Distributed compression-estimation using wireless sensor networks

Abstract

Public Reporting Burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. ABSTRACTIn this paper we consider deterministic parameter estimation problems. We study the intertwining of quantization and estimation in general and shows particular results in i) low SNR situations where the noise standard deviation is in the order of the parameter's dynamic range; and ii) universal estimation when the sensor data and noise model are unknown. The goal is to understand how the signal processing capability of a WSN scales up with its size, and to develop robust distributed signal processing algorithms and protocols with low bandwidth requirement and optimal performance. We show that for universal estimation in low signal to noise ratio (SNR), the universal distributed estimators not only exist but achieve performance close to that of estimators based on the original (un-quantized) observations. We also generalize these results a Bayesian estimation framework with a particular application to state estimation of dynamic stochastic processes.

A Novel Fast Dimension‐Reducing Ranked Query Method with High Security for Encrypted Cloud Data LIU Lianggui and LIU Zhengjin (School of Information Science and Technology, Zhejiang Sci‐Tech University, Hangzhou 310018, China)

“…Since Σ is positive definite, Σ can be expressed as Σ = