Data centric storage with geographic hash table for wireless sensor networks using Residue Number Systems

In-Network storage model is effective to reduce communications overhead in the wireless sensor networks because it stores the sensed data in the sensors inside the network instead of sending them to the sink node. When the sink node wants to get some stored data it can restore it by sending request to the sensors that has stored the required data. Recently, the erasure codes are used to store the sensed data within the network. In these codes the sensed data is partitioned into several parts in which that sensed data can be restored from some subsets of these parts. In other word; the sink node can restore the sensed data even if some parts are lost. In this paper we provide a new model for Innetwork storage based on Reside Number System (RNS). This new proposed model offers an accurate analysis for the required number of parts that must be stored to ensure that the sensed data can be correctly restored even if some of sensors that stores partitions of sensed data broke down. In addition, this model reduces the number of parts needed to recover the sensed data compared with the previous approaches. This new approach locally distributes the modules between sensors by geographically clustering the sense field instead of randomly choosing them as in the previous models. We have defined a module to evaluate the mathematical model and to proof that the communications overhead becomes less.