Improved Whale Optimized MLP Neural Network-Based Learning Mechanism for Multiuser Detection in MIMO Communication System

Abstract

Direct-Sequence Code Division Multiple Access (DS-CDMA) is a digital method to spread spectrum modulation for digital signal transmission. We propose to detect signal in DS-CDMA communication using the learning mechanism. Initially, the user signals are spread using the respective pseudo-noise (PN) code where the input signal is multiplied with the code which is then modulated using the quadrature phase shift keying (QPSK) modulator. The modulated signal is then transmitted in a 3G/4G channel considering all types of fading. The transmitted signal is received by the antenna array which is performed by demodulation. We propose to adaptively assign the weights by employing Improved Whale Optimized Multi-Layer Perceptron Neural Network (IWMLP-NN)-based learning mechanism. To design IWMLP-NN, Improved Whale Optimization Algorithm is combined with multilayer perceptron neural network. This is used instead of the normal Multiple Signal Classification (MUSIC) and least mean squares (LMS)/root-mean-square (RMS) algorithms used in beam-forming networks. After assigning weight through IWMLP-NN-based learning mechanism, we de-spread to get the original user data. We have compared our proposed technique with the normal techniques with the help of plots of Bit Error Rate (BER) versus Signal-to-Noise Ratio (SNR). We use both the AWGN channel and fading channel for analysis. Experimental results prove that our proposed method achieves better BER performance results even with deep fading.