Optimize Aggregation Latency of Battery-Free Sensor Network: Constructing the BF/BP Hybrid Network

With the development of energy harvesting technol-ogy, energy-harvesting devices have been integrated into Wireless Sensor Network (WSN), named as Battery-Free WSN (BF-WSN). The aggregation latency is an important issue in BF-WSN. Many researches study the problem of Minimum Latency Aggregation Scheduling (MLAS) to achieve as small latency as possible. However, due to the poor ambient energy condition, the latency is still high even when the best scheduling strategy is applied, compared with the Battery-Powered WSNs (BP-WSNs). Therefore, we first propose the BF-BP Hybrid WSN (HWSN) to reduce the aggregation latency of BF - WSN under the constraint of costs of replacing BF -nodes with BP-nodes and prove that the construction of HWSN is NP-Hard. We introduce a method based on the Hidden Markov Model to construct a HWSN starting from a BF - WSN and give the proof of correctness and approximation ratio of the construction algorithm. The theoretical analysis and simulation results verify that our proposed HWSN can achieve high performance in terms of latency with acceptable costs compared to that of BF - WSNs.