DNN-VTRM Channel Estimation for OFDM Underwater Acoustic Communication System
Nonlinear factors can adversely affect underwater acoustic (UWA) channel estimation, and the multipath and large delay extension of the UWA channel still causes residual inter-symbol interference (ISI) for orthogonal frequency division multiplexing (OFDM) underwater acoustic communication systems. I...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10804150/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Nonlinear factors can adversely affect underwater acoustic (UWA) channel estimation, and the multipath and large delay extension of the UWA channel still causes residual inter-symbol interference (ISI) for orthogonal frequency division multiplexing (OFDM) underwater acoustic communication systems. In this study, the Deep Neural Network (DNN) and Virtual Time Reversal Mirror (VTRM) technologies are jointly used as channel estimation and equalization algorithms in the OFDM underwater acoustic communication system. VTRM technology is used to solve the issues of OFDM system performance degradation caused by the residual ISI. The DNN network is used to solve the nonlinear factor from the UWA channel and the floor effect of the LFM matched filtering algorithm in the channel estimation part of the VTRM, so that it can complete the UWA channel mapping better, including sparsity, large delay expansion, and multipath. First, the DNN time-domain channel estimator is trained using the received random probe signal and sparse UWA multipath channel impulse responses generated by Bellhop. The output value of the DNN channel estimator is then processed by the filtering algorithm to restore the sparsity of the channel, which prepares as the UWA channel impulse response. When the received signal is preprocessed by VTRM equalization, the focusing principle of time reversal can be used to cancel the distortion caused by the UWA sparse multipath channel. Finally, this study uses the BELLHOP model to simulate the UWA test channel to compare the DNN-VTRM with traditional channel estimation methods such as the Least Square (LS) algorithm, the Linear Minimum Mean Square Error (LMMSE) algorithm, the Orthogonal Matching Pursuit (OMP) algorithm, and the DNN frequency domain channel estimator. The experimental results show that the estimated effect of the DNN time-domain channel estimator is more accurate than that of the LFM matched filtering and OMP algorithms. Under the low signal-to-noise ratio (SNR) condition, the OFDM underwater acoustic communication system based on DNN-VTRM has a better bit error rate (BER) performance than the traditional algorithm. |
|---|---|
| ISSN: | 2169-3536 |