Simulative analysis of channel and qo s aware scheduler to enhance the capacity of multimedia lte networks

International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
6480(Print), ISSN 0976 – 6499(Online), Volume 6, Issue 5, May (2015), pp. 01-08 © IAEME
1
SIMULATIVE ANALYSIS OF CHANNEL AND QoS
AWARE SCHEDULER TO ENHANCE THE CAPACITY OF
MULTIMEDIA LTE NETWORKS
Savitha Suresh1
, Lethakumary.B2
1,2
Department of Electronics and Communication
University College of Engineering, Muttom, Kerala, India
ABSTRACT
Here a new MAC scheduling mechanism for the downlink of LTE systems named Channel
and Qos Aware Scheduler is analyzed. This scheduler is based on a Channel and QoS aware
algorithm which performs joint time and frequency scheduling. The relevance of the scheduler
comes in to play in a situation in which the number of data hungry users are at the rising edge and
they demand for traffics that have very tight QoS requirement in terms of bit rate and delay.eg:-
VoIP, Video conferencing & Online Gaming. The performance of the scheduler is evaluated by
means of network simulations in LTE single cell scenario with mixed traffic and compared the
results with state of the art LTE downlink schedulers. The results shows that in a realistic scenario in
which quality of channel varies over time as well as frequency, CQA scheduler significantly
outperforms other schedulers in terms of provided QoS.
Keywords : Channel and Qos Aware Scheduler, LTE, Multimedia Traffic, Scheduling, Single Cell
Networks
I. INTRODUCTION
For many years, voice calls dominated the traffic in mobile communication network. The
growth of mobile data was initially slow, but in the past couple of years, its use started to increase
dramatically. A part of this growth was driven by the increased availability of 3.5G communication
technology. More important, however was the introduction of Apple iPhone in 2007, followed by
devices based on Google’s Android operating system from 2008. These smart phones were more
attractive and user friendly than their predecessors and were designed to support the creation of
application by third party developers. The result was an explosion in the number and use of mobile
applications. As a contributory factor, network operators had tried to encourage the growth of mobile
INTERNATIONAL JOURNAL OF ADVANCED RESEARCH IN ENGINEERING
AND TECHNOLOGY (IJARET)
ISSN 0976 - 6480 (Print)
ISSN 0976 - 6499 (Online)
Volume 6, Issue 5, May (2015), pp. 01-08
© IAEME: www.iaeme.com/ IJARET.asp
Journal Impact Factor (2015): 8.5041 (Calculated by GISI)
www.jifactor.com
IJARET
© I A E M E

