Robust Scheduling and Power Control for Vertical Spectrum Sharing in STDMA Wireless Networks

• Published in: IEEE transactions on wireless communications Vol. 11; no. 5; pp. 1850 - 1860
• Main Authors: Phunchongharn, P., Hossain, E., Long Bao Le, Camorlinga, S.
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01-05-2012; Institute of Electrical and Electronics Engineers
• Subjects: Applied sciences; cognitive radio; Exact sciences and technology; Heuristic algorithms; Interference; Power control; Pricing; Radiocommunication specific techniques; Radiocommunications; robust optimization; Robustness; scheduling; spatial reuse time-division multiple access (STDMA); Systems, networks and services of telecommunications; Telecommunications; Telecommunications and information theory; Transmission and modulation (techniques and equipments); Uncertainty; Vertical spectrum sharing; Wireless networks; Wireless telecommunication; robust optimization; Iterative method; Scheduling; Resource sharing; Perturbation; Power allocation; Penalty method; Optimization; Time division multiple access; cognitive radio; Algorithm performance; Simulation; Power control; Vertical spectrum sharing; Pricing; Heuristic method; Greedy algorithm; Wireless network; spatial reuse time-division multiple access (STDMA); Optimal planning; Tariffication; Numerical convergence; Software radio
• ISSN: 1536-1276; 1558-2248
• DOI: 10.1109/TWC.2012.030812.111341

We study the robust transmission scheduling and power control problem for spectrum sharing between secondary and primary users in a spatial reuse time-division multiple access (STDMA) network. The objective is to find a robust minimum-length schedule for secondary users (in terms of time slots) subject to the interference constraints for primary users and the traffic demand of secondary users. We consider the fact that power allocation based on average (or estimated) link gains can be improper since actual link gains can be different from the average link gains. Therefore, transmission of the secondary links may fail and require more time slots. We also consider this demand uncertainty arising from channel gain uncertainty. We propose a column generation-based algorithm to solve the scheduling and power control problem for secondary users. The column generation method breaks the problem down to a restricted master problem and a pricing problem. However, the classical column generation method can have convergence problem due to primal degeneracy. We propose an improved column generation algorithm to stabilize and accelerate the column generation procedure by using the perturbation and exact penalty methods. Furthermore, we propose an efficient heuristic algorithm for the pricing problem based on a greedy algorithm. For the simulation scenario considered in this paper, the proposed stabilized column generation algorithm can obtain the optimal schedules with 18.85% reduction of the number of iterations and 0.29% reduction of the number of time slots. Also, the heuristic algorithm can achieve the optimality with 0.39% of cost penalty but 1.67×10 -4 times reduction of runtime.