Research On Energy Efficient Neighbor Discovery Mechanism In Wireless Ad Hoc Networks

With the continuous development of wireless communications technology, sensor technology, embedded computing, distributed information processing technology and small wireless devices widely used, wireless self-organizing network has attracted much attention of the current international and involved a highly crossed multidisciplinary and highly integrated knowledge. Wireless ad hoc network is an infrastructure independent multi-hop self-organized network. Wireless ad hoc network has important application prospect in the field of commercial and military. In the construction of smart city, wireless ad hoc networks are widely used in intelligent transportation systems, environmental monitoring systems. Wireless ad hoc networks are applied to track wild animals and configure networks with handheld device. Wireless ad hoc networks are used for dangerous relief and battlefield communication in complex environment. In all these network applications, neighbor discovery is the first step towards self-organization and is the premise and foundation to realize routing and topology. Although neighbor discovery plays an important role in wireless ad-hoc network, it has not got enough attention.Most nodes in wireless ad hoc networks are supported with battery. Low duty cycle operations are widely used in wireless ad hoc networks, which can effectively save energy. But it will bring new challenges that asynchronous being awake and asleep make it difficult for node to discover neighbors. Therefore, neighbor discovery problem is further studied on the base of the existing. The main research activities include the following aspects:(1) Research on a common framework of Quorum-based neighbor discovery algorithm. First, we study Quorum system, propose a theory that Quorum systems which satisfy the rotation closure property can be used to design the neighbor discovery algorithm. At the same time, it proves that grid quorum, cyclic quorum and torus quorum satisfy the rotation closure property and all can be used to design neighbor discovery algorithm.(2) Study on the optimal energy consumption value for a given discovery latency. Several classical algorithm assume time slots always perfectly aligned, but in the actual scenario, time slots are in a non-aligned state most of the time. This results that optimal energy consumption value in classical algorithm is not a real optimal. To this end, we take full advantage of the characteristics of time slot non-alignment and get the optimal energy consumption value.(3) Research on E-cyclic quorum algorithm. Based on the optimal energy consumption value and combined with cyclic quorum system, we propose a new neighbor discovery algorithm. This algorithm can simultaneously realize the neighbor discovery in the symmetric and asymmetric scenarios.(4) Energy and latency comparison of performance of neighbor discovery algorithms. First, determine the experimental performance indicators, including cumulative distributed function of discovery latency and power-latency product; second, design validation experiments, compare the performance of different algorithms neighbor discovery and make guidance of choosing discovery algorithm for different application scenarios.