Adaptive blind equalization of fast time‐varying channel with frequency estimation in impulsive noise environment

Abstract

In this paper, a novel source signal recovery method for fast time-varying channels described by complex exponential-basis expansion model (CE-BEM) in the impulsive noise environment is proposed. This method consists of two phases. The first phase is the equalization of fast time-varying channels, in this phase, a novel algorithm FSE-FLOS-CMA is proposed. The convergence performance of this newly proposed algorithm is much better than that of existing fractionally spaced equalizer-constant modulus algorithm (FSE-CMA) in impulsive noise environment. In the second phase, a novel frequency estimation method is proposed. The estimated frequency in impulsive noise environment is obtained by calculating a specific p-order fractional low-order cyclic moment of the equalized signal. Simulation results show that the proposed FSE-FLOS-CMA and frequency estimation method can effectively estimate the source signals transmitted through the fast time-varying channels in impulsive noise environment.

Trigonometric Coordinate Transformation Blind Equalization Algorithm Based on Bi-Direction Long and Short-Term Memory Neural Networks

“…The number of neurons in the input layer of the BLSTM neural network is the tap length of the transverse filter, and the weight vector in the network is adaptively adjusted through the loss function so that the network output sequence is gradually close to the original transmission sequence [15], [16], [17], [18]. BLSTM neural network hidden layer unit structure, as shown in FIGURE 2.…”

“…Although the enhanced coordinate transformation constant modulus blind equalization algorithm can be applied to MQAM and MPAM signals, the loss function term needs to be adjusted as the signal order increases. The expression of the loss function becomes more complex with higher order, making it less universally applicable and requiring more computational resources [17], [21]. In this paper, a constant modulus blind equalization algorithm based on triangular coordinate transformation is proposed for MQAM and MPAM signals.…”

“…When the transmitted signal is a 64QAM modulated signal, the number of taps of the CMA blind equalizer is 17, and the iteration step is 0.0000008. The network structure of SCA-BP-CMA is (17,13,1), and the learning rate of the neural network is µ SCA−BP = 0.0005. The network structure of TAF-CBP-CMA is (17,9,1), and the learning rate of the neural network is µ TAF−CBP = 0.0065.…”

“…The network structure of SCA-BP-CMA is (17,13,1), and the learning rate of the neural network is µ SCA−BP = 0.0005. The network structure of TAF-CBP-CMA is (17,9,1), and the learning rate of the neural network is µ TAF−CBP = 0.0065. The network structure of BLSTM-TCT-CMA is (17,29,29,1), the transfer constant is β = 12, and the learning rate of the neural network is µ BLSTM −TCT = 0.00015.…”

“…The network structure of TAF-CBP-CMA is (17,9,1), and the learning rate of the neural network is µ TAF−CBP = 0.0065. The network structure of BLSTM-TCT-CMA is (17,29,29,1), the transfer constant is β = 12, and the learning rate of the neural network is µ BLSTM −TCT = 0.00015. Monte Carlo simulation experiments were performed 500 times, and the results are shown in FIGURES 9 and 10.…”