1. AUTONOMOUS DISTRIBUTED POWER CONTROL FOR COGNITIVE RADIO NETWORKS Sooyeol Im; Jeon, H.; Hyuckjae Lee; IEEE Vehicular Technology Conference, 2008. VTC 2008-Fall 1

2. Outline 2  Introduction  System Model  Related Works  Autonomous Distributed Power Control for Cognitive Radio Networks  Simulations  Conclusions  Comments

3. Introduction 3  In 2004, FCC proposed that unlicensed devices flexibly utilize the TV spectrum with no harmful interference [2][3].  The cognitive radio allows the secondary user (SU) to opportunistically access the spectrum licensed to the primary user (PU) without interfering with the PU.  Power control of CR networks is more complex than cellular networks  The quality of service requirements of the secondary user.  The protection of the primary users’ communication.  Secondary users should always check the estimated interference at the primary receiver after determining their transmission power. [2] FCC, “Notice of Proposed Rule Making,” ET Docket No. 04-113, May 2004. [3] M. J. Marcus, “Unlicensed Cognitive Sharing of TV Spectrum: The Controversy at the Federal Communications Commission,” IEEE Commun. Mag., vol. 43, no. 5, pp. 24-25, May 2005.

4. Introduction (cont’d) 4  Centralized Power Control  A central manager controls the transmission power of all users within its coverage.  Fuzzy Logic System (built-in fuzzy power controller) [5] Dynamically adjust its transmission power in response to the changes of the interference level caused by the SU to PU.  In [6], the optimal power control problem was modeled as a concave minimization problem.  Drawback all information required for managing the network should be known to the central entity very heavy signaling is inevitable [5] J. T. Le and Q. Liang, “An Efficient Power Control Scheme for Cognitive Radios,” in IEEE Wireless Communications and Networking Conference, 2007 [6] W. Wang, T. Peng, and W. Wang, “Optimal Power Control under Interference Temperature Constraint in Cognitive Radio Network,” in IEEE Wireless Communications and Networking Conference, 2007,

5. Introduction (cont’d) 5  Distributed Power Control  Each user controls its transmission power by itself using only local information.  An additional process to satisfy QoS for the PU Let the SUs recognize the interference temperature at the primary receiver. In [7], a joint coordination and power control algorithm was proposed. Coordination phase and power control phase In [8], a genie-aided distributed power control algorithm was proposed.   They are not fully autonomous distributed power control algorithm. [7] Y. Xing and R. Chandramouli, “QoS Constrained Secondary Spectrum Sharing,” in IEEE Dynamic Spectrum Access Networks, 2005 [8] L. Qian, X. Li, J. Attia, and Z. Gajic, “Power Control for Cognitive Radio Ad Hoc Networks,” in IEEE Local and Metropolitan Area Networks, 2007

6. Introduction (cont’d) 6  In this paper,  Fully autonomous distributed power control scheme without an additional process for CR networks  The constraint for the sum of the interference of PU induced by all SUs in the network is replaced by new one which limits the individual transmission power.  The individual transmission power constraint and the proposed scheme are numerically derived.  The simulation results demonstrate that the proposed scheme never exceed the interference temperature limit (ITL) of the PU

7. System Model 7  Network Architecture  One Primary transmitter  N secondary users  P TV : TX power of TV STA  P i : TX power of SU  α 1, α 2 : the path loss factor of the TV and the SU. α 1 < α 2 Primary users Secondary users h

8. System Model (cont’d) 8  Two QoS requirements  Interference temperature for primary user PU should be able to communicate whenever it wants. The total amount of interference at the primary receiver caused by all SU’s opportunistic communications should be less than the ITL (Interference Temperature Limit). G TV,i : the link gain from SU transmitter i to the TV receiver Pi: the tx power of SU transmitter ξ TV : the interference at PU

