flow
forwarding and resource scheduling decisions are performed
centrally by a global controller implemented on the cloud. The

remainder of the paper is organized as follows. In section 2, the
related research literature is revisited while section 3 describes
the proposed scheduler. Furthermore, section 4 presents the
performance evaluation of the FLSA-CC in comparison with
existing scheduling models and section 5 concludes our work.
II. RELATED WORK
Several downlink packet schedulers have been proposed in
the current literature. This section makes a brief overview of
available scheduling strategies applying to LTE-A systems.
In [4] the authors propose an Iterative Parallel Grouping
Algorithm (IPGA)-based flow scheduler which aims at achiev-
ing service differentiantion while at the same time reducing the
system energy consuption. The algorithm is applied to an SDN
architecture and the scheduling process is performed within
a centralized SDN controller. As follows, a unique priority
value assigned to each routing path is determined in respect of
QoS requirements and OpenFlow switches energy consuption.
Simulation results showed that both priority differentiantion
and energy saving are performed.
The authors in [5] study the performance of the Propotional
Fairness (PF) and Round Robin (RR) non-QoS aware sched-
ulers in relay enhanced LTE-A networks. The PF scheduler
aims at a fair distribution of the network resources to users. Its
metric is estimated using formula (1), where di
k(t) represents
the available throughput for the ith
flow in the kth
RB of the
tth
TTI and ¯Ri
(t − 1) denotes the past average throughput.
Both inband and outband LTE relaying schemes are consid-
ered. More specificaly, in an inband relaying scheme the relay
access link operates in the same frequency with the backhaul
link that connects the RN with the donor eNB (DeNB). On
the contrary, in an outband relaying scheme, the relay access
link operates in a different frequency than the backhaul link
connecting the RN with the DeNB. Simulation results showed
that the PF scheduler achieves better performance that the RR
scheduler in both inband and outband scenarios.
mP F
i,k =
di
k(t)
¯Ri(t − 1)
(1)
In [6] an analysis of the interference phenomenon that
occures in LTE-A networks with type-1 inband RNs is
performed. Also, a Two-Hop Proportional Fairness (2H-PF)
downlink scheduler is introduced. Its metric is estimated using
formula (2), where AL and BH designate the Access Link and
Backhaul Link, respectively. Also, the CR parameter defines
the composite rate estimated by formula (3). The proposed
scheduler is evaluated in scenarios with 1, 2 and 4 RNs per
DeNB while the DeNB activity is deactivated by 0%, 20%
and 40% of the TTI time respectively. Simulation results in
terms of throughput, Signal to Interference & Noise Ratio
(SINR) and Block Error Rate (BLER) showed that the network
performance is quite sensitive when a level of DeNB muting
is applied.
m2H−P F
i,k = max
CR(t)
¯Ri(t − 1)AL
,
CR(t)
¯Ri(t − 1)BH
(2)
CR =
1
(di
k(t))AL
+
1
(di
k(t))BH
(3)
The Modified Largest Weighted Delay First (M-LWDF)
and the Exponential/PF (EXP/PF) QoS aware schedulers are
described in [7] and [8], respectively. Both algorithms extend
the PF metric by taking into consideration network factors
such as delays and packet losses that affect the service quality.
Specifically, the M-LWDF metric is estimated using formula
(4), while the EXP/PF metric is calculated by formula (5).
The DHOL,i parameter represents the head of line delay.
Additionally, the αi value is determined by formula (6), where
δi is the target packet loss ratio and τi is the delay constraint.
Finally, the X value is calculated by formula (7), where Krt is
the number of active real time flows. Numerical results showed
that the M-LWDF scheduler performs better as the network
load is low, while the EXP/PF algorithm gives better results
as the load increases.
mM−LW DF
i,k = ai · DHOL,i · mP F
i,k (4)
m
EXP/P F
i,k = exp(
ai · DHOL,i − X
1 +
√
X
) · mP F
i,k (5)
αi = −
logδi
τi
(6)
X =
1
Krt
·
Krt
i=1
ai · DHOL,i (7)
The LOG RULE and the EXP RULE schedulers presented
in [9] and [10] take into account the head of line packet delay
and the channel quality reported by UEs, to support delay
sensitive flows. Especially, the LOG RULE metric is estimated
using formula (8), where Γi
k is the spectral efficiency for the
ith
user on the kth
subchannel. Correspondingly, the EXP
RULE metric is estimated using formula (9), where bi and
c are configurable parameters. Simulation results showed that
both LOG RULE and EXP RULE could guarantee delay
constraints by configuring scheduler parameters according to
users target delays.
mLOGRULE
i,k = bi · log(c + ai · DHOL,i) · Γi
k (8)
mEXP RULE
i,k = bi·exp(
ai · DHOL,i
c + (1/Krt) j DHOL,j
)·Γi
k (9)
In [11] the authors evaluate the FLS QoS aware downlink
scheduler in an LTE-A topology with RNs. This scheduler
is implemented in two levels. The upper level uses formula
(10) to estimate the ui(x) quota of data that the ith
real
time flow must transmit at the xth
frame to succeed its QoS
constraints. In this formula, qi(x) represents the queue length
in the xth
frame, Mi the number of coefficients used and
ci(n) the nth
coefficient value. Coefficients are used in order
to guarantee the required delay constraints for real time flows.
The target delay τi is determined by formula (11), where Tf
is the frame length. Additionally, the coefficient value ci(n) is
determined by formula (12). Accordingly, the lower level uses