2
data by introduction of flat rate charging schemes that permitted unlimited data downloads. That led
to a situation where neither developers nor users were motivated to limit their data consumption.[1].
The addition of multimedia services such as Skype, GTalk, WhatsApp, multi-user interactive gaming
etc in cellular communication systems has created new challenges for resource allocation, making it
as one of the key component that affects the performance of an LTE system. Due to the higher traffic
flows, the requirement of efficient resource allocation is more important in downlink than uplink. In
short, we need an efficient and smart downlink packet scheduler in the eNB because it simply
maintains the balance between the operator- customer equations. The operator wants to maximize
revenue by utilizing its limited resources to the fullest to accommodate as many users as possible
whereas the users need these applications to the top quality with minimum amount of interruption and
delay.
The LTE systems can have simultaneous delivery of large amount of consumer multimedia
content to vast number of wireless devices which increased its popularity worldwide. As a result there
has been a growing interest in the design of LTE Packet Scheduling algorithms & in the recent
scientific literature, several downlink packet scheduling algorithms have been proposed that focuses
on different aspects of QoS. A very recent and vast survey on the topic is provided in [2].However,
most of the scheduling algorithms mentioned in this survey are not QoS aware. So they are not
suitable for LTE systems. [2]
QoS-aware LTE downlink scheduling algorithms that are aiming at satisfying the delay
requirement of real-time traffic, is proposed in [4]. Here the data flows are prioritized and will be
scheduled based on the Head-of-line (HOL) delay parameter. A disadvantage of this approach is that
it does not take into consideration the variable channel conditions; in particular, in realistic scenarios
in which the presence of fast and frequency-selective fading is expected, assigning radio resources
based only on the HOL metric often results in the selection of lower modulation and coding schemes,
which is spectrally inefficient and thus does not allow to achieve a high capacity. [3]
Among the channel aware approaches, the Token Bank Fair Queue (TBFQ) scheduler can be
considered. This is a queue and channel-aware scheduling algorithm which attempts to maintain
fairness among users. TBFQ is based on the leaky bucket principle, and it is mainly designed to
support bursty traffic. It assigns higher amount of resources to the users that have more data in the
queues. This feature of the TBFQ approach is not adequate for voice traffic, since it is characterized
by small packet sizes and low expected queue fill levels. Furthermore, TBFQ does not explicitly take
into account the delay requirements. [5][3]
Another important downlink scheduler is the Priority Set Scheduler (PSS), which is a channel-
aware scheduler that aims at guaranteeing a predefined bit rate to each user. This algorithm has a very
good performance because it successfully combines TD and FD scheduling in order to achieve a
higher spectral efficiency and increase the overall system capacity. The main drawback of this
scheduler is that it only considers the Guaranteed Bit Rate (GBR) parameter specified within the EPS
bearer. This means that delay sensitive classes of traffic, such as voice, video and gaming, may suffer
poor quality even if their GBR requirement is satisfied. This limits the application of this scheduler to
delay insensitive traffic. [6][3]
A step forward in this research line, [3] proposes a new LTE downlink scheduling algorithm
called Channel and QoS Aware (CQA) scheduler. The QoS parameters that it considers are the HOL
and the GBR parameters. The CQA scheduler performs the scheduling according to different criteria
in the TD and FD, in order to achieve a high spectral efficiency while at the same time taking care of
satisfying the delay requirements of the traffic. The disadvantage is that, only VoIP services are
considered and the simulation scenario is limited to static and pedestrian which is far away from a
realistic scenario. [3]

3
The CQA scheduler proposed in [3] is analyzed based on simulations in order to prove that the
scheduler provides good performance in a realistic LTE scenario with multimedia traffic & results are
compared with state of the art LTE downlink schedulers.
II. CHANNEL AND QOS AWARE (CQA) SCHEDULER
The CQA scheduler, as the name implies, is based on Channel and QoS aware algorithm,
which performs joint TD and FD scheduling. This approach is more efficient than only TD or FD
scheduling [7]. In the TD, at each TTI, the CQA scheduler selects from all the users j = 1, ..., N those
that did not yet reached the maximum bit rate (MBR) and groups them by HOL delay calculating the
metric mtd in the following way[3]:
= ┌ ┐ (1)
Where dj
HOL(t) is the current value of HOL delay of flow j, and g is a grouping parameter that
determines granularity of the groups, i.e. the number of the flows that will be considered in the FD
scheduling iteration. The grouping is used to select the most urgent flows, i.e., with the highest value
of HOL delay, and to enforce the scheduling mechanism to consider those flows in the following FD
scheduling iteration. [3]
The groups of flows selected by TD iteration are then forwarded to FD scheduling starting
from the flows with the highest value of the mtd metric until all RBGs are assigned in the
corresponding TTI. In the FD, for each RBG k = 1, ..., K, the CQA scheduler assigns the current RBG
to the user j that has the maximum value of the FD metric which we define in the following way[3]:
,
= . .
,
(2)
where mj
GBR (t) is calculated as follows[3]:
= = . .!
(3)
where"#$ is the bit rate specified in EPS bearer of the flow j, $% is the past averaged
throughput that is calculated with a moving average, rj
(t) is the throughput achieved at the time t, and
α is a coefficient such that 0 ≤ α ≤ 1.
The purpose of mca
(k,j)
(t) is to add channel awareness to the system in order to maximize the
capacity by assigning the resources to the flows that can use them more efficiently. For mca
(k,j)
(t) two
different metrics are considered: mpf
(k,j)
(t) and mff
(k,j)
(t). The mpf is the Proportional Fair metrics
which is defined as follows [3]:
&
,
= '
(,
(4)
where Re
(k,j)
(t) is the estimated achievable throughput of user j over RBG k calculated by the
Adaptive Modulation and Coding (AMC) scheme that maps the channel quality indicator (CQI) value
to the transport block size in bits. The mpf metric is a good channel awareness metric since it aims at
simultaneously achieving the fairness among flows and maximizing system capacity by prioritizing
the users that have suffered lower channel quality and the users that have extremely good
instantaneous channel quality; the CQA scheduler that uses this channel awareness metric is denoted

