Energy-efficient discovery strategies for WSNs with mobile elements

In sparse wireless sensor networks, data collection is carried out through specialized Mobile Elements (MEs) thatvisit sensor nodes, gather data, and transport them to the collection point. Since visit times are typically unpredictable, one of the main challenges in this kind of networks is the energy-efficient discovery of MEs by sensor nodes. This thesis focuses on adaptive discovery schemes, where the sensor node’s duty cycle is adjusted over time according to the probability that the ME is nearby. Initially, a hierarchical approach is proposed based on two different Beacon messages emitted by the ME (Long Range Beacons and Short Range Beacons). Simulation results show that the proposed scheme can provide a significant energy reduction with respect to a single Beacon, especiallywhen the discovery phase is long. Later, two different adaptive discovery schemes are considered (a learning-based approach and a hierarchical approach). And their performance in different mobility scenarios is evaluated. Simulation results show that a learningbased approach is not suitable when the ME moves in an irregular pattern, and a hierarchical approach is not able to learn and exploit information about the specificmobility pattern of the ME. Finally, a hybrid discovery algorithm is proposed that combines a learning-based approach with a hierarchical approach. The proposed algorithm is very flexible as it can adapt to different mobility patterns of the MEs. The performance of the proposed approach has been evaluated through extensive simulation analysis and is compared with the existing adaptive algorithms, that only leverage either a learning-based approach or a hierarchical approach. The results show that the proposed hybrid discovery algorithm outperforms all other discovery schemes for all the considered scenarios.