Robust Calibration for Localization in Clustered Wireless Sensor Networks
Additional Document Info
This paper presents a robust calibration procedure for clustered wireless sensor networks. Accurate calibration of between-node distances is one crucial step in localizing sensor nodes in an ad-hoc sensor network. The calibration problem is formulated as a parameter estimation problem using a linear calibration model. For reducing or eliminating the unwanted influence of measurement corruptions or outliers on parameter estimation, which may be caused by sensor or communication failures, a robust regression estimator such as the least-trimmed squares (LTS) estimator is a natural choice. Despite the availability of the FAST-LTS routine in several statistical packages (e.g., R, S-PLUS, SAS), applying it to the sensor network calibration is not a simple task. To use the FAST-LTS, one needs to input a trimming parameter, which is a function of the sensor redundancy in a network. Computing the redundancy degree and subsequently solving the LTS estimation both turn out to be computationally demanding. Our research aims at utilizing some cluster structure in a network configuration in order to do robust estimation more efficiently. We present two algorithms that compute the exact value and a lower bound of the redundancy degree, respectively, and an algorithm that computes the LTS estimation. Two examples are presented to illustrate how the proposed methods help alleviate the computational demands associated with robust estimation and thus facilitate robust calibration in a sensor network. Note for Practitioners-Wireless sensor network is an emerging technology that finds numerous civilian and military applications lately. When utilizing the information from a sensor network for decision making, a statistical estimation procedure is often a necessary intermediate step in order to know about critical system parameters or state variables. For the estimation purpose, the commonly used least-squares estimation (LSE) mechanism is not robust at all against data corruptions or outliers caused by sensor and communication failures. Any single corrupted data point may cause considerable deterioration in an LSE. Applying the robust estimators available from robust statistics research to a wireless sensor network, however, faces a number of computational challenges. This paper offers several useful algorithms to address thesechallenges, making the application of a robust estimator easier and more efficient. 2009 IEEE.