ECCA: An Energy and Context-Aware Clustering Algorithm for Energy-Efficient Cyber-Physical Sensor Networks

CPSNs, or Cyber-Physical Sensor Networks (CPSNs), have become the backbone of modern CPS in our age of cyber-physical systems (CPS) in the field of smart healthcare, industrial automation, environmental management and intelligent transportation. Despite the widespread use of CPSNs, the sensor nodes have limited energy capacity and these are the critical factors that affect communication, network stability and operating lifetime. Current clustering strategies are based on random, probabilistic or single-parameter cluster head selection models and are not suited to the dynamic network environment and the distribution of energy in different clusters. We propose an energy-efficient and context-aware clustering algorithm for CPSNs (ECCA) to improve energy utilization at each clustering cycle. We propose an analytical Context Score (CS) that combines the four key aspects of the node with respect to the residual energy, neighbourhood density, link quality and predicted remaining lifetime, and finds the best cluster head. Unlike optimization-based clustering methods that require more computation and more processing overhead with each iteration, ECCA has a deterministic analytical model with linear complexity, which makes it suitable for sensor nodes with limited resources. The adaptive cluster head selection strategy can avoid re-formation of clusters, minimize energy consumption and increase communication reliability. From the conceptual evaluation presented in this work, we found that ECCA can reduce residual energy consumption, improve node survivability, system stability and lower computational overhead compared to traditional clustering. The proposed framework provides a practical and scalable clustering solution for next generation energy-aware cyber-physical sensor networks.