Cooperative Communication-Aware Spectrum Leasing in Cognitive Radio Networks
Y. Yi†, J. Zhang†, Q. Zhang†, T. Jiang‡, J. Zhang╫ †Hong Kong University of Science and Technology ‡ Huazhong University of Science and Technology ╫ Huawei Technologies Co., Ltd.
Presenter: Yi Youwen (
[email protected]) April 5, 2010
Outline
Introduction
Game Theory Analysis
Background System Model Utility Functions Stackelberg Game
Simulation Results Conclusions
Outline
Introduction
Game Theory Analysis
Background System Model Utility Functions Stackelberg Game
Simulation Results Conclusions
Cognitive Radio
Fixed spectrum allocation is inefficient
Spectrum scarcity: most of the spectrum band is allocated Spectrum hole: spectrum is not fully utilized
One solution: cognitive radio (CR)
Primary users (PU): own the spectrum Secondary users (SU): opportunistically access without interference
1
CCRN
Cooperative cognitive radio networks (CCRN)
PU leverages SU as cooperative relay Data rate of PU will increase, and results in more access opportunities Secondary Network
Payment
Access Opportunity Primary Network
2
Network Cooperation
Challenges:
How much opportunity the secondary network can obtain, and the according payment? What is the optimal relay selection?
Cooperative Communication-Aware Spectrum Leasing Framework:
Design a negotiation mechanism for each network to maximize its utility with regards to data rate and revenue. Both have motivation to cooperate, and relay selection can be done within polynomial time. 3
Outline
Introduction
Game Theory Analysis
Background System Model Utility Functions Stackelberg Game
Simulation Results Conclusions
System Model
Transmission Mode Primary transmission
Cooperative transmission
SU -> PAP
Secondary transmission
PU -> SU, PU->PAP
SU ->SAP
Access Mode TDMA 4
Outline
Introduction
Game Theory Analysis
Background System Model Utility Functions Stackelberg Game
Simulation Results Conclusions
Game Theory Analysis Utility Functions
1 0.9
a=2 a=6
0.8
Primary network
0.7
U
d
0.6 0.5 0.4 0.3 0.2 0.1 0
0
0.5
1
1.5
2
7 b=0.4 b=0.8
6
5
Secondary network
4
us
3 2 1 0 0
2
4
6
8
Rs
5
10
Outline
Introduction
Game Theory Analysis
Background System Model Utility Functions Stackelberg Game
Simulation Results Conclusions
Game Theory Analysis
Spectrum Leasing Framework
Backward Induction
Given primary network’s strategy, find the optimal reaction of secondary network’s Assuming the secondary network will act as its optimal reaction, the primary network selects its strategy to maximize its utility 6
Game Theory Analysis
The best strategy of the secondary network
Given price c, the SN determines A convex optimal problem
The best response function:
7
Game Theory Analysis
The Best Strategy of Primary Network
Given the best response function of the SN , the PN determines the relay selection and the price per unit access time c Total access time constrain
Relay selection constrain
Primary transmission mode
8
Game Theory Analysis
Relay Selection User
……
……
Virtual node
……
……
Min-cost Problem & Hungarian Method 9
Game Theory Analysis
Price Determination
A convex problem, and unique solution exist
10
Outline
Introduction
Game Theory Analysis
Background System Model Utility Functions Stackelberg Game
Simulation Results Conclusions
Simulation Results
Setup
Assumptions
PU
Fading channel No power control
SU
Some key parameters
Kp = 10 Ks = 5 ~ 20 rp = 100m rs = 70m d = 20m
d PU
rp
PAP
SU
SAP
rs
PU
SU
PU
11
Simulation Results
Both PN and SN have motivation to share the spectrum, and involve cooperative transmission. 200
14 CSL NCSL NSL
190
12 10
170
8
Us
Up
180
CSL NCSL NSL
160
6
150
4
140
2
130
5
10
15 Ks
20
0 5
10
15
20
Ks
Fig. 1 The utility of PN and SN versus the number of secondary users
12
Simulation Results
15
210
CSL NCSL NSL
CSL NCSL NSL
200 190
10
180 170
Us
With heavier traffic load, the PN gain less from spectrum leasing, while gain more from cooperative transmission. Cooperative transmission still benefits both networks
Up
160
5
150 140 130 120 0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
0 0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
Fig. 2 The utility of PN and SN versus the traffic demand of the primary users
13
Outline
Introduction
Game Theory Analysis
Background System Model Utility Functions Backward Induction
Simulation Results Conclusions
Conclusions
Cooperative Communication-Aware Spectrum Leasing Framework
Game Theory Analysis
Two collocated infrastructure-based primary and secondary networks can share the spectrum
The spectrum leasing problem is modeled as a stackelberg game. An unique Nash Equilibrium exists which gives the optimal strategies for both networks.
Cooperation Benefits
Simulation results show that both primary and secondary networks achieve higher utility when leveraging cooperative transmission.
14
Thank you! Q&A Email:
[email protected]
Simulation Results (cont.) Different number of secondary users 0.7
100 CSL NCSL
0.65
CSL NCSL
95
0.6
90
The spectrum price c
The spectrum leased to secondary network s
0.55 0.5 0.45
85 80 75
0.4
70
0.35
65 5
10 15 The number of secondary uers Ks
20
60
5
10 15 The number of secondary uers Ks
20
Simulation Results (cont.)
Different traffic loads of primary network 1
The spectrum leased to secondary network s
90 CSL NCSL
The spectrum price c
85
80
75
70 0.2
0.4
0.6 0.8 1 1.2 The traffic demand coefficient
1.4
1.6
CSL NCSL
0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.2
0.4
0.6 0.8 1 1.2 The traffic demand coefficient
1.4
1.6