Content uploaded by Nadeem Javaid
Author content
All content in this area was uploaded by Nadeem Javaid on Feb 18, 2020
Content may be subject to copyright.
2/18/20
Buffer-aided cooperative communication in wireless networks
PhD thesis defense by Hina Nasir
1
Buffer-Aided Cooperative Communication
in Wireless Networks
Hina Nasir
136 - FBAS / PHDCS / F15
Doctoral Thesis Defense Presentation
Supervisor:
Dr. Nadeem Javaid,
Associate Professor,
CUI, Islamabad
Department of Computer Science and Software Engineering,
International Islamic University, Islamabad
Co-supervisor:
Dr. Husnain Abbas Naqvi,
Assistant Professor,
IIUI, Islamabad
January 28, 2020
Agenda
Buffer-aided cooperative
communication
Existing literature
Problem statement
Proposed solution and results
Conclusion and future works
Q & A session
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
2
Analogy
3
Sara
Dawood
Rose
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
Analogy
explained
4
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
Introduction (1/2)
Buffer-Aided Cooperative Relaying
Transmission schedule
Fixed
Non-fixed
Relaying techniques
Amplify and forward (AF)
Decode and forward (DF)
Challenges involved [1]
Channel state information
Buffer state information
oFull buffer cannot receive data
oEmpty buffer cannot transmit data
Delay
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
5
!"#$# ##%$$ ##
&#'''( ##)&*#+,-$$-."+/$#,0-
Fig. 1: Cooperating relaying techniques
Amplify and forward Decode and forward
Introduction (2/2)
Energy Harvesting Cooperative Communication
Radio frequency (RF) signals carry energy and information simultaneously
Simultaneous wireless information and power transfer (SWIPT)
Energy harvesting is performed using time switching or power splitting
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
6
Fig. 2: SWIPT system
Literature Review
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
7
Literature Review (1/4)
Link Quality based Relay Selection
Max-max relay selection (MMRS)
scheme [2]
Fixed transmission schedule
oBest quality SR link in odd time-slot
oBest quality RD link in even time-slot
Max-link relay selection (MLRS)
scheme [3] and [4]
Non-fixed transmission schedule
oBest quality SR or RD link in each time-slot
Hybrid relay selection [5]
Fixed transmission schedule
Reduced delay relay selection
scheme [6]
Delay is 2 time-slots
Generalized relay selection [7]
Multiple activation of SR links
simultaneously
Loco-link relay selection scheme
[8]
Deals with outdated CSI
Switched MLRS [9]
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
8
,1)#$ )2")23#34# !#'''#5#( ##
/$$,-"/+),0,
/6###( 7$8 !"#3#$###334# 3#9'''
#5#( ##+$$:+;":.;,0,
-#<(=<&(7( !"#2"#3##$#3""4$#4'''
#># &3.-,$$++/"+.+,0+
+)?#4( &# #3#@3# !"#($#A)2B)2)2BC#3
#'''#># &3.+$$:,/.":,-.,0.
.#=<&<(7( !"#3#5# A13#($#4'''#
># &3../$$,+.;",+;+),0;
;)?#4 &# "A"13<#* !"#3#&A%# ##) #$ ""3
C##'''-$$/./,"/.-.,0.
D#( 1> 3# <66&###C4"($2#3 !"#C##5#? (E
'#'''#( ##..D$$/.:-"/;0. &,0D
:EC1 (C4#)2"C# !"#3#($#) #$"3((,0:F,
#<(3( ##(#&<3,0:$$".
Fig. 3: Link quality based relay
selection
Limitations
Liable to full or empty buffers
Large queuing delay
Homogenous buffer size
Limitations
Liable to full or empty buffers
Large queuing delay
Homogenous buffer size
Literature Review (2/4)
Buffer Status based Relay Selection
Max-weight relay selection (MWRS) scheme [10] and [11]
Weight assignment on the basis of buffer occupancy
oAvailable buffer space is SR weight
oOccupied buffer space is RD weight
Non-fixed transmission schedule
Diversity gain is 2K at small buffer size
Delay (1+KL / 2) time-slots
Generalized buffer status based rely selection [12]
Shortest in longest out [13]
Fixed transmission schedule
Diversity gain is K at small buffer size
Priority based max-link [14]
Priority class to full empty and partially filled buffer
Link with the highest quality is selected among the priority class
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
9
0AG 1#&6###G1###&?$#1##3<## !"#345# !#*'''
#># &3.+0$$D;DD"D;:-?,0.
