A Filter-based Approach for Optimizing Top-k Queries in Wireless Sensor Networks

This paper presents an efficient approach for top-k query processing in wireless sensor networks (WSN) called FaVN-K. Our approach is similar to the FILA proposed by Wu et al. in that each sensor node is assigned with a value range called filter. What is different is that in FaVN-K virtual node is introduced to solve the problem of overlapping ranges. FaVN-K can achieve exact top-k result if all the nodes follow the filtering rules strictly. The experimental results show that FaVN-K can reduce transmission volume dramatically, which is especially important to WSN