Advancements in MANETs and WSNs: Exploration of Performance Optimization through Parameter Tuning and Intelligent Algorithms

ABSTRACT

This paper investigates the performance optimization of Mobile Ad Hoc Networks (MANETs) and Wireless Sensor Networks (WSNs) through the utilization of learning algorithms, quality of service (QoS) parameter discovery, and a novel CSO-based energy-efficient reliable sectoring scheme. The proposed MANET model undergoes an in-depth analysis of over 10,000 unique situations, employing diverse configurations for the MANT routing protocol. Comparative evaluations with well-established metaheuristic algorithms, Particle Swarm Optimization (PSO) and Clonal Selection Optimization (CSO), showcase the efficacy of the proposed model. In the context of QoS parameter discovery, we employ the K-means clustering technique for effective clustering of QoS values. This method proves particularly valuable for service requestors seeking an optimal selection of services in competitive contexts. The approach excels in identifying services that meet the specific criteria of service requestors, ensuring efficient decision-making in dynamic and highly competitive environments. Addressing the challenges in WSNs, our proposed CSO-based energy-efficient reliable sectoring scheme achieves several objectives: enabling single-hop communication between Cluster Head (CH) and Sink, facilitating multi-hop communication between sensor nodes and CH, ensuring balanced cluster density and predefined path discovery, optimizing cluster count, location, and the election process of CH, and leveraging swarm-based optimization as a computationally efficient alternative to analytical methods. Additionally, the integration of a prediction algorithm mitigates the constraints associated with multi-hop communication, further enhancing the overall efficiency and reliability of the WSN.

FULL TEXT