)?#4# &# !"""#&"$#3""E4($#4#'''5#
( ##C.+$$.;-".;;?,0;
,*# &# <#* !""3##($#(&###4#
'''D$$.--".+;,0,0
/CC#6@C# 3#$#3#&44# !'''#># &3
.+$$.+.,B.+;, &,0.
-)H6)#)&A3# !"#A##3")2"C#3##1E
($#4#'''#( ##..;$$,D,.",D/:7 3,0D
Fig. 4: Buffer status based relay selection
Limitations
Random link selection in case of equal weight links
Large queuing delay
Limitations
Random link selection in case of equal weight links
Large queuing delay
Literature Review (3/4)
Packet Selection Scheme
Inverse channel packet matching [14]
Packets that experienced bad channel conditions in SR hop go through good channel
conditions in RD hop and vice versa
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
10
+<C#(1&1C# <C#=&? &3# "$##4# !"#$#3""
43,0-'''D0># &3(I>(,0-"EJ''',0-
Fig. 5: Packet selection on buffer-aided relay
Limitations
Scheme operation in two segregated phases is a
barrier in diversity gain and delay
Limitations
Scheme operation in two segregated phases is a
barrier in diversity gain and delay
Literature Review (4/4)
Energy Harvesting Cooperative Communication
Time switching based relaying
(TSR) [16]
Fraction of time for energy harvesting
Remaining time for information processing
Power splitting based relaying
(PSR) [17]
Fraction of power for energy harvesting
Remaining power for information processing
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
11
.#G 1 #6383#&$4#&3#&##
$#&K9'''#5#( ##,$$/.,,""/./.7 3,0/
; A& 3)>*A3#&3#"$&3#&$
$#3""4" $23#&47 '&##&($ #,,
$$,"/0,0D
Fig. 7: Power splitting based relaying
Fig. 6: Time switching based relaying
Limitations
TSR reduces time for information processing
PSR reduces received signal power
Limitations
TSR reduces time for information processing
PSR reduces received signal power
Hybrid of TSR and PSR [17]
Research Gaps
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
12
Research Gaps
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
13
Link quality based relaying
Work either focuses on relay selection [2-5] or on packet selection [14]
The buffer size is considered homogenous [2-13]
The use of sequential access buffers bounds the bad data lying at head of the buffer to be
transmitted to the destination [2-8]
Diversity gain is only a function of number of relays [2-9]
The relays remain silent in case of corrupted signal because the schemes are based solely on
AF [2] or DF [3,4] modes
Link weight based relaying
Random link selection does not guarantee the selection of best link [10-11]
Traditional weight allocation increases the queuing delay [10-12]
Energy harvesting cooperative communication
TSR compromises on the throughput
PSR compromises on the outage probability
There is a need for unified framework that acts as TSR, PSR and hybrid of them [16,17]
Statement of Problem
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
14
H#2$#3$#
! ##3&# #
3 3
##33$ 3L #
3###4
@ ##$!#3#
# &$##3
1# &$
###4#&3
# #$$$###
#$$$###&
##
@ ##$!#3#
# &$##3
#3
1# &$
#M4
$#N2###3 3
A3#4##3
Aims and Objectives of Research
To devise techniques to improve the outage probability, end-
to-end delay and throughput performances
1. To analyse the state-of-the-art buffer-aided cooperative communication
techniques and gain insight with respect to relay and packet selection schemes
2. To investigate relay selection schemes and identify research gaps which
impacts on the performance of outage probability, delay and throughput
3. To propose and design the relay selection schemes and to derive the closed-
form expressions for the outage probability, average end-to-end delay and
throughput
4. To validate and evaluate the performance of proposed schemes by comparing
them with the state-of-the-art schemes on buffer-aided relay and packet
selection
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
15
Proposed Solutions
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
16
Solution 1
Design and Analysis of Buffer-Aided Packet and Relay Selection Schemes
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
17
Challenge addressed:
1. Joint relay and packet selection
2. Random access buffers
!"#!$
!$%&'()!*!!+!
+#!"+,++-.&'(##/0) /00
& 1*234%!5!
