Research On The Fast Aggregation Convergecast Problem In Wireless Sensor Networks

One of the core functions of the wireless sensor networks is the monitoring data collection.To fulfil the task,all the sensor nodes in the network need to transmit their monitoring data to a sink node.The many-to-one transmission pattern is called convergecast.While convergecasting,we can have the intermediate relay nodes merge their received data packets with their own ones by means of data aggregation scheme,and in this way we implement aggregation convergecast.Aggregation convergecast is an efficient way of data collection.We investigate the aggregation convergecast scheduling problem in this paper,aiming at reduce the time needed for data collection in wireless sensor networks.We first design a pollution monitoring system based on wireless sensor network,and take it as an instance of the aggregation convergecast problem.Then,we provide the node communication model and the network structure model,and formulized the problem based on the models.For the transmission target choosing problem in aggregation convergecast,we provide an aggregation tree construction algorithm based on minimum lower bound spanning tree.To achieve a small delay lower bound,we construct the aggregation tree gradually by adding to it a link in each iteration,which enables the convergecast delay lower bound of the tree maintains minimum.The aggregation tree built with this algorithm has a small delay lower bound,which enables it to have more chance to achieve a low convergecast delay.For the time slot allocation problem in aggregation convergecast,we provide the Neighbor Degree Ranking algorithm together with a supplementary scheduling scheme.The algorithm takes the sum of the neighbors' degree as the ranking parameter for the candidate senders and allocate time slot to them successively.In this way the time slot is preferentially allocated to the nodes in the key region of the network and the conflict probability of the subsequent transmissions is reduced.In addition,the parallel transmissions in the same time slot are added by unsing the supplementary scheduling scheme.At last,we build the experimental sense for simulation and analyze the aggregation tree construction phase and the time slot allocation phase of our algorithm and compare it with other algorithms.The experiment shows that the proposed scheduling algorithm outperforms other work in most cases,and the performance advantage of our algorithm mainly comes from more disperse tree links and more parallel transmissions.In a word,the experimental results indicate the feasibility of our approach for fast data collection in wireless sensor networks.