Study On Delay-Constrained And High-Reliable Data Collection In Mobile Sensor Network

Wireless sensor network is a multi-hop network, which consists of abundant nodes with sensing, data processing, wireless communicating, and limited energy. In recent years, many applications related to sensor network constantly appeared in our real life, such as health monitoring for patients and so forth. Wireless sensor network has become a new system of information acquisition, which has been applied in many fields for data collection. The concept of data collection is that nodes send data sensing by themselves to a Sink so that users can further processes data. Data collection protocols are critical to the lifetime of network. Therefore, reasonable and feasible data collection protocols should be designed so that nodes can preserve their energy, as well as to achieve a longer lifetime and reduce the data collection delay. However, earlier research work is mainly around the static wireless sensor network. The static wireless sensor network means that there is not a node with mobile ability in wireless sensor network. The static wireless sensor network exist a series of problems, such as energy hole. In order to overcome the problems existing in the static wireless sensor network, the mobile sensor network has been proposed. The mobile sensor network means that there at least exists a node with mobile ability in wireless sensor network. Recently, the data collection protocols designed for the mobile sensor network also have lots of disadvantages.To this end, this paper makes further research on data collection protocols in mobile network and designs different data collection protocols under different network scenes.(1) For the stactic wireless sensor network, we utilize a mobile data collector to collect data and propose a delay-constrained algorithm named DOTE (delay-constrained data collection without aggregation).DOTE selects some nodes as collective nodes and lingering nodes from a shortest path tree, making lingering nodes close to the Sink as much as possible. Theoretical analyses and massive simulations show that the DOTE can not only balance the energy consumption of nodes to prolong the network lifetime, but also shorten the path length that the mobile data collector walks to reduce the data collection delay, compared with existing algorithms.(2) For the mobile sensor network in which the Sink and all nodes are constantly moving, an energy-efficient algorithm ROAR(random orientation angle routing data collection algorithm) is proposed, which is based on orientation angle routing strategy. In the ROAR, nodes exactly predict the position of the mobile Sink by the predicted algorithm. Besides, nodes utilize the strategy of orientation angle routing to transmit data, which can control the dircection of data transmission. Therefore, nodes can quickly transmit their data to the Sink, and efficiently balance energy of nodes. Theoretical analyses and massive simulations show that the ROAR can not only extend the network lifetime, but also can reduce the energy consumption of nodes and shorten the duration of time in each round of data gathering, compared with existing algorithms.