Volume 4, Issue 1
network;
Performance
I. INTRODUCTION
15
Section V.
UE1
Frequency
UE0
SCHEDULER
UEn
Time Varying
Fluctuating
Channel
Time
Tk( S ) Tk( P )
where
(S )
k
( P)
k
scheduler S.
This scheduling algorithm assigns the RBs to the UE with
the best relative channel quality, i.e., a combination of CQI
and level of fairness desired. There are various versions of PF
scheduling based on values that it takes into account. The main
goal of this scheduling algorithm is to achieve a balance
between maximizing the cell throughput and fairness by letting
all users achieve a minimum QoS (Quality of Service). Such
an algorithm is designed to be better in terms of average user
throughput, as well as being fair to most of the users and
meeting the minimum QoS requirements during the scheduling
process.
TTI Start
check if (N_RB/N_UE) is
integer ?
NO
MaxMin Scheduler:
Maximizing the minimum of the UE throughputs is the main
task of MaxMin scheduler. MaxMin scheduler is a scheduler
scheme that is able to maximize the minimum data rate of
resources. The fairness of MaxMin provides lower average
throughputs where UE located far from base station (least
expensive data flow) is assigned all the capacity that it can use.
Based on Pareto optimal, without decreasing the rate of other
UE that has a lower rate, the rate of one UE cannot be
increased.
Resource Fair Scheduler (RF):
The RF scheduler scheme allocates an equal amount of
resources for all UEs. It mainly aims to maximize the sum rate
of all UEs while ensuring fairness with respect to the number
of RBs assigned to a UE. To achieve this goal the following
additional constraint is imposed:
bk
N
K
YES
NO
for all k
N
If N/K is non-integer, some UEs will get while others
K
N
K . This decision should be made randomly in
order to guarantee fairness.
UEs get
A. Performance measures
The main purpose of the simulations was to compare the
performance of six different scheduling schemes in addition to
18
Throughput
B
t sim
Schedulers
Value
10MHz
100
600
varies
varies
1
3GPP TU
1 transmit, 1 receive (1 x 1)
Zero Forcing ZF
Round Robin (RR)
Best CQI (BCQI)
Approximate Max. Throughput (AMT)
Kwan Max. Throughput (KMT)
Proportional Fair (PF)
MaxMin.
Resource Fair
Proposed (RR-CQI)
Rk
k 1
f ( R1 , R2 ,....., RK ) K
K ( Rk ) 2
k 1
22
30
20
18
16
RR
BCQI
KMT
PF
MaxMin
RF
20
Throughput [Mbps]
Throughput [Mbps]
25
15
RR
BCQI
KMT
PF
MaxMin
RF
14
12
10
8
6
4
10
2
5
10
15
Number of UEs
20
25
10
15
20
SNR [dB]
25
30
35
40
0.99
0.95
RR
BCQI
KMT
PF
MaxMin
RF
0.9
0.85
0.8
0.98
0.97
RR
BCQI
KMT
PF
MaxMin
RF
0.96
0.95
0.94
0.93
0.92
0.75
0.91
0.7
0
10
20
30
Number of UEs
40
50
10
15
SNR [dB]
20
25
30
1.05
Subframes=25, cqi=7=16QAM, SNR=10, MUSISO-1x1, PedB, 10MHz
0.9
0.85
0.75
10
20
30
Number of UEs
40
50
Fig. 10 Fairness achieved versus SNR for RR-CQI, RR and BCQI schedulers
Throughput [Mbit/s]
35
30
RR
RR-CQI
BCQI
25
20
15
10
5
35
10
15
20
25
30
SNR [dB]
35
40
45
50
25
20
1.05
Subframes=50, cqi=7=16QAM, 10Users-MUSISO-1x1, PedB, 10MHz
1
15
Throughput [Mbit/s]
RR
RR-CQI
BCQI
0.8
10
0.95
10
15
20
25
30
Number of UEs
35
40
45
50
RR
RR-CQI
BCQI
0.95
0.9
0.85
10
15
20
25
30
SNR [dB]
35
40
45
50
Fig. 12 Fairness achieved versus SNR for RR-CQI, RR and BCQI schedulers
[4]
[5]
[6]
[7]
V. CONCLUSION
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
ACKNOWLEDGMENT
[20]
[21]
REFERENCES
[1]
[2]
[3]
[22]
22