4
as CQAPF. The other channel awareness metric is mff [6] and it represents the frequency selective
fading gains over RBG k for user j and is calculated in the following way [3]:
,
=
)*+ (,
∑ )*+ (,-
(./
(5)
where CQI(k,j)
(t) is the last reported CQI value from user j for the kth
RBG. This is considered
as good channel awareness metric since it aims at increasing the overall system capacity by
prioritizing users that can use available resources more efficiently. CQA scheduler that uses this
channel awareness metric is denoted as CQAFF. [3]
III. PERFORMANCE EVALUATION
1. DESCRIPTION OF THE SCENARIOS
To evaluate the response of the CQA scheduler on multimedia traffic, we have simulated a
typical outdoor scenario in which N UE’s are attached to a single eNB. Here, we have considered a
single cell scenario and hence intercell interference is not considered. As per 3GPP specification [8],
the fading scenario considered is Extended Vehicular A (EVA) model, in which the user moves with a
velocity 60 km/h. The modeled traffic are: VoIP, video streaming, Downloads and Real time gaming.
Schedulers used other than CQA are Round Robin (RR), Blind Equal Throughput (BET),
Proportional Fair (PF) and Priority Set (PS) scheduler.
2. SIMULATION SET UP
We have used the LTE-EPC network simulator (LENA) [9] to carry out the performance
evaluation. The four traffics (VoIP, Video, Gaming and Downloads) were introduced together in the
scenario in order to analyze the response of scheduling algorithms under consideration. The strain in
system was evaluated by increasing the number of UE’s. It has been taken care to keep the ratio of
different traffic users constant. The generic settings used in the simulation are listed in Table I.
TABLE I. SIMULATION GENERIC SETTINGS
Parameter Value
Simulator NS3
Bandwidth 5MHz
AMC PiroEW2010
Path Loss Model Friis propagation Loss model
Fading Model Trace Fading Loss Model
Number of RBs 25
Transmission Power of eNB 30 dBm
Transmission Power of UE 23 dBm
Mobility model Constant Position
Noise figure eNB:5; UE:5
TTI 1 ms
Simulation Time (Per User) 10 seconds
The random generator used in placing the user equipments (UE) is based on a prior fashion
for each iteration. Hence the result obtained with one parameter can be reliably compared with
another. The simulation parameters are given in Table II. Output is taken with the support of flow
monitor [10].

International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976
6480(Print), ISSN 0976 – 6499(Online), Volume 6, Issue
3. RESULTS
The performance of the CQA scheduler& t
terms of QoS satisfaction. To measure the QoS satisfaction, we
average throughput when number of UE increases. Throughput is calculated as:
012345
=
07 8 9 :9; <9=>8
TABLE
Parameters
Cell Architecture
Number of eNB
Number of UE
Fading scenario
Cell size
Figure 1 shows the throughput analysis of single cell voice users in a mixed traffic scenario.
Up to 60 users, CQA scheduler outperforms all other schedulers that are taken for comparison. The
degraded performance beyond 60
of service like video, gaming and download are getting more resources than voice traffic
Figure 1: Throughput analysis of Single cell voice users in a mixed traff
6499(Online), Volume 6, Issue 5, May (2015), pp. 01-08 © IAEME
5
The performance of the CQA scheduler& the state of the art schedulers we
satisfaction. To measure the QoS satisfaction, we have evaluated the variation in
when number of UE increases. Throughput is calculated as:
23451?4 0 =
$8=87@8 #A 8; ∗ 8
D7 4E9 73F <8273
G?;
$8=87@8 #A 8; ∗ 8
<9=>8 $8=87@8 H 07 8 9 I72; <9=>8 D8F
TABLE II. SIMULATION PARAMETERS
Parameters Value
Cell Architecture Single Cell
Number of eNB 1
Number of UE
VoIP: 40% 10-100
varied
in steps
of 10
Video:30%
Gaming:10%
Downloads:20%
Fading scenario Vehicular @ 60km/h
Cell size 15 km
degraded performance beyond 60 users justifies the fact that classes of traffic that need more quality
gaming and download are getting more resources than voice traffic
Throughput analysis of Single cell voice users in a mixed traffic with Vehicular
trace
© IAEME
he state of the art schedulers were evaluated in
evaluated the variation in
when number of UE increases. Throughput is calculated as:
D8F
G?;
users justifies the fact that classes of traffic that need more quality
gaming and download are getting more resources than voice traffic.
ic with Vehicular scenario

