Reduction of Large Scale Linear Dynamic MIMO Systems Using Adaptive Network Based Fuzzy Inference System

Large Scale Multiple Input Multiple Output (MIMO) technology is a promising technology in wireless communications, and it is already at the heart of many wireless standards. MIMO technologies provide significant performance improvements in terms of data transfer rate and reduction the interference. However, MIMO techniques face large-scale linear dynamic problems such as system stability and it will be possible to overcome this problem by tuning the proportional integral derivative (PID) in continuous systems. The aim of this paper is to design an efficient model for MIMO based on Adaptive Neural Inference System (ANFIS) controller and compare it with a traditional PID controller. and evaluated by objective function as integral time absolute error (ITAE). ANFIS is used to train fuzzy logic systems according to the hybrid learning algorithm. The training involves the fuzzy logic parameters through simulating the validation data to represent a model to know the correctness and effectiveness of the system. It is optimizes the system performance in real time, however, to avoid potential problems such as easy local optimality. In the proposed approach stability is guaranteed as the initial steady-state scheme. ITAE is combined with ANFIS to minimize the steady-state transient time responses between the high-order initial pattern and unit amplitude response. The proposed ANFIS self-tuning controller is evaluated by comparing with the conventional PID. MATLAB simulink is used to illustrate the results and demonstrate the possibility of adopting ANFIS controller. The simulation results showed that the performance of ANFIS controller is better than the PID controller in terms of settling time, undershoot and overshoot time.