ISSN 2278-3091

Volume 10, No.3, May - June 2021

International
Shivanandaswamy V. Mahapurush etJournal of Advanced
al., International Trends
Journal of Advanced inComputer
Trends in Computer Science
Science and and
Engineering, Engineering
10(3), May - June 2021, 1832 – 1836
Available Online at http://www.warse.org/IJATCSE/static/pdf/file/ijatcse481032021.pdf
https://doi.org/10.30534/ijatcse/2021/481032021

Link Scheduling Algorithm to Mitigate Worst-Case

Interference Problems for LTE-A
Shivanandaswamy V. Mahapurush*1, Prathima Mahapurush2, Priyatam Kumar3
1
Assistant Professor, Department of Electronics and Communication Engineering, Smt Kamala & Sri
Venkappa M. Agadi College of Engineering & Technology, Lakshmeshwar, Gadag, Karnataka, India
2
Assistant Professor, Department of Information Science and Engineering, Smt Kamala & Sri Venkappa M.
Agadi College of Engineering & Technology, Lakshmeshwar, Gadag, Karnataka, India
3
Professor, School of Electronics and Communication Engineering, KLE Technological University,
Vidyanagar, Hubli, Karnataka, India
*Corresponding Author: [email protected]

ABSTRACT uplink transmission speeds of up to 1 Gbps and 500

Mbps, respectively. By merging the unused or surplus
As the number of apps increases, the traffic of multimedia secondary carrier components of one cell with the
services in wireless networks grows exponentially. primary carrier component of a geographically different
Because of the proportional fair algorithm's shortcomings cell (both of which are supplied by distinct base stations)
in terms of achieving Quality of Service (QoS) [2,] the unused or surplus secondary carrier components
requirements such as delay and packet loss, it is not of one cell can be efficiently utilized. This allows for
extensively employed in modern wireless communication. more efficient use of the wireless world's limited radio
This study provides a scheduling method based on the resources (spectrum).
QoS utility function, with a parameter added to improve
the priority mechanism in order to overcome such flaws. 2. LITERATURE SURVEY
Our scheduling algorithm can match the QoS criteria of
real-time traffic, according to simulation results, and the A practical scheduler should be capable of increasing
overall system transmission performance has increased overall system capacity while maintaining some level of
significantly. user fairness [4, 5, 8, 11]. Several packet scheduling
algorithms have recently been proposed in order to
Key words: Quality of service, Scheduling algorithm, deliver faster throughput while ensuring fairness to all
Utility function, Proportional fair users. The scheduling mechanism in [6], for example,
uses Assignment Model to achieve this balance by
1. INTRODUCTION effectively mapping UEs to RBs at each TTI. The
suggested algorithm in [6] modifies the Best-Channel
To meet the demands of fast-processing applications Quality Indicator (Best-CQI) method, which is
operating on a wide range of internet-connected wireless characterized by high data rates at the cell level but poor
devices such as smart phones, laptops, and palmtops, fairness, in order to achieve fairness in resource
among others. In LTE Release 10, the Third distribution while keeping system capacity utilization as
Generation Partnership Project (3GPP) proposed carrier high as possible. The authors of [9] present a multi-user
aggregation technology. ITU has standardized carrier scheduling system that operates on the edge of the
aggregation as a fourth-generation technological feasible multi-user rate region while ensuring a desirable
component for achieving the target data rates of 1 Gbps long-term average fairness. [1] implemented another
in the downlink [1] and 500 Mbps in the uplink [2]. scheduling algorithm based on the utility function in
Wider bandwidths of up to 100 MHz spectrum are order to enhance the performance of the LTE system,
required to achieve such large data rates [2]. The 3GPP specifically the throughput and fairness performance. By
LTE systems with transmission bandwidths of 1.4 MHz, splitting the packet scheduler into a time domain and a
4 MHz, 5 MHz, and 20 MHz served as the foundation frequency domain and using separate algorithms in both
for the planned 100 MHz IMT-Advanced mobile domains, as described in [2], throughput fairness among
systems, also known as LTE Advanced. Multiple LTE users may be successfully controlled. [4] Describes a
carrier components (CCs) with varying bandwidths and Generalized Proportional Fair (GPF) scheduling
situated in the same or distinct frequency bands can be methodology and its application to OFDMA frequency
aggregated in CA to obtain a bandwidth of up to 100 scheduling. When compared to a system without
MHz for supplying user equipment with downlink and

