Red Panda Addax Optimizer_ Bidirectional Long Short-Term Memory for cluster head selection and data aggregation in wireless sensor networks

Abstract

Wireless Sensor Networks (WSNs) rely on efficient data collection and transmission, where selecting an optimal Cluster Head (CH) is critical for managing energy and ensuring reliability. However, existing methods face challenges in handling dynamic network conditions, energy prediction, and secure data aggregation, which affect overall network performance. Here, the Red Panda Addax Optimization (RPAO) is implemented for CH selection and routing and Red Panda Addax Optimization_Bidirectional Long Short-Term Memory (RPAO_BiLSTM) is introduced for data aggregation in WSN. The RPAO is modeled with the integration of Red Panda Optimization (RPO) and Addax Optimization Algorithm (AOA). This process begins with the simulation of a WSN. Here, energy prediction is conducted by utilizing a Recurrent radial basis function (RRBF). After that, CH selection is done using the RPAO based on the fitness parameters. Thereafter, secure routing is conducted based on RPAO with the above fitness factors. Lastly, data aggregation is conducted using Bidirectional Long Short-Term Memory (BiLSTM), which is trained by the RPAO. It is identified that RPAO_BiLSTM has gained a Data Packet Delivery Rate (DPDR) of 93.676%, energy of 0.145J, delay of 0.662 ms and trust of 87.884, highlighting its effectiveness in improving the performance, reliability, and energy efficiency of WSNs.

The authors' abstract, as published at the source. Computers & Electrical Engineering, 2026 · DOI ↗