Cloud-Assisted Remote Sensor Network Virtualization for Distributed Consensus Estimation

TL;DR

This paper introduces a cloud-assisted sensor network virtualization and distributed estimation framework that efficiently coordinates sensors for geographic parameter estimation, reducing communication overhead and maintaining linear convergence.

Contribution

It presents a novel virtualization algorithm for sensor selection and a distributed estimation method that outperforms traditional approaches in communication efficiency.

Findings

Reduces communication overhead compared to ADMM

Maintains linear convergence time

Effective in large-scale sensor networks

Abstract

We develop cloud-assisted remote sensing techniques for enabling distributed consensus estimation of unknown parameters in a given geographic area. We first propose a distributed sensor network virtualization algorithm that searches for, selects, and coordinates Internet-accessible sensors to perform a sensing task in a specific region. The algorithm converges in linearithmic time for large-scale networks, and requires exchanging a number of messages that is at most linear in the number of sensors. Second, we design an uncoordinated, distributed algorithm that relies on the selected sensors to estimate a set of parameters without requiring synchronization among the sensors. Our simulation results show that the proposed algorithm, when compared to conventional ADMM (Alternating Direction Method of Multipliers), reduces communication overhead significantly without compromising the estimation error. In addition, the convergence time, though increases slightly, is still linear as in the case of conventional ADMM.