Hopfield Nearual Networks And Its Applications In Communication Systems

Artificial neural network is an abstraction and simulation of the basic characteristics of the human brain; and also it is a kind of imitating the behavior characteristics of Animal neural networks for distributed parallel algorithm which is for mathematical model of information processing. Recently, Artificial neural networks are deeply studied and widely used in combinatorial optimization problems and a lot of successful application instances and good results are reported.With the improvement of users'requirements, traditional OSI architecture is not suitable to wireless network with the development of wireless communication systems. So Cross-layer design is proposed, whose main content is that the protocol stack can realize self-adaptive optimization of resources allocation according to the changes of wireless environment through transmitting specific information between the layers of protocol stack, in order to utilize wireless networks resources effectively and improve the performance of the system.Multi-user OFDM will be one of the key technologies in the next generation of wireless communication systems. In order to fully exploit the advantages of OFDM in cellular systems, resource allocation techniques, which efficiently use the resources such as bandwidth, power, and modulation to increase the spectral efficiency of the system need to be devised. There are two kinds of dynamic resource allocation schemes existing in current multi-user OFDM system:margin adaptive (MA) and rate adaptive (RA).In this paper, these two kinds of dynamic resource allocation schemes have been optimized by Hopfield neural network(HNN). We utilized HNN's characteristics such as parallel processing, fast convergence speed and easy convergence to the optimum to solve the problems:(1) satisfying system performances and users'requirements for MA Optimization in multi-user OFDM communication system, (2) under the conditions of the signal-to-interference-plus-noise ratio (SINR) constraint, power constraint and time delay constraint for cross-layer dynamical resource allocation in Orthogonal Frequency Division Multiple Access (OFDMA)-based Wireless Mesh Networks (WMN), (3) under the conditions of proportional fairness, satisfying system performances and users' requirements for adaptive cross-layer resource allocation problem with the fairness in multi-user OFDM-MISO communication systems. (2) and (3) are the RA optimization problem. The method is simple in the computation by dividing the bit-loading matrix into three matrixes. The simulation results show that HNN can effectively solve optimization problems of resource allocation in such system, and it is more effective than the selected traditional method.