Scholarlink

<p>The Non-orthogonal multiple access (NOMA) systems have become a hopeful one that addresses the need for fifth-generation (5G) communication while resolving the issues with spectrum scarcity. NOMA’s major objective is to improve the spectrum utilization while sacrificing an effective utilization of resources. Therefore, this work designed an efficient user clustering as well the power allocation scheme with the aid of deep learning (DL) enabled White Feedback Sea Lion Optimization (WFSLnO). Here, the downlink femtocell NOMA power consumption scheme includes one macro-base Station (BS) contained by a cluster of femtocell BSs. In addition, user clustering is accomplished by Deep Fuzzy Clustering (DFC), in which the user grouping parameters like Signal-to-Interference-plus-Noise-ratio (SINR), position, initial power, and channel gain are utilized. Moreover, the Backpropagation Neural Network (BPNN) is employed for the power allocation process. Furthermore, the proposed WFSLnO optimized the BPNN’s hyperparameters. Here, the WFSLnO enabled BPNN’s power allocation performance is revealed by considering the metrics including energy efficiency, achievable rate, throughput, and sum rate, as well as the corresponding values achieved are 2.975 Mbits/sec, 0.039 Mbits/Joules, 18.49 Mbits/sec and 0.631Mbps.</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

User clustering and power allocation based deep learning enabled hybrid feedback shark Lion optimization

  • 期刊论文

  • 发表日期：

    2025年8月7日

查看全文
  • Kasula Raghu,
  • Puttha Chandrasekhar Reddy

摘要

The Non-orthogonal multiple access (NOMA) systems have become a hopeful one that addresses the need for fifth-generation (5G) communication while resolving the issues with spectrum scarcity. NOMA’s major objective is to improve the spectrum utilization while sacrificing an effective utilization of resources. Therefore, this work designed an efficient user clustering as well the power allocation scheme with the aid of deep learning (DL) enabled White Feedback Sea Lion Optimization (WFSLnO). Here, the downlink femtocell NOMA power consumption scheme includes one macro-base Station (BS) contained by a cluster of femtocell BSs. In addition, user clustering is accomplished by Deep Fuzzy Clustering (DFC), in which the user grouping parameters like Signal-to-Interference-plus-Noise-ratio (SINR), position, initial power, and channel gain are utilized. Moreover, the Backpropagation Neural Network (BPNN) is employed for the power allocation process. Furthermore, the proposed WFSLnO optimized the BPNN’s hyperparameters. Here, the WFSLnO enabled BPNN’s power allocation performance is revealed by considering the metrics including energy efficiency, achievable rate, throughput, and sum rate, as well as the corresponding values achieved are 2.975 Mbits/sec, 0.039 Mbits/Joules, 18.49 Mbits/sec and 0.631Mbps.