the PF metric to allocate network resources to real time flows
for transmitting their quota of data, whereas the remaining
resources are allocated to best effort flows. Simulation results
showed that the FLS scheduler succeeds better performance
for real time flows compared to the PF and LOG RULE
schedulers.
ui(x) = qi(x) +
Mi
n=2
[qi(x − n + 1)
− qi(x − n + 2) − ui(x − n + 1)] · ci(n)
(10)
τi = (Mi + 1) · Tf (11)
ci(n) =
n where 0 ≤ n ≤ 1
ci(n − 1)/2 where n ≥ 2
(12)
III. THE PROPOSED SCHEDULER
This section describes the proposed FLSA-CC scheduler
(Fig. 1) which aims at the optimization of system perfor-
mance using carrier aggregation in the LTE downlink. More
specifically, the FLSA-CC improves the FLSA [3] scheduler
in a cross carrier manner adapting the resource allocation
process at different channel conditions of the aggregated
component carriers. Furthermore, due to the fact that in the
CCS methodology adopted by this scheduler, only the PCC
uses the PDCCH channel for transmission of scheduling
information, interference decrement is observed resulting in
better channel conditions in terms of SINR, increasing thus
the overall system capacity. The FLSA-CC has been built
upon three distinct levels, which cooperate with each other for
dynamically assigning radio resources to users in each TTI.
The upper level of FLSA-CC uses the formula (10) of the
FLS [12] to estimate the ui(x) quota of data that the ith
real
time flow should transmit in each xth
TTI, in order to succeed
its QoS constraints. In other words, ui(x) quota is estimated in
each xth
TTI of a frame, whereas in FLS it is estimated once,
at the beginning of each frame. The FLSA-CC estimates the
coefficient value ci(n) using formula (13), where N represents
the number of component carriers. Performance improvement
has been observed due to the fact that in FLS, when a real time
flow transmits its ui(x) quota of data, it loses the opportunity
to continue the transmission until the beginning of the next
frame. By recalculating the formula (10) in each TTI (instead
of estimating it only at the beginning of each frame), the
FLSA-CC provides more resources to real time flows that have
remaining data for transmission.
ci(n) =
n where 0 ≤ n ≤ 1
ci(n − 1)/(2 · N) where n ≥ 2
(13)
In each TTI, the middle level uses a cross carrier version
of the MLWDF scheduler, called MLWDF-CC, to allocate
RBs to real time flows for transmitting their ui(x) quota of
data obtained from the upper level. This scheduler extends
the PF-CC metric [13] [14] mP F −CC
i,k given in formula (14).
Accordingly, the MLWDF-CC metric is evaluated by formula
Fig. 1. The FLSA-CC scheduler design
(15) where the αi value is determined by formula (6). As
follows, the use of the cross carrier QoS aware MLWDF-CC
scheduler realizes improved resource distribution among the
real time flows in comparison with the FLS scheduler which
at the second level uses the non-cross carrier principle as well
as the non-QoS aware PF algorithm.
The third level has been added to allocate the remaining
RBs of each TTI to best effort flows using formula (14) of
the PF-CC algorithm.
mP F −CC
i,k =
di
k(t)
N
j=1
¯Ri,j(t − 1)
(14)
mMLW DF −CC
i,k = ai · DHOL,i · mP F −CC
i,k (15)
IV. SIMULATION ENVIRONMENT
The simulation environment is implemented using an ex-
tended version of the Lte-Sim [15] simulator. More specif-
ically, the iCanCloud [16] and the OpenFlow [17] modules
of the Omnet++ [18] simulator have been configured and
embedded to the Lte-Sim enabling the ability to include
cloud infrastructure and SDN controllers to the simulated LTE
topologies.
The simulation environment consists of an LTE Evolved
UMTS Terrestrial Radio Access Network (E-UTRAN) and a
Cloud infrastructure. The E-UTRAN includes 7 DeNBs and
28 RNs (4 per DeNB). The transmission radius is equal to
1 kilometer for each DeNB and 100 meters for each RN
respectively. Also, each RN is positioned at 90% of the
transmission radius of its DeNB. Furthermore, Type-1 Outband
Relaying is applied where two link types are defined, the
access link and the backhaul link. The access link is used
for communication between a UE and an RN or a DeNB
using a frequency f1, while the backhaul link is used for
communication between an RN and a DeNB using a frequency
f2 = f1.
Inter-band CA is applied in each DeNB and RN. According
to this CA configuration, two component carriers, which

