Scholarlink

<p>In today’s world the MIMO based system provides the minimum fading effects for the information travels from transmit end to receive end. Therefore, filter-bank multi-carrier with offset quadrature amplitude modulation (FBMC/OQAM) exhibits superior spectrograph shape and improves mobility support in MIMO Comparing wireless technology to OFDM. However, the traditional Channel Estimation (CE) techniques cannot be used because of the inherent imaginary interference. do not achieve high CE performance. To address these limitations, this paper proposes a novel deep learning-based channel estimation framework that integrates Long Short-Term Memory (LSTM) networks with Ant Colony Optimization (ACO) for hyperparameter tuning. The LSTM architecture is specifically designed to overcome the vanishing gradient problem inherent in traditional RNNs, while ACO enhances model convergence and generalization. Additionally, a hybrid spectral preprocessing technique combining Fast Fourier Transform (FFT) and Short Chirp Z-Transform (SCZT) is introduced to reduce computational complexity and improve spectral resolution. The proposed LSTM-ACO model is implemented using Python and evaluated under sparse pilot configurations. Comparative analysis with conventional CE methods such as Least Squares (LS) and Linear MMSE (LMMSE) demonstrates significant improvements in Bit Error Rate (BER) and throughput. Simulation results confirm the robustness of the proposed approach in doubly-selective and high-speed channel conditions, establishing its potential for deployment in 5G and beyond wireless systems.</p>

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

Evaluation of channel estimation for MIMO-FBMC/OQAM system based on deep learning approach (LSTM-ACO technique)

  • 期刊论文

  • 发表日期：

    2025年11月17日

查看全文
  • T. Padmavathi,
  • Ramarakula Madhu,
  • Kusma Kumari Cheepurupalli

摘要

In today’s world the MIMO based system provides the minimum fading effects for the information travels from transmit end to receive end. Therefore, filter-bank multi-carrier with offset quadrature amplitude modulation (FBMC/OQAM) exhibits superior spectrograph shape and improves mobility support in MIMO Comparing wireless technology to OFDM. However, the traditional Channel Estimation (CE) techniques cannot be used because of the inherent imaginary interference. do not achieve high CE performance. To address these limitations, this paper proposes a novel deep learning-based channel estimation framework that integrates Long Short-Term Memory (LSTM) networks with Ant Colony Optimization (ACO) for hyperparameter tuning. The LSTM architecture is specifically designed to overcome the vanishing gradient problem inherent in traditional RNNs, while ACO enhances model convergence and generalization. Additionally, a hybrid spectral preprocessing technique combining Fast Fourier Transform (FFT) and Short Chirp Z-Transform (SCZT) is introduced to reduce computational complexity and improve spectral resolution. The proposed LSTM-ACO model is implemented using Python and evaluated under sparse pilot configurations. Comparative analysis with conventional CE methods such as Least Squares (LS) and Linear MMSE (LMMSE) demonstrates significant improvements in Bit Error Rate (BER) and throughput. Simulation results confirm the robustness of the proposed approach in doubly-selective and high-speed channel conditions, establishing its potential for deployment in 5G and beyond wireless systems.