Research On Adaptive Resource Allocation And Optimization In Power Line Communication System

Power line communication technology is one of the core technologies forthe smart grid application. As power line network is widely distributed and easyto access, there is no need to lay extra lines. It arouses great attention in thearea of broadband access, remote automatic reading meter and smart homenetwork. Power line communication has positive development prospect. Butdue to the limitation of channel conditions, it is affected by channel noise andmultipath fading. So anti interference technology is used to combat theseeffects. Orthogonal Frequency Division Multiplexing technology is widely usedin power line communication system. Combining OFDM with adaptive resourceallocation technology can effectively improve the influence of frequencyselective fading to realize high speed power line communication, and improvethe bandwidth efficiency of the system and quality of service.The resource allocation problem was researched in this paper, includingsubcarriers, bits and power allocation. According to the power minimizationcriterion, to solve the resource allocation problem for multiuser OFDM system,fractional step method was adopted. The first step was for subcarrier allocationto decide the assigned subcarrier number for each user, and which carrier eachuser could use. The second step was for bit and power allocation.To solve the conflict of subcarriers allocation and high computationalcomplexity, on the basis of rate-craving greedy algorithm and amplitude-cravinggreedy algorithm, an improved subcarrier allocation scheme was proposed in thispaper. The algorithm adopted iteration allocation method, and allocated thesubcarrier to the user with the highest priority to make sure the total transmissionpower was minimal. The method reduced the computational complexity and the required transmission power of the system. Besides, the utilization ratio of thefrequency band was improved. In view of the traditional particle swarmoptimization alogrithm for bit and power allocation existing problems of slowconvergence and getting local optimum solution, improved particle swarmoptimization was proposed. In the process of particle swarm algorithm updating,the crossover operation of genetic algorithm was made to the position of theparticles, and the mutation operators of genetic algorithm was made to theparticles whose speed was below the threshold. In this way, the crossoveroperation increased the diversity of the particles and speeded up theconvergence speed. Through the mutation particles can jump out of localoptimal solution and improve the ability of global optimization. In the end,Theoretical analysis and simulation have been carried out to verify theproposed algorithm. Results showed that using new algorithm for resourceallocation can effectively solve the problem of premature convergence andimprove the signal-to-noise ratio characteristics of the system and reduce thetransmission power.