Mobile sinks are extensively used for data gathering in wireless sensor networks (WSNs). This method avoids imbalances in energy consumption caused by multihop transmission but may cause an extended delay time. In this paper, we focus on how to shorten the length of the traveling path to reduce the delay time of data gathering. We propose that the mobile sink visits the overlapping areas of communication ranges of sensors instead of sensors one by one. Next, we determine the visiting point of each overlapping area and use the traveling salesperson problem (TSP) algorithm to plan a traveling path. Because the visiting point is a point within the overlapping area of communication ranges of sensors, it is possible that the length of the traveling path can be reduced further. Hence, we attempt to shorten the traveling path obtained by the TSP algorithm. The benefit of the proposed method is that the number of visiting points is reduced after integration of visiting points. This method not only shortens the length of the traveling path for the mobile sink but reduces the computational effort required for traveling-path planning by the TSP algorithm as well. Moreover, we also consider data transfer rate in traveling-path planning to obtain a path that satisfies the constraint of the data transfer rate. Our experimental results show that the proposed algorithm delivers good results in terms of the computational effort and length of the traveling path.
IEEE Transactions on Vehicular Technology 65(4), p.2309-2324