Research On Collaborative Neighbor Discovery Technology Based On Circle

Due to the mobility of devices in wireless sensor network,each device needs to keep discovering who are in its neighborhood,referred to as neighbor discovery.Considering the limited battery capacity of such network nodes,keeping the node’s duty cycle and discovery latency at a low level is the main challenge of the neighbor discovery protocol.Especially in a network scenario with a large number of nodes,not only the pairwise neighbor discovery protocol is needed to ensure that the two nodes discover each other quickly,but also the nodes that have discovered each other need to accelerate the process of discovering new neighbors through collaboration.After at least one collaborative node and the new node discover each other,other collaborative nodes and the new node can use the information carried in the beacon to discover with each other faster or communicate in a multi-hop manner.However,state-of-the-art collaborative neighbor discovery protocols do not research and optimize this discovery process.How to collaborate to reduce the latency of at least one collaborative node discovering a new node is a problem to be solved urgently.To solve this problem,this thesis studies and proposes a collaborative scheme based on Circle model and Circle protocol.The proposed scheme reduces the collaborative discovery latency by reasonably adjusting the cycle length and phase of the collaborative nodes,and the performance of the scheme is evaluated on smartphones and Telos B nodes.The main research work of this thesis includes the following two aspects:(1)The collaborative nodes,which are scheduled by Circle protocol,have their own cycle lengths,and the time difference at the start of the cycles is called phase.Different phases and cycle lengths result in different collaborative discovery latencies.In order to explore the relationship between phase,cycle length and the average collaborative discovery latency,this thesis proposes collaborative Circle model based on Circle model.The collaborative Circle model can be used to analyze and calculate the collaborative discovery latency.Based on this model,this thesis proves the relationship between the lower bound of the average collaborative discovery latency and the sum of the duty cycles of the collaborative nodes,and gives the necessary conditions for obtaining the lower bound.According to the necessary conditions,the cycle lengths of the collaborative nodes should be the same.(2)According to the necessary conditions for obtaining the lower bound of the average collaborative discovery latency,this thesis proposes a collaborative scheme called Circlebased.Circle-based can reduce the average collaborative discovery latency by reasonably adjusting the phases of the collaborative nodes.In particular,in a scenario where the cycle length of the new node is known,when the collaborative node uses Circle-based collaborative discovery scheme,the lower bound of the average collaborative discovery latency and the maximum collaborative discovery latency can be achieved.For collaborative discovery scenarios,this thesis also proposes a low-power Quorum mechanism,which stipulates that only one collaborative node uses hyper cycle,ensuring that at least one collaborative node can discover each other with a new node with the same cycle length,while reducing the total energy consumption of the collaborative nodes.Finally,experiments were conducted on smartphones and Telos B nodes.The experimental results show that the performance of Circle-based is better than other collaborative discovery schemes.