Achieving optimal data storage position in wireless sensor networks

Data storage in wireless sensor networks (WSNs) involves producers (such as sensor nodes) storing in storage positions a large amount of data which they have collected and consumers (e.g., base stations, users, and sensor nodes) then retrieving that data. When addressing this issue, previous work failed to utilize data rates and locations of multiple producers and consumers to determine optimal data storage positions to be communication cost-effective in a mesh network topology. In this paper, we first formalize the data storage problem into a one-to-one (one producer and one consumer) model and a many-to-many (m producers and n consumers) model with the goal of minimizing the total energy cost. Based on above models, we propose optimal data storage (ODS) algorithms that can produce global optimal data storage position in linear, grid, and mesh network topologies. To reduce the computation of ODS in the mesh network topology, we present a near-optimal data storage (NDS) algorithm, which is an approximation algorithm and can obtain a local optimal position. Both ODS and NDS are locality-aware and are able to adjust the storage position adaptively to minimize energy consumption. Simulation results show that NDS not only provides substantial cost benefit over centralized data storage (CDS) and geographic hash table (GHT), but performs as well as ODS in over 75% cases.