Research On Adaptive Resource Allocation For Ofdma System Based On Intellight Optimization Algorithms

OFDMA is a wireless access method based on OFDM, its'resource allocation mechanism is so flexible that it can allocate the sub-carrier dynamically by the user'businesses. Nowadays, OFDMA has been the key technique of wireless communication system. What's more, the concern of resource allocation is growing. So, some algorithms based on adaptive resource allocation have generated successively, such as the sub-carriers allocation algorithms based on the classical greedy algorithms, utility function, genetic algorithms and some other cross-layer resource allocation algorithms. Compare with the current algorithms, these algorithms have improved the resource allocation algorithm's performances, and get achievement to some extent, but still have problems of large amount of calculation, practice hardly, low spectrum utilization. Therefore, the resource allocation algorithms based on OFDMA is still the hot issue. On the other hand, the resource allocation algorithms based on intelligence algorithms is one of the most important trends.In order to further improve the availability of frequency spectrum, the system capacity and so on, after studying the adaptive resource allocation algorithms deeply, this paper applied the particle swarm optimization (PSO) algorithm and the adaptive genetic algorithm into adaptive resource allocation algorithms based on physical layer and cross-layer.First of all, this paper proposed PSO algorithm in OFDMA system to allocate the adaptive sub-carrier. The PSO has the characteristics of fewer parameters, rapid speed of convergence and strong capability of global search, so, made it being a better solution in physical layer to allocate sub-carrier. In order to reduce the total transmitting power, in the process of resource allocation, considering the fairness among the users, this algorithm allocated the sub-carriers to one user by the number of his sending bits, and found the best sub-carriers program according to every sub-carrier's fading condition. According to the experimental and simulation results, it was proved that, the performance had been improved when using the PSO algorithm to allocate the adaptive sub-carrier, in the case of minimizing the overall transmit power and being more rapidly while needing the fulfillment of each user's bit-rate and bit error rate (BER) requirement.Secondly, for the shortcoming of the cross-layer resource allocation algorithms, this paper proposed an adaptive genetic algorithm based on cross-layer, put the sum of all the users'sent bits as the utility function. The adaptive genetic algorithm has the characteristics of population diversity and comprehensive ability to global search, so, made it being a better solution to the cross-layer resource allocation. Being content with the fairness among users, the algorithm generated the excellent sub-carriers program depending on the information of state channel and the user's queue states to allocate some sub-carriers for every user. At the same time, the algorithm made full use of the adaptive cross probability and mutate probability to find the best solution. According to the experimental and simulation results, it was proved that, the improved algorithm in this paper enhanced the availability of frequency spectrum and throughput maximumly, was also shorter the user's average wait delay, satisfied the user's quality of service.