Study On Several Computational Intelligent Multiuser Detection Algorithms In DS-CDMA System

Multi-user detection technology is one of the effective ways to solve the problem of multiple access interference in the field of wireless communications. The principle is to make full use of the information such as the size of all users, the user code and time delay which caused the multiple access interference to reduce negative impact brought by the incompletely orthogonal of spread spectrum code. It can greatly improve the system performance.In this paper, we study the optimal and suboptimal multi-user detection technology.And it discusses and analyzes the traditional detection, de-correlation multi-user detection and MMSE multi-user detection in detail. The simulation results verify that the multi-user detection has the characteristics of soft capacity, and the multiple access interference and near-far effect resistance of the detection is stronger.In order to solve the problems of poor robustness and channel tracking ability about the non-standard constrained constant modulus algorithm blind multi-user detection algorithm, we define a Rayleigh distribution variable-step size for the step function formula, and this new algorithm combines the new step function formula and differential constant modulus which we just need to know the amplitude differences of two adjacent signals. It can clevely avoid the search for target amplitude. On this foundation, we propose a non-standard constrained differential constant modulus algorithm which based on Rayleigh distribution variable-step. The simulation results show that the abilities of resisting multiple access interference, the near-far effect and channel tracking of RDV-NSCDCMA are superior to NSCCMA even in the conditions of low signal to noise ratio and strong multi-access interference.In order to solve the problems of poor searching precision and poor ability to keep the development and quest for balance about the adaptive artificial fish swarm algorithm, we firstly use the sub-optimal solution or the mutation operation result of MMSE multi-user detection as the initial value of every artificial fish. Secondly, it uses the objective function and constraint condition as antigen, and the candidate solution as antibodies. Tirdly, it clones high affinity antibodies according to the proportional affinity. Fourthly, the adaptive mutation operator is introduced, and it mutates with the method of inverse ratio affinity. For the sake of antibody diversity, the algorithm reinitializes the low affinity antibodies by certain proportion. On this foundation, we propose an improved artificial fish swarm algorithm based on adaptive clonal selection and mutation, and it’s a multi-user detection algorithm based on minimum mean square error and improved artificial fish swarm algorithm. The simulation results show that the abilities of resisting multiple access interference and the near-far effect, the rate of convergence of ACSM-IAFSA are superior to the other two artificial fish swarm algorithms. And it has better capacilities to maintain the population diversity, maintain the development and explore the balance.