belong to different frequency bands, are aggregated. Each
component carrier bandwidth is equal to 20MHz and contains
100 resource blocks. Thus, 40MHz bandwidth is assigned to
each cell (RN or DeNB) and 200 RBs are totally available
for scheduling in each TTI. Table 1 presents the system sub-
frequencies assigned to each cell.
TABLE I
THE SUB-FREQUENCIES THAT ASSIGNED TO EACH CELL
Cell Aggregated Component Carriers
DeNB0
PCC: 1805 MHz − 1825 MHz (band 3)
SCC: 760 MHz − 780 MHz (band 28)
DeNB1,3,5
PCC: 1825 MHz − 1845 MHz (band 3)
DeNB2,4,6
PCC: 2110 MHz − 2130 MHz (band 1)
Relay nodes
Case 1
PCC: 2620 MHz − 2640 MHz (band 7)
Case 2
PCC: 2640 MHz − 2660 MHz (band 7)
The 3GPP urban channel model [19] [20] is considered.
Due to the assumption that the channel between DeNB or RN
and UE encounters non-line-of-sight (NLOS) transmission, its
propagation loss is estimated using formula (16) and (17),
respectively, where d represents the distance among the nodes
and fc the carrier frequency. Correspondingly, due to the
assumption that the channel between a DeNB and an RN
encounters line-of-sight (LOS) transmission, its propagation
loss is estimated using formula (18).
PLNLOS
DeNB→UE = 37.6·log10(d)+58.94+21·log10(fc) (16)
PLNLOS
RN→UE = 36.7 · log10(d) + 22.7 + 26 · log10(fc) (17)
PLLOS
DeNB→RN = 22 · log10(d) + 28 + 20 · log10(fc) (18)
The Cloud contains a set of virtual machines (VMs) and
implements the functionalities of the LTE Evolved Packet Core
(EPC). Additional VMs with user applications are created.
Specifically, one VM is created for each UE running three
applications namely one VoIP, one video and one best effort.
Flow forwarding as well as resource scheduling in each
DeNB and RN are performed using a centralized global
controller placed into the SGW having a wide view of the
entire system. The simulated topology is presented in figure
2.
V. PERFORMANCE EVALUATION
The performance of the FLSA-CC was evaluated against the
PF, MLWDF, EXP/PF, FLS, FLSA, EXP-RULE and LOG-
RULE schedulers. Especially for the EXP-RULE metric the
used parameter set is ai ∈ [5/(0.99 · τi), 10/(0.99 · τi)], bi =
1/E[Γi
] and c = 1 as proposed in [10] for best performance.
Table II summarizes the factors considered by each scheduler
for resource allocation, demonstrating that the FLSA-CC is
the most complete strategy. Furthermore, the full band periodic
Channel Quality Indication (CQI) reporting scheme is applied.
Thus each UE reports its downlink SINR to RN, for each
Fig. 2. The simulated topology
component carrier in every TTI. The RN quantizes the reported
SINR value and calculates the CQI as described in [15]. Then,
it uses the CQI to guarantee a maximum BLER less than 10%
regardless of the scheduling strategy applied.
TABLE II
THE PARAMETERS CONSIDERED IN EACH SCHEDULER
Scheduler SINR Throughput
HOL
Delay
Max.
Delay
Max.
PLR
Queue
Length
CCS
PF
MLWDF
EXP/PF
FLS
FLSA
FLSA-CC
EXP RULE
LOG RULE
A number of users, move inside the borders of each RN
according to the random way-point mobility model. Each user
receives two real time flows, an H264 video with bitrate equal
to 440 kbps and a Voice over IP (VoIP) using the G.729 codec.
Furthermore, one best effort flow is added as background
traffic. Table III summarizes the simulation parameters.
A. Real time services results
Due to the fact that the simulation environment inlcudes an
LTE topology with RNs, a two hop target delay for real time
flows τi = τi,DeNB→RN + τi,RN→UE is considered, where
τi,DeNB→RN represents the target delay between a DeNB and
an RN, the τi,RN→UE represents the target delay between an
RN and a UE. Also we consider τi,DeNB→RN = τi,RN→UE.
In general, QoS aware schedulers increase the packet loss
ratio (PLR) to maintain the required τi. This strategy is based