6 4"1!4!"*7"
+*8!!1!7)'/
,#!!!!# $-
System Model 1: Packet Selection
S, R and D
R is AF relay with buffer of size L
Nodes are half duplex
Buffers allow random access
Transmission scheme
Incremental relaying model
Phase 1: S transmits
Phase 2: R transmits
Maximal ratio combining at D
Packet selection criteria
(s1.9)
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
18
Fig. 8: System model for buffer-aided packet selection using
incremental relaying
Performance Analysis
Outage Probability
(s1.11)
Average end-to-end delay
(s1.13)
(s1.14)
is outage of direct link
Throughput
(s1.15)
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
20
System Model 2: Joint Packet and
Relay Selection
S, D and
Finite and homogenous buffer at R
Full buffer cannot receive
Empty buffer cannot transmit
When a packet enters, buffer is incremented by 1
When a packet leaves, buffer is decremented by 1
Relays are HD and follow AF relaying mode
Each node has single antenna resource
Channel Model
All links follow i.i.d Rayleigh fading
Fading envelop is constant for one time-slot and change independently from one time slot
to another
Link outage when equivalent SNR is less than
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
21
Fig. 9: System model for joint packet
and relay selection
Joint Packet and Relay Selection
(JPARS) Scheme
SNR based JPARS scheme
Phase 1: Relay selection
o (s1.16)
Phase 2: Packet selection
o (s1.17)
Outage probability
o (s1.18)
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
22
Fig. 9: System model for joint
packet and relay selection
Joint Packet and Relay Selection
(JPARS) Scheme
BSB based JPARS scheme
Phase 1: Relay selection
(s1.19)
Phase 2: Packet selection
oSame as in JPARS SNR
Outage probability
o (s1.20)
oWhere, (s1.21)
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
23
Fig. 10: System model for joint
packet and relay selection
Simulation Setup
=3 dB
K=1, 2 and 3
L=2, 8 and 16
P
s
=P
r
=P =1 Watt
SNR 1 dB-25 dB
Analytical is verified by Monte Carlo simulations
10
6
simulation runs
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
25
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
26
Fig. 11: Comparison of outage probability against average SNR for K=1, L=1,2,16 Fig. 12: Significance of direct link: outage probability of AF-BAS
against average SNR for K=1, L=2
Fig. 13: Average end-
to-end packet delay and
throughput of AF-BAS
against average SNR
for increasing buffer
size
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
27
Fig. 14: Outage probability of the proposed JPARS, AF Max-Link and AF-dual-
hop schemes with the combinations of (K,L) as (1,2), (2,2) and (2,8).
Fig. 15: Outage probability of JPARS-SNR scheme for L = 4, K = 2
and 3, = 2 and 3 and = 1.
Fig. 16: Outage probability of JPARS-SNR scheme for L = 4, K = 2 and 3, =
1and = 2 and 3.
Average Delay
Symmetric channel Asymmetric channel
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
28
Summary of Findings
The idea of packet selection over multiple relays improved the overall outage
probability of the system.
The symmetric channel configuration shows the better outage probability as
compared to the asymmetric channel configuration.
The maximum packet delay is reduced in the symmetric and asymmetric
channel conditions, however, when RD link is given priority over the SR
link, the improvement in the delay is not very significant.
Limitations
The packet selection over multiple relays gives a nominal decrease in outage
probability.
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
29
Solution 2
Virtual Link Relay Selection Scheme for Buffer-Aided Relaying
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
30
Challenges addressed:
1. Can the diversity gain be a function of both the number of relays and the buffer size in
case of buffer-aided CR?
2. How can we look at buffers so that the diversity gain becomes the function of buffer size
also?
