Signal detection of massive MIMO systems using LSTM-based signal detectors for beyond 5G radio
Massive multiple inputs and multiple outputs (M-MIMO) are expected to play a vital role in radio systems beyond fifth-generation (B5G) by enhancing the throughput and spectral access of the framework. The use of multiple antennas in MIMO upsurges the complexity of accurate signal detection and degra...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-08-01
|
| Series: | Alexandria Engineering Journal |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1110016825007926 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Massive multiple inputs and multiple outputs (M-MIMO) are expected to play a vital role in radio systems beyond fifth-generation (B5G) by enhancing the throughput and spectral access of the framework. The use of multiple antennas in MIMO upsurges the complexity of accurate signal detection and degrades the overall system’s performance. The conventional zero force equalizer (ZFE), Maximum likelihood detection (MLD), Minimum mean square error (MMSE), and so on upsurge the complexity while detecting the signal. In this article, we proposed a signal detection algorithm based on a long short-term memory (LSTM) neural network known as MLD-LSTM, MMSE-LSTM, and ZFE-LSTM for MIMO frameworks (16 X 16, 256 X 256, and 512 X 512). The parameters include bit error rate (BER), power spectral density (PSD), and complexity of conventional MLD, MMSE, and ZFE methods. The simulation results reveal that the proposed methods effectively enhance the BER and PSD performance at low complexity. The proposed methods obtain the BER and PSD improvement of 30% to 45% as compared with the contemporary methods. |
|---|---|
| ISSN: | 1110-0168 |