Robust Distributed Estimators for Wireless Sensor Networks with One-Bit Quantized Data

In this letter, the problem of distributed estimation in a wireless sensor network with uncertain observation noise distributions is considered, where each sensor only sends one-bit quantized data to a fusion center. Two robust estimators called quantized mean estimator and trimmed mean estimator are proposed. The asymptotic relative efficiency and influence function of the proposed estimators are derived. Numerical results illustrate the performance advantages of the proposed estimators over the maximum likelihood estimator.