Heuristic algorithms for scheduling in centralized cognitive radio networks

Abstract

In this thesis we propose heuristic scheduling algorithms for the centralized, time slotted cognitive radio network (CRN) model in [1]. In the considered CRN scheme, a frequency f and time slot t pair constitutes a resource r and a centralized cognitive base station (CBS) coordinates the assignment of the resources to the secondary users (SUs) possibly with multiple antennas. The proposed algorithms di er in their problem objectives, but they all make sure that none of the primary users in the service area of the CBS is disturbed, no collisions occur among the SUs and each SU gets at least one time slot per scheduling period. For the throughput maximizing scheduling (TMS) problem, an auction theory based algorithm is proposed whereas Best First Resource Assignment (BFRA) and Resource Assignment With Partial Backtracking (RAPB) algorithms are proposed for the maximizing the number of satis ed users (MNSU) problem whose objective is to increase the number of satis ed SUs where an SU is satis ed only if its minimum data rate requirement is met.