Scholarlink

<p>Non-Orthogonal Multiple Access (NOMA) has gained prominence as a preferred technology due to its superior Spectral Efficiency (SE) and its pivotal role in expanding the capacity of future networks. However, the high Outage Probability (OuP) encountered by the far user in NOMA necessitates the implementation of Cooperative NOMA (C-NOMA). C-NOMA is a technology employed in wireless systems to improve the sum rate, SE, boost overall system performance, and reduce OuP for the far user. This research investigates an Optimization problem (OP) focused on minimizing the total transmission power at the transmitter in a Downlink (DL) Multi-Carrier (MC) C-NOMA that incorporates Simultaneous Wireless Information and Power Transfer (SWIPT). First, we propose a User Assignment Algorithm (UAA) based on users’ Best Signal-to-Interference-plus-Noise Ratio (B-SINR). Following this, we formulate an OP to achieve minimum transmission power at the transmitter, constrained by a minimum SINR requirement. In a C-NOMA system, a Deep Neural Network (DNN) is employed to allocate power optimally among users. To optimize the Power Allocation (PA) process using the DNN model, a Feed-Forward Neural Network (FWNN) is used to fine-tune the network weights. This approach significantly enhances the sum rate, SE and EE in C-NOMA. The integration of optimal relay selection and DNN-based PA improves the OuP performance for far users in challenging channel conditions. Finally, we present representative numerical results to validate our findings.</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

Improving System Efficiency and Remote User Outage Probability in MC Cooperative NOMA Using Deep Learning Technique

  • 期刊论文

  • 发表日期：

    2025年12月2日

查看全文
  • Pawan Gupta,
  • Ashok Kumar,
  • Rakesh Sharma

摘要

Non-Orthogonal Multiple Access (NOMA) has gained prominence as a preferred technology due to its superior Spectral Efficiency (SE) and its pivotal role in expanding the capacity of future networks. However, the high Outage Probability (OuP) encountered by the far user in NOMA necessitates the implementation of Cooperative NOMA (C-NOMA). C-NOMA is a technology employed in wireless systems to improve the sum rate, SE, boost overall system performance, and reduce OuP for the far user. This research investigates an Optimization problem (OP) focused on minimizing the total transmission power at the transmitter in a Downlink (DL) Multi-Carrier (MC) C-NOMA that incorporates Simultaneous Wireless Information and Power Transfer (SWIPT). First, we propose a User Assignment Algorithm (UAA) based on users’ Best Signal-to-Interference-plus-Noise Ratio (B-SINR). Following this, we formulate an OP to achieve minimum transmission power at the transmitter, constrained by a minimum SINR requirement. In a C-NOMA system, a Deep Neural Network (DNN) is employed to allocate power optimally among users. To optimize the Power Allocation (PA) process using the DNN model, a Feed-Forward Neural Network (FWNN) is used to fine-tune the network weights. This approach significantly enhances the sum rate, SE and EE in C-NOMA. The integration of optimal relay selection and DNN-based PA improves the OuP performance for far users in challenging channel conditions. Finally, we present representative numerical results to validate our findings.