Optimization on Burst Allocation in D2DWRAN

Abstract

With rapid development of digital devices with wireless services, communication networks play a significant role in our daily lives. The explosive data demand makes it urgent to extend the capacity of the existing networks. Cognitive Radio (CR), which can sense and use the frequency band dynamically, can provide an efficient radio frequency usage. IEEE 802.22 working group has developed several standards based on CR technology leading to Wireless Regional Area Networks (WRANs). However, WRANs adopt the cellular topology in which all communication is in a point-to-multi-point (P2M) manner. This P2M communication limits the network capacity because all messages need to be routed by the Base Station (BS) even for intra-cell communication. Therefore, Device-to-Device (D2D) WRAN (D2DWRAN) has been proposed, which employs D2D communication into the WRAN. A burst allocation is the allocation of resources for wireless communication to the users, and influences the network capacity significantly. With D2D communication, the burst allocation problem in D2DWRAN is different from it in WRAN. This thesis attempts to solve this new burst allocation problem. After introducing the progression from WRAN to D2DWRAN, the new burst allocation problem is stated and formulated mathematically in different scenarios. Then the existing optimization methods are studied and evaluated, such as Greedy Algorithm, Simulated Annealing Algorithm, Genetic Algorithm and Ant Colony Algorithm. Some modifications based on these algorithms are also proposed to solve the burst allocation problem in D2DWRAN. These modified algorithms are also simulated and the advantages and disadvantages of each of them are discussed and analysed.