Research On AP Optimize Of WLAN Planning Based On PSO Algorithm

In recent years, WLAN has been a rapid development and has become one of the main broadband access techniques. With the WLAN network expansion and the number of users increase, large-scale WLAN network planning becomes more and more important. If the WLAN network planning unreasonable, easily lead to mutual interference between networks, reduce the user experience. Therefore, in recent years, the relevance technology of WLAN network planning becomes one of the hot studies.Each step of WLAN network planning is facing the enormous data and workload, which required spending substantial time and manpower. In this paper, we introduce particle swarm optimization algorithm to the field of WLAN network planning, for improve the efficiency and quality of Network planning, and expand the scope of application and development space at the same time.In this paper, we research the main WLAN technology, applications and development, network components and WLAN network topology, design a more rational AP optimization model of WLAN network planning. The model includes coverage and throughput, channel configuration and equipment costs, be able to better express WLAN network performance.We compare several common optimization algorithms, the results showed that the convergence speed of particle swarm optimization is much larger than other algorithms because of less parameter and easy implementation. Depending on in-depth study of PSO and its improved algorithm, we advance a improved PSO based on genetic strategy. The algorithm introduce crossover operator in iterative process of particles to keep variety of particles and enhance algorithm search capabilities. Comparison with the standard PSO has proved that improved algorithm has better optimization performance and faster convergence speed.We complete AP simulation optimization of WLAN network planning in Matlab environment with the improved PSO algorithm. And comparing with Prun+RCC algorithm, we proved that the optimal simulation of this article can obtain better network performance, which allowing users a more equitable access to system resources.