Using end-to-end data to infer sensor network topology

Knowledge of sensor network topology is useful for understanding the structure of the sensor network, and also important for resource management and redeployment. Additionally, it is a crucial component of sensor network tomography techniques. In this paper we propose a new algorithm, namely hamming distance and hop count based classification algorithm (HHC), to infer network topology by using end-to-end data in sensor network. Specifically, we consider the case of inferring sensor network topology during the aggregation of the data from a collection of sensor nodes to a sink node. The HHC algorithm identifies sensor network topology using hamming distance of the sequences on receipt/loss of data maintained in the sink node and incorporating the hop count available at each node. We implement the algorithms in a simulated network and validate the algorithm's performance in accuracy and efficiency.