TABLE III
THE SIMULATION PARAMETERS
Parameter Value
Simulation time 100 seconds
Downlink bandwidth 2*20 = 40 MHz
Modulation QPSK, QAM-16 and QAM-64
DeNBs number / radius 7 / 1 km
Relay nodes number / radius 4 per DeNB / 100 m
Number of users up to 100 users per relay node
Users mobility Random way point
Traffic models
Real time:
H264 video at 440 kbps
VoIP using G.729 codec
Best effort: Web
Fig. 3. Real time flows packet loss ratio using different target delays
on the assumption that real time services such as VoIP and
Video can not make use of expired packets. Thus, since the
delay constraint is satisfied, the algorithms are evaluated in
terms of PLR, so as to have a comprehensive view about the
performance improvements. Figure 3 illustrates the impact of
the target delay parameter τi in the PLR for VoIP and video
flows, respectively, for the case of having 100 users per RN.
While the target delay increases from 50ms to 150ms, the
PLR decreases. Additionally it may be observed that FLSA-
CC compared with FLSA exhibits lower PLR independed of
the target delay parameter.
In figure 4, the PLR for VOIP and video flows is presented
while the number of cell RN users varies from 20 to 100. In
this case, the considered target delays are set to 100ms and
150ms for VoIP and video flows respectively, as determined
by the LTE QoS class specifications for these service types.
As shown, FLSA-CC results in a lower PLR than the rest of
the algorithms. Specifically, FLSA shows a marginal decrease
of its PLR for VoIP flows as well as up to 3% lower PLR for
video flows compared to FLSA.
The analysis of the throughput offered to real time services
provides an important insight on the performance of the FLSA-
CC in comparison with the other schedulers. As presented in
figure 5 the FLSA-CC outperforms the rest of the schedulers,
independently of the number of users for VoIP and video flows.
This is expected due to the cross carrier scheduling operating
principle applied as well as due to the recalculation of formula
(10) in each TTI by the upper level of the FLSA-CC. More
specifically, the FLSA-CC succeeds higher throughputs than
the rest of the algorithms providing rates of up to 800kbps for
Fig. 4. Real time flows packet loss ratio
Fig. 5. Real time flows throughput
VoIP and up to 28Mbps for video services.
The proposed scheduler is also evaluated in terms of Jain
fairness index, which is estimated using formula (19) where
n is the number of the service flows and xi is the throughput
of the ith
flow.
Jain Fairness =
(
n
i=1 xi)2
n ·
n
i=1 x2
i
(19)
Flows with the same service constraints must receive similar
QoS to avoid the situation of having satisfied users against
dissatisfied ones of the same service type. The maximum value
of fairness is 1 while the more a scheduler accomplishes a
value close to 1, the more the resource allocation is fair.
Figure 6 demonstrates that the FLSA-CC scheduler improves
the fairness for both VoIP and video flows.
Fig. 6. Real time flows fairness index

