An analytical model for transport layer caching in wireless sensor networks

Reliable transport protocols have traditionally been designed to perform end-to-end error control transparently to the intermediate nodes (e.g., TCP). However, the resource constraints featured by Wireless Sensor Network (WSN) require a different paradigm where intermediate nodes are able to cache packets, retransmitting them on-demand in order to avoid incurring on costly end-to-end retransmissions. This paper presents an analytical model of end-to-end delivery cost for WSN reliable transport with intermediate caching. The model calculates the cost as the total number of physical layer transmissions using a probabilistic formulation that has been validated through network simulation. Although the model is based on a specific transport protocol (DTSN), the addressed mechanisms are more generic, allowing it to be easily adapted to other WSN transport protocols that also feature intermediate caching. Numerical results confirm the improved efficiency introduced by a transport layer with intermediate caching in comparison with end-to-end approaches that are based exclusively on MAC layer reliability. Different cache partitioning policies were tested, and it is shown that cache partitioning policies should take into account the network conditions experienced by concurrent flows, namely the status of the radio links and the flow lengths.