3. Asymmetric buffers
1*4$"*!9$
6 4"+#!!$%&'0/!*
!!+!:+#!"+,++-
;&'0##0)< 0//
& 4=*.! ;%
1!1*6 4+#!!$%>>>
4?##(/?/0 (/?@<A&'07/'0
System Model 3: Virtual Relaying
S, D and
Finite and non homogenous buffer at R
Buffers allow random access
Full buffer cannot receive
Empty buffer cannot transmit
When a packet enters, buffer is incremented by 1
When a packet leaves, buffer is decremented by 1
Relays are HD and DF having antennas
Virtual relays
Number of packets in buffer
Relay selection
(s2.1)
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
31
Fig. 17: System model for virtual link relay
selection scheme
Performance Analysis
Outage probability of system using Markov Chain
(s2.2)
Where,
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
32
Delay
(s2.3)
Average throughput
(s2.4)
Simulation Setup
=3 dB
Monte-Carlo simulations
1000000 runs
Rayleigh fading channels is considered
K, L are specified for each result
Ps=Pr=P =1 Watt
SNR 1dB -25 dB
Four scenarios
Symmetric channel symmetric buffer
Symmetric channel asymmetric buffer
Asymmetric channel symmetric buffer
Asymmetric channel asymmetric buffer
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
33
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
34
Fig. 20: Outage probability for asymmetric buffer size and symmetric
channels, Lmax = 3, K = 2
Fig. 18: Symmetric buffer size and symmetric channels, K = 2 and variable
buffer L = 1, 2, 3, 4 and 5 Fig. 19: Symmetric buffer size and asymmetric channels, = 1,2 and 3 and =
1, 2 and 3
Fig. 21: Outage probability for asymmetric buffer size and asymmetric
channels, Lmax = 3, K = 2, = 1 and 2 and = 1 and 2
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
35
Fig. 22: Symmetric buffer size and symmetric channels, K=2, L=1, 2, 3, 4
and 5
Fig. 23: Symmetric buffer size and asymmetric channels, V=6, = 1,2 and 3
and = 1, 2 and 3
Fig. 24: Asymmetric buffer size and symmetric channels, Lmax = 3, K = 2 Fig. 25: Asymmetric buffer size and asymmetric channels, Lmax = 3, K = 2, =
1 and 2 and = 1 and 2
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
36
Fig. 26: symmetric buffer size and symmetric channels Fig. 27: Symmetric buffer size and asymmetric channels
Fig. 28: Asymmetric buffer size and symmetric channels, Lmax = 3, K = 2 Fig. 29: Asymmetric buffer size and asymmetric channels, Lmax = 3, K =
2, = 1 and 2 and = 1 and 2
Summary of Findings
The virtual relay design considerably increases the diversity gain using a
small number of relays.
The same diversity gain is achieved by either increasing the number of
relays keeping the buffer size constant or increasing the buffer size keeping
the number of relays constant.
The symmetric channel conditions and buffer sizes perform better as
compared to the asymmetric case.
Limitations
Maximum achievable diversity gain is L instead of 2L
The buffer size cannot be very large
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
37
Solution 3
HDAF based Relay Selection Scheme for Buffer-Aided Relaying
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
38
Challenges addressed:
1. The relays remain silent on corrupted signal quality because of
working in only DF or AF relaying modes.
*
B!"!5 4# 7$"!
6 4@)>>>!!+
+!,++-&'<
System Model 4: HDAF Relaying
Relay selection
(s3.1)
Transmission Scheme
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
39
Fig. 30: System model for HDAF
relaying
Fig. 31: Proposed HDAF relaying protocol
Performance Analysis
Outage probability of system
(s3.2)
Where
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
40
Simulation Setup
=3 dB
T=3 dB
K=3
L=3
Monte-Carlo simulations
10 00 000 runs
P
s
=P
r
=P =1 Watt
SNR 1dB -25 dB
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
41
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
42
Fig. 32: Outage probability of MLRS-HDAF and MMRS-HADF
schemes for K = 3 and L = 3
Fig. 35: Average throughput of MLRS-HDAF, MLRS-DF and MLRS-AF schemes
for K = 3 and L = 3
Fig. 34: Average end-to-end delay of MLRS-HDAF and MMRS-HDAF
schemes for increasing number of relays and L = 10
Fig. 33: Outage probability of MLRS-HDAF, MLRS-DF and
MLRS-AF schemes for K = 3 and L = 3
Summary of Findings
The proposed scheme using HDAF mode achieved the better outage probability as compared to
SNR based buffer-aided relay selection schemes using only AF or DF relaying protocols.
The average end-to-end queuing delay of HDAF is same as in SNR based AF or DF schemes i.e.,
1 + KL at high SNR.
Limitations
High probability of buffer overflow
High delay
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
43
Solution 4
Weight based relay selection schemes
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
44
A4=*6 4
1!!*>2 "*!;%+#!!$%
>>>+!;!!&&##)) )?A&'0A7&(
& AC4"D4C*1
691E6 9**1!1*+#!
