The Effect of Popularity Rule on Capacity and Delay in Multi-Sink WSNs

摘要

Wireless sensor networks (WSNs) have a wide spectrum of applications, such as disaster relief operations, intelligent building, medicine and health care, etc. WSNs are growing in size and sensory traffic and thus, it is necessary to use multiple sinks to gather such traffic with minimal loss. In a network, there are some locations that nodes visit more often (popular) during their movements. We place sinks' nodes (base stations) on the most popular locations. In this paper, we study the asymptotic throughput-delay performance of a WSN with multiple static sinks utilizing location based popularity rule. Our network is divided into C non-overlapping cells to manage the signal interference between network cells. Also, the partitioning was applied in a way that every cell in the network has an average probability of visit. Moreover, sinks were placed in the most popular cells. According to a modified version of the Grossglauser-Tse 2-hop relay algorithm [1] for information delivery, we derive analytical bounds on throughput capacity and average network delay. We show that by utilizing location popularity rule, network delay has improved significantly, and throughput capacity has increased by a factor of 1.3 compared to randomly distributed sinks over network cells.