Scholarlink

<p>Load Balancing (LB) extends the lifespan of the Wireless Sensor Network (WSN) by reducing hotspot congestion and wireless collisions. However, the prevailing works did not consider dynamic traffic load balancing regarding the data types. Thus, the balancing of WSN’s traffic loads using the Temporal-Difference Q-Linear Programming Optimization Algorithm (TDQ-LPOA) based on message types is proposed in this paper. The WSN is initialized, and the parameters are set by using the Objective Modular Network Testbed in C +  + (OMNET). Then, the heterogeneity of the Sensor Nodes (SNs) is controlled using Density Gradient Field Clustering of Applications with Noise (DGFCAN). The data is then sensed. Based on the extracted attributes, the Weighted Hebbian Learning Principles Echo State Network (WHLPESN) is utilized to classify the data. Further, Cross-Layer (CL) optimization is carried out using Cross-Layer Adaptive Sampling Feedback Control Protocol (CLASFCP). Also, the traffic is analyzed using a Sliding Window (SW). Next, the traffic loads are balanced using TDQ-LPOA. Afterward, to transfer the data in WSN, the node failure is detected using WHLPESN. Then, the optimal route is identified using the Ad Hoc weighted Aging Priority Multipath Distance Vector (AHAPMDV). Thus, the effective LB with an Average Latency (AL) of 2409.9273&#xa0;ms and routing of data with an Average Delay (AD) of 13.5556&#xa0;ms are attained.</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

Data type classification-based dynamic traffic load balancing in wireless sensor network using WHLPESN and TDQ-LPOA

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月19日

查看全文
  • Erick J. Machiwa,
  • Verdiana G. Masanja,
  • Joseph W. Matiko,
  • Michael F. Kisangiri

摘要

Load Balancing (LB) extends the lifespan of the Wireless Sensor Network (WSN) by reducing hotspot congestion and wireless collisions. However, the prevailing works did not consider dynamic traffic load balancing regarding the data types. Thus, the balancing of WSN’s traffic loads using the Temporal-Difference Q-Linear Programming Optimization Algorithm (TDQ-LPOA) based on message types is proposed in this paper. The WSN is initialized, and the parameters are set by using the Objective Modular Network Testbed in C +  + (OMNET). Then, the heterogeneity of the Sensor Nodes (SNs) is controlled using Density Gradient Field Clustering of Applications with Noise (DGFCAN). The data is then sensed. Based on the extracted attributes, the Weighted Hebbian Learning Principles Echo State Network (WHLPESN) is utilized to classify the data. Further, Cross-Layer (CL) optimization is carried out using Cross-Layer Adaptive Sampling Feedback Control Protocol (CLASFCP). Also, the traffic is analyzed using a Sliding Window (SW). Next, the traffic loads are balanced using TDQ-LPOA. Afterward, to transfer the data in WSN, the node failure is detected using WHLPESN. Then, the optimal route is identified using the Ad Hoc weighted Aging Priority Multipath Distance Vector (AHAPMDV). Thus, the effective LB with an Average Latency (AL) of 2409.9273 ms and routing of data with an Average Delay (AD) of 13.5556 ms are attained.