!$%>>>4A(## &)'0< &)'<<&'<7/'0
Challenges addressed:
1. Multiple links with equal weights
2. How to control the selection of SR and RD links without increasing
delay?
System Model 5: Weight based Relay
Selection
S, D,
Total number of relays K
Direct link does not exist
Channel gain gi
Available buffer space (ABS) is
Occupied buffer space (OBS) is =
A link is
unqualified if
Available if corresponding buffer is neither full nor empty
Success if link is available and qualified
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
45
Fig. 36: System model for buffer status
based relay selection
Multiple Links with Equal Weights
Multiple links have equal weights
Single side equal weight
Both side equal weight
For K=2, L=2, 0 out of 9 states have distinct weights
For K=2, L=3, 8 out of 16 states have distinct weights
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
46
Fig. 37: Case of multiple links with equal
weights
Improved Max-Weight Schemes
Link selection factor
imax-weight quality
(s4.1)
imax-weight priority
(s4.2)
where, = /L is link prioritization factor
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
47
Performance Analysis
o
Delay
(s4.4)
Where,
Throughput
(s4.5)
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
48
Link to Weight Analysis
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
49
Fig. 38: Weights of SR and RD links with respect to buffer size Fig. 39: Proposed weight assignment using Lth.
(s4.6)
(s4.7)
Outage Probability Analysis
(s4.8)
{A link is chosen from },
{All the links are unqualified in },
{At least one link is decode-able from }.
and are independent events
happens only, if both and happen
Occurrence probability of is expressed as,
(s4.9)
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
50
Average Delay and Throughput
Analysis
Throughput
(s4.10)
(s4.11)
Delay
(s4.12)
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
51
Simulation Setup:
Monte-Carlo simulations
10 00 000 runs
K= 2, 3 and 4
L= 4
Rayleigh fading channel
L
th
=1, 2 and 3
P
s
=P
r
=P =1 Watt
SNR 1-25 dB
=3
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
52
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
53
Fig. 42: Average end-to-end delay with respect to average SNR for K
= 2, L = 3 Fig. 43: Average throughput with respect to average SNR for K = 2, L = 3
Fig. 41: Asymmetric conditions, = 0.5 and = 0.5 for K = 2, L =3.
Fig. 40: Symmetric conditions, = for all i = [1 : 2K] and K = 2, L = 3.
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
54
Fig. 46: The average end-to-end delay against the average SNR for L
= 4 and K = 2 of the MWRS, MLRS and the proposed BTRS schemes Fig. 47: Average throughput against average SNR for L = 4 and K = 2 of the
MWRS, MLRS and the proposed BTRS scheme
Fig. 44: Outage probability against the average SNR for L = 4 and K =
2 of the MWRS, MLRS and BTRS schemes Fig. 45: Outage probability against the average SNR for increasing values of
K at L = 4 of the MWRS, MLRS and BTRS schemes
Summary of Findings
Considering SNR as the second selection metric, the outage probability is improved
Considering RD links prioritization as the second selection metric, the delay and
throughput performances are improved
Buffer threshold improved the performance of the relaying system for the average end-to-
end queuing delay and the average throughput at the cost of the increased outage
probability
Limitations
Outage probability is traded with delay
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
55
Solution 5
Unification of Energy Harvesting Schemes
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
56
F!7>"!"1*
G"* 7"+*1!!!!
!+!7&(
,F$-
Challenges addressed:
1. Designing of a unified model
System Model 6: Energy Harvesting
Cooperative Communication
Dual-hop cooperative system
S, R, D
Channel gain g
Distance d
Hybrid relay
Harvest energy
Process information
Fig. 48: System model for energy harvesting
cooperative communication
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
57
Energy Harvesting Model
Harvested energy
(s5.1)
(s5.2)
Harvester power
(s5.3)
(s5.4)
(s5.5)
(s5.6)
Fig. 49: GEHR model
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
58
Performance Analysis
? &A##3
= (s4.11)
(s4.12)
Average Throughput
Delay limited case
o(s5.10)
Delay tolerant case
o(s5.11)
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
59
Simulation Setup
Ps=1 Watt
I=0.1
h1= 1 dB-25 dB
d1=3 m, d2=1 m
=3 dB
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
61
Average Throughput in Rayleigh
Fading
2/18/20
Buffer-aided cooperative communication in wireless networks
PhD thesis defense by Hina Nasir
62
DF AF
Fig. 50: GEHR working as TSR (= 0.3)
and PSR (= 0.3) in DF relaying
Fig. 51: GEHR working as TSR (= 0.3)
and PSR (= 0.3) in AF relaying
Fig. 52: GEHR working as HEHR (= 0.3)
and PSR (= 0.3) in AF and DF relaying
AF and DF
Summary of Findings
The proposed GEHR scheme can be set to work as not only for the conventional TSR and
PSR schemes, but also for hybrid TSR and PSR scheme.
