Heuristic Algorithm Based Adaptive Resource Allocation Technique For Cognitive Radio

As the rapid development of wireless radio communication technology, theshortage of spectrum resource is gradually obvious, because traditional staticspectrum allocation method has caused a lot of waste of spectrum. The advance ofcognitive radio technology can effectively enhance the spectrum utilization, andsolve the unbalance in spectrum utilization. As the key technology of cognitive radio,the adaptive resource allocation technique, based on the channel property and thespectrum demand, can adaptively allocate the idle spectrum to second users,enhance the spectrum utilization and improve the system performance.Among the current studies about the adaptive resource allocation in thetraditional cognitive radio, most of the work are based on the single primarynetwork. To ample the result of adaptive resource allocation, this paper discusses theadaptive resource allocation for the cognitive system with multiple primarynetworks solve the problems caused by single primary networks. The highlight ofthe paper are as following:The carrier aggregation technology used by LTE-Advanced is introduced intothe multiple primary networks. Under cognitive radio network with multiple primarynetworks environment, adaptive resource allocation is concluded as a mixed integralnon-linear programming problem. Because the problem is a non-deterministicpolynomial problem, a series of simplifications are conduced, and the heuristicoptimization algorithms are introduced to pursue the solution.Kinds of heuristic optimization algorithms are introduced to conduct theadaptive resource allocation, and the features of different stages of the algorithm areconcluded. Based on the features of different stages of adaptive resource allocationunder multiple primary networks, to enhance the convergence speed and improvethe system performance, a synthesis algorithm is proposed.A variable fitness program is introduced into adaptive resource allocation undermultiple primary networks; with the different fitness in different stages, the systemperformance can be controlled conveniently, and the flexibility can be enhancedgreatly.