Resource Optimization Schemes Based On Metaheuristic Algorithms In Wireless Communication Systems

3GPP Long Term Evolution (LTE) is commonly considered as the most mature communication technology for the next generation communication system. The large-scale technical innovation in LTE has exhausted most of the new signal processing techniques accumulated in the past decades, such as OFDM, MIMO, adaptive technique and so on. Moreover, the introduction of new technologies also puts forward the great challenges to the radio resource management and optimization. Therefore, the research on the future communication system will more focus on the resource management technologies and the network layer optimization aspects. This thesis mainly studies on the radio resource management and optimization of the next generation wireless communication system, which aims at achieving the objective of improving spectrum efficiency, and enhancing users’ performance and service quality of the system. The main research contents of this thesis are summarized as follows:Considering downlink discrete power allocation problem in multi-cell OFDM systems, an ant colony optimization (ACO) algorithm based solution is proposed. Via transmitting signal on discrete power levels, multi-cell power allocation under per-base-station power constraint is modeled as a combinatorial optimization problem, which was proved to be NP-hard. ACO is applied to get near-optimal solution of the problem. In particular, a multiple power level based searching graph is designed, and conventional ACO is improved that ant colonies in each cell cooperate to maximize system throughput. Simulation results indicate the best number of power levels, and prove that the proposed scheme achieves a significant rate gain over the existing schemes. Considering scheduling and resource allocation probelm with adaptive modulation in downlink coordinated multi-point transmission (CoMP) systems, a genetic algorithm based solution is proposed. The joint scheduling and resource allocation to maximize the system throughput is formulated as a combinational optimization problem under constraints. In particular, a two-dimension-binary chromosome coding scheme is designed to denote potential user selection and bit loading strategies across multiple subchannels. To handle per-base-station power constraint, a penalty based fitness function is used, and in doing so GA can search for optimal solution in a larger region. To ensure convergence, a super individual named elite is added to each population. Simulation results indicate that the proposed algorithm provides close to ES performance but with much lower computational complexity.A new resource dimension-antenna/antenna array is added to the resource pooling. To solve the multiple-dimension resource allocation problem, a genetic based resource allocation strategy is proposed. The new resource pooling concept includes power, frequency, antenna/antenna array, codeword and so on. Resource allocation based on resource pooling can be modeled as a complex combinational problem under constraints. And two kinds of resource allocation criteria are proposed. In particular, an integer coding strategy is proposed in the scheme. Simulation results prove that the scheme can achieve significant performance gain compared to existing schemes under both criteria. Furthermore, the research result is extended that a frameless network structure is proposed.A summary is given at the end, and the future research directions related to this master’s thesis are also given.