Fig. 7. Best effort flows throughput and fairness index
B. Best effort services results
In this subsection FLSA-CC, FLSA and FLS which accom-
plish better performance for real time services are evaluated
for best effort flows in terms of throughput and fairness index.
As presented in figure 7, the FLSA-CC outperforms the other
two schedulers and provides throughput up to 1.5Mbps for
best effort flows even when the number of users increases,
while the FLSA accomplishes only a 100kbps throughput.
Additionally, the FLSA-CC scheduler significantly improves
the fairness index of best effort flows.
VI. CONCLUSION
In this work FLSA-CC QoS aware cross carrier downlink
scheduler is proposed as an extended version of the FLSA [3].
FLSA-CC operates in an LTE-A network with relay nodes in
a CA mode. The proposed scheduler has been built upon three
distinct levels which cooperate with each other to allocate the
network resources to users in a manner that the requirements
of strict real times services are satisfied while starvation of
best effort traffic is avoided. The performance of FLSA-CC is
compared against other scheduling algorithms in terms of PLR,
throughput and fairness index, in a cloud assisted software
defined architecture. Simulation results showed that the FLSA-
CC scheduler outperforms the rest of the scheduling schemes
by achieving better resource allocation for both real time and
best effort services..
ACKNOWLEDGEMENT
The publication of this paper has been partly supported by
the University of Piraeus Research Center (UPRC) and the
Technological Educational Institute of Western Macedonia.
REFERENCES
[1] “TS 36.322 (V8.8.0): Evolved Universal Terrestrial Radio Access (E-
UTRA), Radio Link Control (RLC) protocol specification(Release 8),”
Technical Specification, 3GPP, 2010.
[2] “TS 36.300 (V13.2.0): Evolved Universal Terrestrial Radio Access
Network (E-UTRAN) (Release 13),” Technical Specification, 3GPP,
2016.
[3] E. Skondras, A. Michalas, A. Sgora, and D. D. Vergados, “A downlink
scheduler supporting real time services in LTE cellular networks,” in
Information, Intelligence, Systems and Applications (IISA), 2015 6th
International Conference on. IEEE, 2015, pp. 1–6.
[4] B.-Y. Ke, P.-L. Tien, and Y.-L. Hsiao, “Parallel prioritized flow schedul-
ing for software defined data center network,” in High Performance
Switching and Routing (HPSR), 2013 IEEE 14th International Confer-
ence on. IEEE, 2013, pp. 217–218.
[5] M. Minelli, M. Ma, M. Coupechoux, and P. Godlewski, “Scheduling
Impact on the Performance of Relay-Enhanced LTE-A Networks,” IEEE
Transactions on Vehicular Technology, vol. 65, no. 4, pp. 2496–2508,
2016.
[6] P. Arnold, V. Rakocevic, and J. Habermann, “Simulation Study on 2
Hop-Proportional Fair Scheduling and eICIC in Relay Type 1 Extended
LTE-A Networks,” in Vehicular Technology Conference (VTC Fall),
2015 IEEE 82nd. IEEE, 2015, pp. 1–5.
[7] M. Andrews, K. Kumaran, K. Ramanan, A. Stolyar, P. Whiting, and
R. Vijayakumar, “Providing quality of service over a shared wireless
link,” Communications Magazine, IEEE, vol. 39, no. 2, pp. 150–154,
