Scholarlink

<p>The swift advancement of Internet of Things (IoT) applications has intensified the demand for efficient wireless communication systems capable of handling high-rate data transmissions. This goal can be achieved with MIMO-NOMA systems. This paper introduces a channel estimation method for a Multiple-Input Multiple-Output (MIMO) system using Non-Orthogonal Multiple Access (NOMA) based on Vertical-Bell Laboratories Layered Space-Time Zero Forcing (V-BLAST ZF). Additionally, it presents a detection technique that employs a Long Short-Term Memory (LSTM) as a deep learning model to tackle the problem of incorrect signal detection, which arises due to an imperfect Channel State Information (CSI), user interference, and channel noise. The results indicate that a higher operational power can enhance the Signal-to-Noise Ratio (SNR) and channel estimation, while a lower power consumption is linked to energy-efficient designs. Spectral efficiency is enhanced when the number of antennas is increased. Nonetheless, this raises power consumption, leading to a compromise between power consumption and spectral efficiency. Moreover, when the number of antennas and bandwidth were increased, the Mean-Squared Error (MSE) in the proposed method decreased from about 0.17 to 0.0000425. Furthermore, with an operational gain of 800 Mbps per cell, the proposed method was about 2.4% and 3.7% better than the (Regularized Zero-Forcing) RZF and M-MMSE methods, respectively. Besides, the proposed method was respectively 6.3% and 33.33% better than the S-MMSE and MR methods in increasing energy efficiency.</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

The Performance Optimization of MIMO-NOMA Network with a Developed V-BLAST ZF Approach Based on LSTM Deep Learning in Internet of Things

  • 期刊论文

  • 发表日期：

    2025年7月1日

查看全文
  • Mehdi Izadi,
  • Gholamreza Farahani,
  • Gholam-Reza Mohammad-Khani

摘要

The swift advancement of Internet of Things (IoT) applications has intensified the demand for efficient wireless communication systems capable of handling high-rate data transmissions. This goal can be achieved with MIMO-NOMA systems. This paper introduces a channel estimation method for a Multiple-Input Multiple-Output (MIMO) system using Non-Orthogonal Multiple Access (NOMA) based on Vertical-Bell Laboratories Layered Space-Time Zero Forcing (V-BLAST ZF). Additionally, it presents a detection technique that employs a Long Short-Term Memory (LSTM) as a deep learning model to tackle the problem of incorrect signal detection, which arises due to an imperfect Channel State Information (CSI), user interference, and channel noise. The results indicate that a higher operational power can enhance the Signal-to-Noise Ratio (SNR) and channel estimation, while a lower power consumption is linked to energy-efficient designs. Spectral efficiency is enhanced when the number of antennas is increased. Nonetheless, this raises power consumption, leading to a compromise between power consumption and spectral efficiency. Moreover, when the number of antennas and bandwidth were increased, the Mean-Squared Error (MSE) in the proposed method decreased from about 0.17 to 0.0000425. Furthermore, with an operational gain of 800 Mbps per cell, the proposed method was about 2.4% and 3.7% better than the (Regularized Zero-Forcing) RZF and M-MMSE methods, respectively. Besides, the proposed method was respectively 6.3% and 33.33% better than the S-MMSE and MR methods in increasing energy efficiency.