9. System Model (cont’d) 9  Received SINR of Secondary User (receiver i) For reliable communication, Let G i,j be the link gain from the secondary transmitter j to the secondary receiver i, G i,TV be the link gain from the secondary transmitter j to the secondary receiver I, P TV be the transmission power of the TV transmitter, N 0 be the receiver noise power. Receiver capture threshold Interference

10. Related Work – DCPC 10  Distributed Constrained Power Control [10]  Goal: keep the transmission power of the users in the network at the minimum level required to achieve the SINR for reliable communication. N x N Matrix H N x 1 Vector U [10] S. A. Grandhi, J. Zander, and R. Yates, “Constrained Power Control,” Wireless Pers. Commun., vol. 1,Dec. 1994.

11. Related Work – DCPC (cont’d) 11  The equation can be rewritten as,   The matrix notation of the linear inequality   P is the transmission power vector P = (P 1,P 2,…,P N ) T  If the maximum eigenvalue of the matrix H is less than one, there exists a non-negative power vector P which satisfies the above equation.

12. Related Work – DCPC (cont’d) 12  The Pareto optimal power vector is,   To solve the equation distributively, a general iterative method was introduced in [9].  

13. Related Work – GDCPC (cont’d) 13  The drawback of DCPC is that  the transmission power of the user reaches the maximum transmission power even if the user cannot achieve the minimum required SINR  Generalized DCPC (GDCPC) [11]  Motivation: to complement the drawback of DCPC When the user cannot achieve the required SINR, it reduces the power to arbitrary level instead of necessarily using the maximum transmission power [11] F. Berggren, R. Jantti, and S. Kim, “A Generalized Algorithm for Constrained Power Control With Capability of Temporary Removal,” IEEE Trans. Veh. Commun., vol. 50, Nov. 2001.

14. Related Work – GDCPC (cont’d) 14  GDCPC can be represented as,   The Power value is chosen arbitrarily within the range of

15. Autonomous Distributed Power Control for Cognitive Radio Networks 15  Since DCPC and GDPC don’t consider the QoS requirement for the PU  the resulting transmission power of them may exceed the ITL of the primary receiver  By translating the interference constraint by the sum of all SUs’ transmission power to the individual constraint for each SU the QoS requirement for the PU can be easily guaranteed

16. Autonomous Distributed Power Control for Cognitive Radio Networks (cont’d) 16  Assumptions  Each secondary user can know the number of secondary users N by using ad-hoc routing protocol.  The primary receiver transmits the beacon including the information about the beacon power and the ITL of itself.  The SUs can know the link gain from itself to the TV receiver and the ITL of the TV receiver

17. Autonomous Distributed Power Control for Cognitive Radio Networks (cont’d) 17  Autonomous DCPC   Autonomous GDCPC  Upper bound power value

18. Simulations 18 ParameterValue N: number of transmitting-receiving pairs50 h x h (the SUs’ region)2000m x 2000m r: transmission range of the SU500m R: transmission range of the PU70km and-100dBm and 3dB : GDCPC arbitrary power level0mW α 1 (PU) and α 2 (SU) : Path loss factor 3 and 4 and100mW and 100kW N 0 : Noise Power10 -11 mW

19. Simulations (cont’d) 19 Comparison of the interference temperature at the TV receiver

20. Simulations (cont’d) 20 Comparison of the number of the supported users

21. Simulations (cont’d) 21  The secondary transmitter nearest to the TV receiver is the worst interferer with the TV receiver  The secondary transmitter satisfies Eq.(18), all secondary transmitters in the network satisfy that.  The distance d satisfies the inequality(19), the proposed schemes are always equal to the conventional schemes.

22. Conclusions 22  Consider that the distributed power control problem in the CR network which shares the primary user’s channel  The transmission power constraint of each secondary user to protect the primary user is numerically derived.  Future work  The convergence rate of he proposed scheme using iterative method

23. Comments 23  Make an improvement of others related work such as DCPC and GDCPC.  Problem Definition  Model Formulation  Scenario  Related method  Improvement  Simulation & Comparison.