Using Back Propagation Neural Network for Channel Estimation and Compensation in OFDM Systems

Abstract

In orthogonal frequency division multiplexing (OFDM) communication systems, due to the environmental impact generated the multipath effect caucused signals distortion and attenuation in transmitted process, and due to relative motion between transmitter and receiver caused the Doppler Effect that makes the signal carrier offset. Therefore, the knowledge of the channel characteristics is very important. To remove the effect from received signal, the receiver needs to have knowledge of channel impulse response (CIR) by channel estimation, and then compensates signals. In this paper, a back propagation neural network (BPNN) is used to estimate channel and compensate signals. Our proposed BPNN channel estimation would compare bit error rate (BER) and mean square error (MSE) with least square (LS) and minimum mean square error (MMSE) algorithms in an existing OFDM channel environment. From the results, our proposed algorithm has better performance than LS algorithm and closes to MMSE algorithm.

Artificial intelligence for channel estimation in multicarrier systems for B5G/6G communications: a survey

“…The BPNN performance has been assessed in terms of BER and MSE compared to other conventional channel estimation approaches. Concerning the MMSE, LS, and LMS methods, the BPNN has underperformed the former while it has outperformed the others [231,232,236].…”

“…Complex-valued BPNNs have been designed based on three layers (input, hidden, and output) of NN for channel estimation purposes, as shown in Fig. 11 [231,232]. The complex signal is decomposed into real and imaginary parts to feed-forward the network.…”

“…The BPNN in Fig. 11 has been used for channel estimation and equalization in FBMC [233] and OFDM systems [231,232,[234][235][236][237][238], employing supervised learning through training sequences. The number of used perceptrons is specific, according to the proposal.…”

Channel estimation using an adaptive neuro fuzzy inference system in the OFDM-IDMA system

“…Although no work has been performed related to channel estimation using the adaptive neuro fuzzy inference system (ANFIS) for OFDM-IDMA system, there are some studies in which the ANFIS and the other heuristic methods are applied to estimate channel frequency responses for single-input single-output (SISO) and multiple-input multiple-output (MIMO) OFDM systems [15][16][17][18][19]. One study was conducted to estimate channel frequency responses in the OFDM-IDMA system by using a radial basis function neural network (RBFNN), which is one of the heuristic methods used for channel estimation [20].…”

“…In [15], an estimator based on a backpropagation neural network with three layers was proposed for SISO-OFDM transmission technology and the system performance was evaluated by using bit error rate (BER) and mean square error (MSE) graphs. In [16] and [17], an ANFIS-based channel estimator was offered for MIMO and SISO OFDM, respectively, and a comparison was made with conventional estimation algorithms such as LS and MMSE with regards to performance and complexity.…”