Figure 2: Throughput analysis of Single cell video users in a mixed traffic with Vehicular scenario
Throughput analysis of video users is given in Figure 2 and F
analysis of gaming users. In both the figures it is obvious that CQA scheduler yields the best
performance. It should also be noted that beyond 60 users, the CQA scheduler yields far better
performance than other schedulers. In other words, the gaming and the video users with CQA
schedulers are getting more resources to meet their requirements in this mixed traffic scenario and
hence proving the fact that classes of traffic with tight QoS requirement will be given more priority.
Figure 3: Throughput analysis of Single cell gaming
6
Throughput analysis of Single cell video users in a mixed traffic with Vehicular scenario
trace
of video users is given in Figure 2 and Figure 3 shows the throughput
In both the figures it is obvious that CQA scheduler yields the best
. It should also be noted that beyond 60 users, the CQA scheduler yields far better
chedulers are getting more resources to meet their requirements in this mixed traffic scenario and
Throughput analysis of Single cell gaming users in a mixed traffic with Vehicular scenario
trace
© IAEME
Throughput analysis of Single cell video users in a mixed traffic with Vehicular scenario
igure 3 shows the throughput
In both the figures it is obvious that CQA scheduler yields the best
. It should also be noted that beyond 60 users, the CQA scheduler yields far better
chedulers are getting more resources to meet their requirements in this mixed traffic scenario and
ic with Vehicular scenario

Figure 4: Throughput analysis of Single cell download users in a mixed traffic with Vehicular
Figure 4 shows the throughput analysis of download users. Here CQA scheduler not only
outperforms other schedulers but also yields an excellent performance on comparison. For instance,
for 50 users in the cell, the download users with other schedulers were getting maximum throughput
of 300 bps whereas CQApf yields throughput more than
beyond 60 users, the download users with CQA scheduler are getting more resources in comparison
with other schedulers. This proves the fact that CQA scheduler is compactable in a situation in which
there is a need to provide multimedia traffics to a large number of
and delay and without compromising the quality of service.
In short the CQA scheduler
download traffic are considered. In the case of voice users, it gives comparatively
performance when numbers of users we
of other traffics.
IV. CONCLUSION
Channel and QoS aware (CQA) scheduling algorithm is a novel scheduling algorithm that is
both channel and QoS aware, which aims to enhance the LTE capacity.
evaluation of CQA scheduler is carried out by means of network simulations in a realistic LTE
scenario and the results are compared with other LTE downlink schedulers. Through analysis of the
results thus obtained we can conclude that, in realistic scenar
over time and frequency, the CQA
terms of provided QoS and system capacity.
balancing of operator -customer equations are of at most importance there is no doubt that the CQA
scheduler will come in handy. There awaits a world with data hungry users and this algorithm is
proven to serve for them.
7
Throughput analysis of Single cell download users in a mixed traffic with Vehicular
scenario trace
Figure 4 shows the throughput analysis of download users. Here CQA scheduler not only
yields throughput more than 400 bps. It should also to be mentioned that
ed to provide multimedia traffics to a large number of users with minimal interruption
and delay and without compromising the quality of service.
the CQA scheduler outperforms all other schedulers when video, gaming and
download traffic are considered. In the case of voice users, it gives comparatively
formance when numbers of users were more than 60 so as to cater for the tight QoS requirements
both channel and QoS aware, which aims to enhance the LTE capacity. Here
results thus obtained we can conclude that, in realistic scenarios in which the channel quality varies
CQA scheduler significantly outperforms the state of the art solutions in
terms of provided QoS and system capacity. Hence, in a commercial LTE network, where the
ustomer equations are of at most importance there is no doubt that the CQA
© IAEME
Throughput analysis of Single cell download users in a mixed traffic with Vehicular
Figure 4 shows the throughput analysis of download users. Here CQA scheduler not only just
It should also to be mentioned that
users with minimal interruption
outperforms all other schedulers when video, gaming and
download traffic are considered. In the case of voice users, it gives comparatively degraded
so as to cater for the tight QoS requirements
Here, the performance
ios in which the channel quality varies
scheduler significantly outperforms the state of the art solutions in
Hence, in a commercial LTE network, where the
ustomer equations are of at most importance there is no doubt that the CQA