The throughput of hybrid framework can be maximized if optimal values of time and power
splitting factors are used
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
65
Conclusions and Future Research Directions
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
66
Conclusions
The idea of joint exploration of relay and packet selection significantly improves the
overall system performance.
The proposed scheme using HDAF mode achieved the better outage probability as
compared to SNR based buffer-aided relay selection schemes using only AF or DF
relaying protocols.
By considering the second selection metric, the system’s performance is improved.
The idea of weight reallocation helps to reduce the queuing delay without increasing
buffer occupancy.
The proposed GEHR scheme can be set to work as not only for the conventional TSR
and PSR schemes, but also for hybrid TSR and PSR scheme.
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
67
Future Work
Buffer-aided SWIPT system with successive relaying along with inter relay interference
capability. The interfering signal can be utilized to charge the node's battery.
Another aspect to extend this work is to consider the secrecy performance analysis in cognitive
radio paradigm using non orthogonal multiple access
The HDAF relaying considering buffer status along with link quality in relay selection is a
possible dimension to explore.
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
68
Other Publications
Journal
1. Nadeem Javaid, Arshad Sher, Hina Nasir and Nadra Guizani, "Intelligence in IoT based 5G Networks: Opportunities and
Challenges, "in IEEE Communications Magazine, vol. 56, no. 10, pp. 94-100, Oct. 2018.
2. Hina Nasir, Nadeem Javaid, Muhammad Sher, Umar Qasim, Zahoor Ali Khan, Nabil Alrajeh, and Iftikhar Azim Niaz.
"Exploiting outage and error probability of cooperative incremental relaying in underwater wireless sensor networks," Sensors
vol. 16, no. 7, pp. 1076, Jun. 2016.
Conference
1. Hina Nasir, Nadeem Javaid, Shaharyar Mahmood, Umar Qasim, Zahoor Ali Khan, and Furqan Ahmed. "Distributed Topology
Control Protocols for Underwater Sensor Networks," 2016 19th International Conference on Network-Based Information Systems
(NBiS), Ostrava, 2016, pp. 429-436.
2. Hina Nasir, Nadeem Javaid, Hifsa Ashraf, S. Manzoor, Zahoor Ali Khan, Umar Qasim, and M. Sher. "CoDBR: cooperative depth
based routing for underwater wireless sensor networks." 2014 Ninth International Conference on Broadband and Wireless
Computing, Communication and Applications, Guangdong, 2014, pp. 52-57.
3. Hina Nasir, Nadeem Javaid, Muniba Murtaza, S. Manzoor, Zahoor Ali Khan, Umar Qasim, and Muhammad Sher. "ACE:
Adaptive cooperation in EEDBR for underwater wireless sensor networks." In Broadband and Wireless Computing,
Communication and Applications (BWCCA), 2014 Ninth International Conference on, pp. 8-14. IEEE, 2014.
4. Shaharyar Mahmood, Hina Nasir, Saad Tariq, Humaira Ashraf, Mahmood Pervaiz, Zahoor A Khan, and Nadeem Javaid.
"Forwarding nodes constraint based DBR (CDBR) and EEDBR (CEEDBR) in underwater WSNs." Procedia Computer Science 34
(2014): 228-235.
5. Hina Nasir, Nadeem Javaid, Muhammad Imran, Muhammad Shoaib, and Mehmoon Anwar, "Simultaneous Wireless
Information and Power Transfer for Buffer-Aided Cooperative Relaying Systems," 2018 14th International Wireless
Communications & Mobile Computing Conference (IWCMC), Limassol, 2018, pp. 845-849.
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
69
Thankyou
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
70
Questions and
Answers Session
2/18/20
Buffer-Aided Cooperative Communication in Wireless Networks
Ph.D. thesis defense by Hina Nasir
71