A Distributed Clustering Algorithm Guided by the Base Station to Extend the Lifetime of Wireless Sensor Networks

Area of Science:

• Computer Science
• Electrical Engineering
• Network Engineering

Background:

• Clustering algorithms are crucial for reducing energy consumption in Wireless Sensor Networks (WSNs).
• Existing centralized and distributed algorithms have limitations in data requirements and inter-node communication.
• Dynamic cluster head selection is key to balancing energy consumption and extending network lifetime.

Purpose of the Study:

• To propose a novel distributed clustering algorithm for WSNs that enhances network operability.
• To dynamically form clusters and optimize cluster head selection using a fuzzy logic system.
• To reduce reliance on extensive data from nodes and minimize control messages.

Main Methods:

• A distributed clustering approach is presented, occasionally supported by the Base Station (BS).
• The Base Station reconfigures the 'skip' value (number of rounds same cluster heads are maintained) via three messages.
• Nodes utilize a Takagi-Sugeno-Kang fuzzy logic model to decide on becoming cluster heads.

Main Results:

• The proposed algorithm demonstrated an extension of network operability in various tested scenarios.
• Performance was compared against representative centralized and distributed fuzzy-logic based algorithms.
• Simulation results validated the effectiveness of the dynamic clustering and fuzzy logic approach.

Conclusions:

• The novel distributed fuzzy logic-based clustering algorithm effectively extends WSN operability.
• The Takagi-Sugeno-Kang fuzzy model offers an effective alternative for cluster head selection.
• The hybrid approach balances distributed decision-making with occasional BS support for network optimization.