1832
Shivanandaswamy V. Mahapurush et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 1832 – 1836

frequency scheduling, this improves system throughput It's worth noting that the FFR or SFR schemes'
while also improving fairness in terms of allotted prospective benefits aren't always achieved [5, 8]. On
resources and obtained data rates per user. [4] Proposes the one hand, ICIC can reduce interference for cell-
an adaptive proportional fair scheduling method for LTE edge users, increasing the bit rate of cell-edge users.
systems, in which the author attempts to create an In a unity FRF system, on the other hand, the lower
algorithm that promotes UE fairness while limiting SINR caused by ICI can be compensated by
overall system throughput degradation by altering assigning extra bandwidth to these users. Even
scheduling priority according to individual user's channel without ICIC, a satisfactory performance can be
condition. attained by compensating for the reduced SINR with
greater bandwidth, that is, resource blocks, to suitable
3.PROPOSED METHODOLOGY users using an intelligent scheduler. In this scenario,
the extra complexity of inter-cell interactions may not
LTE was introduced as an upgrade of the Universal justify the advantages of ICIC [9]. However, as [10]
Mobile Telecommunication Systems (UMTS) in order to points out, the benefit of bandwidth compensation is
provide cellular network users with high data rates in the not completely realized if a limit is set on the peak
uplink and downlink directions, low latency, and rate that can be employed for a specific bearer. ICIC
efficient spectrum usage [3]. The correct scheduling has been proven to be most useful for low-to-
algorithm that satisfies the environment's characteristics moderate traffic levels. CQI feedback frequency is a
and the users' requirements requests could be used to tradeoff between temporal diversity gain and
maximize spectrum utilization. effective interference reduction achieved by a
frequency-selective scheduler. More frequent CQI
feedback allows for better temporal adaptation to
channel variance, resulting in increased spectral
efficiency. The frequency of CQI feedback, on the
other hand, should be determined by the short-term
interference caused by surrounding cells'
instantaneous scheduling decisions. The effectiveness
of CQI feedbacks, particularly those from cell-edge
users, is considerably diminished because this
interference is weakly associated with that in
succeeding sub-frames. As a result, a less frequent
Figure 1: scheduling decision might be beneficial in terms of
System model for SU-MIMO UL transmission interference minimization. CQI feedback filtering,
according to [12], should be used to average out the
The coordinator anticipates the next epoch's temporal changes of the ICI, resulting in a more
interference state based on the SINR, current status, stable CQI value for cell-edge users.
and number of neighbors at each epoch. The
coordinator will broadcast a negotiation message to Because RP offers superior system performance and
the network if the next status of a WBAN is ShortIF. short-term fairness than other algorithms [4], we
The coordinators exchange the TDMA schedule in design an improved PF scheduling algorithm based
the super-frame for sharing the transmission channel on the riding peaks (RP) technique under successive
in the interference avoidance process after receiving RBs allocation restriction [4].
signals from the neighbors.
UE k's improved riding peaks (IRP) PF factor fk (i,t)
at sub-band I and time slot t is defined as:

(1)

Where we introduce a multiplicative factor βk (i, t) in

the traditional PF metric formula. This factor is
defined as:

, θ∈

Figure 2: Network Topology (0,1) (2)

1833
Shivanandaswamy V. Mahapurush et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 1832 – 1836

Where is a constant factor that can be set at eNB. 4. METHODOLOGY:

Each eNB saves the last (time slot t -1) scheduling
ALGORITHM:
results of each sub-band to reduce UL ICI variance
and increase AMC accuracy. The following is a Improved Proportional Fair Scheduling Algorithm for
representation of the scheduling history: LTE

(3) Initialization:

The collection of sub-bands is shown as u(i, t-1), The set of unallocated sub-bands: S = Λ;
where u(i, t-1) is the UE scheduled on sub-band I
during time slot t 1. When eNBs conduct the resource The set of sub-bands allocated to UE k: Μk = φ;
allocation on sub-band I at the present moment (time
Η (t) = φ ;
slot t), the scheduling controllers conduct the
following: Assume that each UE occupies the whole bandwidth
and compute the initial improved PF factor fi,k(t) for
 Calculate the traditional PF metric λt (i ,t) each UE k according to (6);
for UE k on sub-band i at time slot t , as
shown in (7). Let V be the sorted list of all the metric value fi,k(t)
 Calculate the Improved PF metric fk (i ,t) as
shown in (13). Compare UE k with the in decreasing order; m = 1.
history scheduling information u(i , t-1). If k
is not equal u (i , t-1), then we give UE k a While S ≠ φ do
lower priority with the value of θ .
Otherwise, our improved PF metric is Pick sub-band i with m-th largest metric value fi,k ∈V
identical to the traditional one, as shown in ;
(10).
if (i is adjacent to Mk ) or ( Mk = φ ) then

