K-Means Based Bee Colony Optimization for Clustering in Heterogeneous Sensor Network.

Area of Science:

• Computer Science
• Network Engineering
• Artificial Intelligence

Background:

• Wireless Sensor Networks (WSNs) require efficient clustering for energy optimization and extended network lifetime.
• Traditional bee colony optimization in WSNs often faces limitations in energy efficiency and overall network performance.
• Heterogeneous sensor networks present unique challenges for clustering algorithms regarding energy consumption and data transmission.

Purpose of the Study:

• To propose a novel clustering algorithm, K-BCO, integrating Bee Colony Optimization and K-mean algorithms for heterogeneous WSNs.
• To develop a robust and efficient clustering solution addressing energy consumption and network performance challenges in WSNs.
• To enhance the stability and sustainability of wireless sensor network operations.

Main Methods:

• Developed the K-BCO algorithm by synergistically combining Bee Colony Optimization with K-mean clustering.
• Evaluated K-BCO's performance against established algorithms like H-LEACH, DBCP, and ABC-ACO.
• Measured key performance metrics including Average Error Rate (AER), Average Data Delivery Rate (ADDR), and Average Energy Consumption (AEC).

Main Results:

• K-BCO demonstrated superior performance across AER, ADDR, and AEC compared to H-LEACH, DBCP, and ABC-ACO.
• K-BCO achieved an ADDR of 95.00%, significantly outperforming H-LEACH (75.86%), DBCP (72.07%), and ABC-ACO (90.08%).
• The algorithm ensures optimized energy consumption and provides more stable, robust solutions.

Conclusions:

• The K-BCO algorithm effectively optimizes energy consumption in WSNs, extending network lifetime.
• K-BCO offers a robust and efficient clustering solution for heterogeneous wireless sensor networks.
• This approach is recommended for practitioners seeking sustainable and high-performance wireless communication.