The Application In Frequency Assignment By Effective Optimization Arithmetics

In these years,with quick step of mobile communication networks building,The request with networks'operation is asked by people more and more.So the capability of networks is needing to enlarge. But now in this instance the available frequency in cellular mobile communication network is limited, it is imposible to enlarge infinitily.So how to use effectively the available frequency in order to meet increased clients'needs is the key problem,which depends on acquiring a kind of the better frequency assignment and frequency planning method.Frequency assignment is just channel assignment, which is a type of nonlinear combinatorial optimization problem. In cellular mobile communication networks,the reasonable frequency assignment scheme is that allocating the limited frequencies to the cells in the system on the premise of no interference between them. In the GSM networks, the interference between cells primarily is co-channel,adjacent channel and co-site interference. So regarding these interference as the inhibitable condition of the goal funcition, the ultimate aim of this thesis'investigation is to find the best solution of the goal funcition by computing using centain optimization arithmetic.Two optimization arithmetics are used to deal with frequency assignment in this article. One is a modified genetic algorithm, the other is artificial fish swarm algorithm. Firstly, the genetic algorithm is a very classical optimization algorithm, which ever acquired greatly application effect in many areas. A modified genetic algorithm is introduced in this article on the base of genetic algorithm and combining with monkey-king genetic algorithm which is introduced by Guo Chenghai and Xie Jun. This algorithm's characteristic is setting the biggest fitness value in every age as monkey king, and then producing the new individuals by the way of crossover. The simulation result shows that the modified approach effectively deals with the problem of frequency assignment. It not only can improve convergence rating,have a better convergence speed but also can keep diversity of individuals.The artificial fish swarm algorithm is a new bionic optimization algorithm, which simulates fishes prey. This algorithm's theory is easy to realize and has a few parameters. It brings more and more attention by people. The artificial fish swarm algorithm is applied to deal with frequency assignment in this thesis for the first time. The result shows that this algorithm has definite feasibility and superiority. When the practicable frequency is abundance, this algorithm has a high convergence speed. But when the practicable frequency is reduced, the result is not satisfactional. Because the artificial fish swarm algorithm is on the junior research phase, it can improve greatly on the frequency assignment problem.