Assign sub-band i to UE k ;

(4)
We give lesser priority to some UEs who have not
been scheduled on the target sub-band in the past two
steps (time slot t-1). To put it another way, some UEs
have the same PF metric value. The user who was
scheduled on the previous TTI will be scheduled by
the eNB. The ICI variance caused by the frequent
changes of scheduled UEs with identical PF factors
in a few adjacent TTIs can be reduced using this
strategy. The entire algorithm is shown below: m = 1;

else

m= m+ 1.
end if

end while

1834
Shivanandaswamy V. Mahapurush et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 1832 – 1836

4.RESULTS CEUT gain will be reduced. Furthermore, the

increased chances of first transmission success and
The highest throughput is achieved by our suggested
CAUT suggest a better fit between MCS level
Improved Proportional Fair Scheduling Algorithm,
selection and instantaneous channel condition.
which is followed by RP TDA and standard RP. At
Because the IRP method performs better on CAUT
CAUT and CEUT, gains of roughly 29 percent and
and CEUT, we need to examine short-term fairness
35 percent can be reached with an equivalent of 0.8.
(less than 100msec).

Figure 5: Variation of Fairness of scheduler with

users
When the time period is less than 20 msec, our
suggested approach with an equivalent to 0.8 has
Figure 3: Ratio of scheduling vs CDF
greater short-term fairness than RP and RP TDA.
Otherwise, it performs similarly. When the time
interval is smaller than 40 msec, the IRP algorithm is
commendable with a value of less than 0.8. When the
time period is greater than 40 msec, however, it has a
lower short-term fairness. Furthermore, RP TDA
performs worse than RP in terms of short-term
fairness, despite the former's superior performance at
CAUT and CEUT.

REFERENCES
Figure 4: System throughput CDF
[1] 3GPP TS 36.213, “Evolved Universal
Terrestrial Radio Access (E-UTRA), Physical layer
The gain of the CAUT and the chance of initial
procedures (Release 10),” TS 36.213 v10.10.0, 2013.
transmission success grow steadily as the value of
[2] A. Ubisse and N. Ventura, “Modeling a link
falls, whereas the gain of the CEUT declines. These
level simulator for long term evolution uplink,” in
patterns appear to be in line with our assumptions.
The Southern Africa Telecommunication Networks
The lower the value of θ, the lower the priority of the
and Applications Conference (SATNAC ’11), East
UEs that were not scheduled at the previous TTI.
London, South Africa, September 2011.
Because of the frequency selective scheduling, the
1835
Shivanandaswamy V. Mahapurush et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 1832 – 1836

[3] 3GPP TS 36.104, “Evolved Universal [9] G. Boudreau et al., Interference Coordination
Terrestrial Radio Access (E-UTRA), Base Station and Cancellation for 4G Networks, IEEE Commun.
(BS) radio transmission and reception (Release Mag., 2009.
10),” TS 26.104 v10.11.0, 2013. [10] A. Khandekar et al., LTE-Advanced:
[4] A. Sadek and S. Shellhammer, “SUI Path-Loss Heterogeneous Networks, European Wireless Conf.
Model for Coexistence Study,” IEEE 802.19- 2010.
08/0010r0, IEEE P802.19 Wireless Coexistence, [11] O.Simeone, E.Erkip, S.Shamai, Robust
2008. Transmission and Interference Management For
[5] 3GPP TS 36.101, “Evolved Universal Femto-cells with Unreliable Network Access, IEEE
Terrestrial Radio Access (E-UTRA), User JSAC, Dec. 2010.
Equipment (UE) radio transmission and reception [12] S. Annapureddy et al., http://www.ieee-
(Release 10),” TS 36.101 v10.11.0, 2013. ctw.org/2010/mon/Gorokhov.pdf, 2010 IEEE
[6] K. Konstantinou, S. Kang, and C. Tzaras, “A Commun.Theory Wksp., Cancun, Mexico, 2010.
measurement based model for mobile-to-mobile
UMTS links,” in Proceedings of the IEEE 65th
Vehicular Technology Conference (VTC ’07), pp.
529–533, Dublin, Ireland, April 2007.
[7] T. Nihtila and V. Haikola, HSDPA Performance
with Dual Stream MIMO in a Combined Macro-
Femto Cell Network, IEEE VTC 2010.
[8] H. R. Karimi et al., Evolution Towards
Dynamic Spectrum Sharing in Mobile
Communications, IEEE PIMRC 2006.

1836