Decentralized Estimation in an Inhomogeneous Environment 1
Zhi-Quan Tom Luo, Jin‐Jun Xiao · 2004
We consider the decentralized estima- tion of a noise-corrupted deterministic parameter by a bandwidth constrained sensor network with a fusion center. Extending the work of (1, 2), we construct a decentralized estimation scheme (DES) where each sensor compresses its observation to a small number of bits with length proportional to the logarithm of its local Signal to Noise Ratio (SNR). The resulting com- pressed bits from different sensors are then collected and combined by the fusion center to estimate the un- known parameter. The proposed DES is universal in the sense that the local sensor compression schemes and final fusion function are independent of noise pdf. We show that its mean squared error is within a con- stant factor to that achieved by the classical central- ized Best Linear Unbiased Estimator (BLUE).