Scholarlink

By splitting the signal into several narrowband subcarriers, Orthogonal Frequency Division Multiplexing (OFDM) reduces inter-symbol interference (ISI) and efficiently counteracts channel impairments in wireless communication systems. Nevertheless, OFDM still needs equalization techniques to improve performance because it is vulnerable to noise and channel fading. Because it takes noise variance into account and reduces Bit Error Rate (BER), the Minimum Mean Square Error (MMSE) performs better than Zero Forcing (ZF) among equalizers, especially in situations with low Signal-to-Noise Ratio (SNR). In order to reduce channel distortions, we investigated a number of equalization strategies in addition to channel estimation (CE). MMSE showed the best performance. In order to enhance BER prediction even more, we integrated an LSTM network. Improved error prediction was made possible by LSTM’s acquisition of the nonlinear relationship between SNR and BER by training on SNR-BER data. For adaptive communication systems, the MMSE, LSTM method offers a reliable solution. In summary, OFDM guarantees spectral efficiency, while MMSE is the best equalizer for noisy wireless networks. In contemporary wireless networks, LSTM improves system adaptability and is a potent instrument for real-time performance enhancement.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Intelligent Wireless Communication System Optimization Using OFDM, Equalization, and Deep Learning Techniques

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Ahlaam Miftah Saed,
  • Ezuldeen Saeid,
  • Al-Mutasim Abreek,
  • Ibtihal Fawzi Elshami,
  • Asma Ali Budalal,
  • Rabia Al-Mamlook

摘要

By splitting the signal into several narrowband subcarriers, Orthogonal Frequency Division Multiplexing (OFDM) reduces inter-symbol interference (ISI) and efficiently counteracts channel impairments in wireless communication systems. Nevertheless, OFDM still needs equalization techniques to improve performance because it is vulnerable to noise and channel fading. Because it takes noise variance into account and reduces Bit Error Rate (BER), the Minimum Mean Square Error (MMSE) performs better than Zero Forcing (ZF) among equalizers, especially in situations with low Signal-to-Noise Ratio (SNR). In order to reduce channel distortions, we investigated a number of equalization strategies in addition to channel estimation (CE). MMSE showed the best performance. In order to enhance BER prediction even more, we integrated an LSTM network. Improved error prediction was made possible by LSTM’s acquisition of the nonlinear relationship between SNR and BER by training on SNR-BER data. For adaptive communication systems, the MMSE, LSTM method offers a reliable solution. In summary, OFDM guarantees spectral efficiency, while MMSE is the best equalizer for noisy wireless networks. In contemporary wireless networks, LSTM improves system adaptability and is a potent instrument for real-time performance enhancement.