Swarm Intelligence with Adaptive Neuro-Fuzzy Inference System-Based Routing Protocol for Clustered Wireless Sensor...

Area of Science:

• Computer Science
• Electrical Engineering
• Network Engineering

Background:

• Wireless Sensor Networks (WSNs) face challenges in maximizing network lifetime due to energy constraints of sensor nodes.
• Existing clustering and routing techniques aim for energy efficiency but often treat multihop routing as an NP-hard problem.
• Computational intelligence, including fuzzy logic and swarm intelligence (SI), offers potential solutions for complex routing decisions.

Purpose of the Study:

• To design a novel routing protocol, Swarm Intelligence with Adaptive Neuro-Fuzzy Inference System-based Routing (SI-ANFISR), for clustered WSNs.
• To enhance energy efficiency and prolong the operational lifetime of WSNs through optimized cluster head selection and multihop routing.
• To introduce a new approach by integrating the Squirrel Search Algorithm (SSA) for tuning the ANFIS model in WSN routing.

Main Methods:

• A weighted clustering algorithm is employed for electing cluster heads (CHs) and forming clusters.
• An Adaptive Neuro-Fuzzy Inference System (ANFIS) model is designed for optimal route selection, utilizing residual energy, node degree, and node history as inputs.
• The Squirrel Search Algorithm (SSA) is utilized to fine-tune the membership functions (MFs) of the ANFIS model for improved routing performance.

Main Results:

• The proposed SI-ANFISR technique effectively determines cluster heads and optimal multihop routes.
• Experimental validation demonstrates significant performance improvements compared to existing WSN routing techniques.
• The SI-ANFISR technique achieved a maximum throughput of 43838 kbps and a residual energy of 0.4800 J.

Conclusions:

• The SI-ANFISR protocol offers a novel and effective solution for enhancing WSN lifetime and performance.
• The integration of ANFIS with SSA presents a unique contribution to WSN routing strategies.
• The optimized routing approach leads to substantial gains in network throughput and residual energy conservation.