8
V. REFERENCES
1. C. Cox, "System Architecture Evolution," An Introduction to LTE: LTE, LTE-Advanced,
SAE and 4G Mobile Communications, Wiley Publishers, 2012.
2. F. Capozzi, G. Piro, G. L.A, G. Boggia, and P. Camarda, “Downlink Packet Scheduling in
LTE Cellular Networks: Key Design Issues and a Survey,” IEEE Commun. Surveys Tuts., vol.
15, no. 2, pp. 678–700, 2013
3. Biljana Bojovic and Nicola Baldo, “A new Channel and QoS Aware Scheduler to enhance the
capacity of Voice over LTE systems”, Proceedings of 11th
International Multi-Conference on
Systems, Signals & Devices (SSD’14),11-14 February 2014, Spain
4. I. Ahmed, L. Badia, N. Baldo, and M. Marco, “Design of a Unified Multimedia-Aware
Framework for Resource Allocation in LTE Femtocells,” Proceedings of International
Symposium on Mobility Management and Wireless Access (MobiWac), 2011
5. F. Bokhari, W. Wong, and H. Yanikomeroglu, “Adaptive Token Bank Fair Queuing
Scheduling in the Downlink of 4G Wireless Multicarrier Networks,” Proceedings of IEEE
Vehicular Technology Conference(VTC), 2008
6. G. Mongha, K. Pedersen, I. Kov´acs, and P. Mogensen, “QoS Oriented Time and Frequency
Domain Packet Schedulers for The UTRAN Long Term Evolution,” Proceedings of IEEE
Vehicular TechnologyConference (VTC), 2008.
7. K. C. Beh, S. Armour, and A. Doufexi, “Joint time-frequency domain proportional fair
scheduler with harq for 3gpp lte systems,” Proceedings of IEEE Vehicular Technology
Conference (VTC), 2008.
8. LTE: Evolved Universal Terrestrial Radio Access (E-UTRA): Base Station Radio
Transmission and Reception, TS 36.104, 3GPP, Release 8, 2011.
9. “LTE-EPC Network Simulator (LENA),” [Online]. Available: http://networks.cttc.es/mobile
networks/ software-tools/lena/
10. "Network Simulator-3 Model Library," [Online]. Available: http://www.nsnam.org/docs
/models/html/index.html
11. K R Remesh Babu, Alia Teresa T M, A Neela Madheswari and Philip Samuel, “Improved Fss
Algorithm with Qos In Cloud” International Journal of Advanced Research in Engineering &
Technology (IJARET), Volume 5, Issue 2, 2014, pp. 138 - 151, ISSN Print: 0976-6480, ISSN
Online: 0976-6499.
12. Prof. J. R. Pathan, Prof. A. R. Teke, Prof. M. A. Parjane and Prof. P.S. Togrikar, “Dropping
Based Contention Resolution For Service Differentiation To Provide Qos In WDM OBS
Networks” International journal of Computer Engineering & Technology (IJCET), Volume 4,
Issue 1, 2013, pp. 218 - 228, ISSN Print: 0976 – 6367, ISSN Online: 0976 – 6375.
13. N. Devakirubai and s. Palani, “QoS Ensured Optimal Replica Placement In Graph Based Data
Grids” International journal of Computer Engineering & Technology (IJCET), Volume 4,
Issue 6, 2013, pp. 314 - 325, ISSN Print: 0976 – 6367, ISSN Online: 0976 – 6375.

Simulative analysis of channel and qo s aware scheduler to enhance the capacity of multimedia lte networks

More Related Content

What's hot (19)

Viewers also liked (14)

Similar to Simulative analysis of channel and qo s aware scheduler to enhance the capacity of multimedia lte networks (20)

More from IAEME Publication (20)

Recently uploaded (20)

Simulative analysis of channel and qo s aware scheduler to enhance the capacity of multimedia lte networks