A bio-inspired load balancing technique for wireless sensor networks

Wireless Sensor Networks (WSNs) consist of multiple distributed nodes each with limited resources. With their strict resource constraints and application-specific characteristics, WSNs contain many challenging trade-offs. This thesis is concerned with the load balancing of Wireless Sensor Networks (WSNs). We present an approach, inspired by bees’ pheromone propagation mechanism, that allows individual nodes to decide on the execution process locally to solve the trade-off between service availability and energy consumption. We explore the performance consequences of the pheromone-based load balancing approach using a system-level simulator. The effectiveness of the algorithm is evaluated on case studies based on sound sensors with different scenarios of existing approaches on variety of different network topologies. The performance of our approach is dependant on the values chosen for its parameters. As such, we utilise the Simulated Annealing to discover optimal parameter configurations for pheromone-based load balancing technique for any given network schema. Once the parameter values are optimised for the given network topology automatically, we inspect improving the pheromone-based load balancing approach using robotic agents. As cyber-physical systems benefit from the heterogeneity of the hardware components, we introduce the use of pheromone signalling-based robotic guidance that integrates the robotic agents to the existing load balancing approach by guiding the robots into the uncovered area of the sensor field. As such, we maximise the service availability using the robotic agents as well as the sensor nodes.