Modeling optimal dynamic scheduling for energy-aware workload distribution in wireless sensor networks

Energy-aware workload distribution becomes crucial for extending the lifetime of wireless sensor networks (WSNs) in complex applications as those in Internet-of-Things or in-network DSP processing scenarios. Today static workload schedules are well understood, while dynamic schedules (i.e., with multiple partitions) remain unexplored. This paper models the dynamic scheduling by considering both the communication and computation energy consumption. It formulates a series of (integer) linear programming problems to characterize the optimal scheduling strategies. Surprisingly, even 2-partition scheduling can provide the maximum gains. Besides the interest to evaluate the optimality of on-line heuristics for dynamic scheduling, the reported off-line strategies can be immediately applied to WSN applications.