An analysis of loss-free data aggregation for high data reliability in wireless sensor networks

Data aggregation is am important feature in Wireless Sensor Networks, used primarily to reduce energy use. This paper extends our previous results, which showed that data aggregation can improve the reliability of data delivery rather than degrading it as previously assumed. These previous results were based on the use of scaleable aggregation functions, such as SUM, COUNT, MIN, MAX which work independently of the network size. In this paper we extend these results to consider the reliability and energy efficiency of lossless data delivery with the semi-scaleable aggregation function APPEND, and determine the boundary conditions under which the data reliability can be maintained without an increase in the energy cost. These new results show that lossless aggregation using the APPEND aggregation function can provide improved reliability with reduced energy usage in certain conditions.