Energy-efficient computation and communication scheduling for cluster-based in-network processing in large-scale wireless sensor networks

Emerging Wireless Sensor Networks (WSN) applications demand considerable computation capacity for in-network processing. To achieve the required processing capacity, cross-layer collaborative in-network processing among sensors emerges as a promising solution: Sensors not only process information at the application layer, but also synchronize their communication activities to exchange partially processed data for parallel processing. Task mapping and scheduling plays an important role in parallel processing. Though this problem has been extensively studied in the high performance computing area, its counterpart in WSNs remains largely unexplored. Scheduling computation and communication events is a challenging problem in WSNs due to limited resource availability and shared communication medium. This research investigates the energy-efficient task mapping and scheduling problem in large-scale WSNs composed of homogeneous wireless sensors. A hierarchical WSN architecture is assumed to be composed of sensor clusters, where applications are iteratively executed. Given this environment, task mapping and scheduling in single-hop clustered WSNs is investigated for energy-constrained applications. Based on the proposed Hyper-DAG model and single-hop channel model, the EcoMapS solution minimizes schedule lengths subject to energy consumption constraints. Secondly, real-time applications are also considered in single-hop clustered WSNs. Incorporating the novel Dynamic Voltage Scaling (DVS) algorithm, the RT MapS solution provides deadline guarantee with the minimum balanced energy consumption. Next, the task mapping and scheduling problem is further addressed in its general form for multi-hop clustered WSNs. A novel multi-hop channel model is developed, and a multi-hop communication scheduling algorithm is presented, based on which the MTMS solution minimizes application energy consumption subject to deadline constraints. Finally, low-complexity sensor failure handling algorithms are developed to recover network functionality when sensors failures occur in single-hop and multi-hop clustered WSNs.