| Wireless sensor network is a new generation of sensor network. This technology is a combinationof sensor technology, embedded computing technology, distributed information processing technologyand communication technology. Its application and development will have far reaching implicationsto various fields of human life and production. However, energy issue directly affects its performanceand survival time, becomes a crucial restriction of practical application. For this "bottlenecks", manyscholars have made a considerable number of energy management strategies to prolong survival timeof wireless sensor networks to some extent, but have not fundamentally solve this energy problem.This article studies the “self-powered” wireless sensor networks which are powered by theenvironment energy. To solve the energy issue and make a correctly and reasonably using of it are thefocus of this article. The self-powered wireless sensor node is supplemented by the harvested energyfrom environment to supply its energy consumption; which makes a great effort to ease the energybottleneck and also improves the network performance. According to the way of energy harvesting,two different design of low-power self-powered nodes are presented, detailed analysis of itscomposition, unit energy consumption of each sub-model and system level efficiency are also shown,simultaneously, optimized energy harvesting prediction algorithm and system energy managementalgorithm also analyzed. In addition, we research the characteristics of self-powered wireless sensornetwork protocol. Dedicated unequal clustering adaptive routing protocol and cooperativecommunication is successfully applied to the network’s inter-cluster and intra-cluster communication.Specifically, it contains the following aspects:1. Design a solar self-powered wireless sensor node BFCap with adaptive energy acquisitionand management strategy; it is made up of a bidirectional flyback converter. It couldachieve maximum power transfer tracking algorithm, and has a wider range of energyharvesting and coldstart function. Also a WCSAD algorithm can more accurately predict thenext stage of collecting solar energy.2. Design a self-powered wireless sensor node energy management architecture based onreinforcement learning, when faced with unpredictable environmental energy. Thearchitecture can not only make the node to meet "permanent" run energy demand real-time,but also can to meet the quality of service requirements in the specific tasks of theself-powered wireless sensor network applications. 3. Analysis and design of wind-powered wireless sensor node system energy managementalgorithm. By addressing two core issues: the maximum power acquisition and dynamicvoltage regulation, the adoption of self-powered wireless sensing system organically linkedand two aspects of energy consumption, the experiments show that the algorithm canachieve true node-level energy optimization management.4. Designed a self-powered wireless sensor network routing protocol based on geneticalgorithm, the protocol by using the centralized operation clustering of large amount ofcomputation and routing in the gateway building algorithm, self-powered nodes distributedrun cluster head of the low-power adjustment and routing update process, in order toimprove as much as possible the network energy utilization.5. Based on the clustering EH-WSN designed the last chapter, a Cooperative communicationprotocol is used here. The protocol uses a distributed application on DF manner, it is a jointoptimization algorithm by the members’ node running power allocation and relay selection,to optimize the energy consumption between cluster heads communications, and it canimprove the network energy utilization ratio effectively. |
PDF Full Text Request