2001.
[8] K. S. R. Basukala, H. Mohd Ramli, “Performance analysis of EXP/PF
and M-LWDF in downlink 3GPP LTE system,” in AH-ICI 2009. 1st
Asian Himalayas International Conference on Internet. IEEE, 2009,
pp. 1–5.
[9] B. Sadiq, R. Madan, and A. Sampath, “Downlink scheduling for mul-
ticlass traffic in LTE,” EURASIP Journal on Wireless Communications
and Networking, vol. 2009, no. 14, 2009.
[10] S. Bilal, M. Ritesh, and S. Ashwin, “Downlink scheduling for multiclass
traffic in LTE,” EURASIP Journal on Wireless Communications and
Networking, vol. 2009, 2009.
[11] G. Piro, L. A. Grieco, G. Boggia, and P. Camarda, “QoS provisioning in
LTE-A networks with relay nodes,” in Wireless Days (WD), 2012 IFIP.
IEEE, 2012, pp. 1–3.
[12] G. Piro, L. A. Grieco, G. Boggia, R. Fortuna, and P. Camarda, “Two-
level downlink scheduling for real-time multimedia services in LTE
networks,” Multimedia, IEEE Transactions on, vol. 13, no. 5, pp. 1052–
1065, 2011.
[13] M. H. H. Mohammed Abdul Jawad M. Al-Shibly and M. R. Islam,
“Radio resource scheduling in LTE-Advanced system with Carrier
Aggregation,” ARPN Journal of Engineering and Applied Sciences,
vol. 10, no. 22, pp. 17 281–17 285, 2015.
[14] S. Oh, J. Na, and D. Kwon, “Performance Analysis of Cross Component
Carrier Scheduling in LTE Small Cell Access Point System,” in The
Second International Conference on Electrical, Electronics, Computer
Engineering and their Applications (EECEA2015), 2015, p. 146.
[15] G. Piro, L. A. Grieco, G. Boggia, F. Capozzi, and P. Camarda,
“Simulating LTE cellular systems: An open-source framework,” IEEE
Transactions on Vehicular Technology, vol. 60, no. 2, pp. 498–513, 2011.
[16] A. Núñez, J. L. Vázquez-Poletti, A. C. Caminero, G. G. Castañé,
J. Carretero, and I. M. Llorente, “iCanCloud: A flexible and scalable
cloud infrastructure simulator,” Journal of Grid Computing, vol. 10,
no. 1, pp. 185–209, 2012.
[17] D. Klein and M. Jarschel, “An OpenFlow extension for the OMNeT++
INET framework,” in Proceedings of the 6th International ICST Con-
ference on Simulation Tools and Techniques. ICST (Institute for
Computer Sciences, Social-Informatics and Telecommunications Engi-
neering), 2013, pp. 322–329.
[18] A. Varga and R. Hornig, “An overview of the OMNeT++ simulation
environment,” in Proceedings of the 1st international conference on
Simulation tools and techniques for communications, networks and
systems & workshops. ICST (Institute for Computer Sciences, Social-
Informatics and Telecommunications Engineering), 2008, p. 60.
[19] “TR 36.814 (V9.0.0): Further advancements for E-UTRA physical
layer aspects (Release 9),” Technical Specification Group Radio Access
Network, 3GPP, 2010.
[20] “TR 36.942 (V10.2.0): Radio Frequency (RF) system scenarios (Release
10),” Technical Specification, 3GPP, 2011.

QoS-aware scheduling in LTE-A networks with SDN control

More Related Content

What's hot (18)

Similar to QoS-aware scheduling in LTE-A networks with SDN control (20)

More from University of Piraeus (20)

Recently uploaded (20)

QoS-aware scheduling in LTE-A networks with SDN control