Optimal Distributed Fault-Tolerant Sensor Fusion: Fundamental Limits and Efficient Algorithms

by   Marian Temprana Alonso, et al.

Distributed estimation is a fundamental problem in signal processing which finds applications in a variety of scenarios of interest including distributed sensor networks, robotics, group decision problems, and monitoring and surveillance applications. The problem considers a scenario where distributed agents are given a set of measurements, and are tasked with estimating a target variable. This work considers distributed estimation in the context of sensor networks, where a subset of sensor measurements are faulty and the distributed agents are agnostic to these faulty sensor measurements. The objective is to minimize i) the mean square error in estimating the target variable at each node (accuracy objective), and ii) the mean square distance between the estimates at each pair of nodes (consensus objective). It is shown that there is an inherent tradeoff between satisfying the former and latter objectives. The tradeoff is explicitly characterized and the fundamental performance limits are derived under specific statistical assumptions on the sensor output statistics. Assuming a general stochastic model, the sensor fusion algorithm optimizing this tradeoff is characterized through a computable optimization problem. Finding the optimal sensor fusion algorithm is computationally complex. To address this, a general class of low-complexity Brooks-Iyengar Algorithms are introduced, and their performance, in terms of accuracy and consensus objectives, is compared to that of optimal linear estimators through case study simulations of various scenarios.


page 1

page 2

page 3

page 4


Linear TDOA-based Measurements for Distributed Estimation and Localized Tracking

We propose a linear time-difference-of-arrival (TDOA) measurement model ...

Lower Bounds for the Minimum Mean-Square Error via Neural Network-based Estimation

The minimum mean-square error (MMSE) achievable by optimal estimation of...

Distributed Estimation Recovery under Sensor Failure

Single time-scale distributed estimation of dynamic systems via a networ...

Scheduling of Sensor Transmissions Based on Value of Information for Summary Statistics

The optimization of Value of Information (VoI) in sensor networks integr...

Estimation-Energy Tradeoff for Scalar Gauss-Markov Signals with Kalman Filtering

In this letter, we investigate a receiver architecture, which uses the r...

Distributed Regression in Sensor Networks: Training Distributively with Alternating Projections

Wireless sensor networks (WSNs) have attracted considerable attention in...

An Information-theoretic Method for Collaborative Distributed Learning with Limited Communication

In this paper, we study the information transmission problem under the d...

Please sign up or login with your details

Forgot password? Click here to reset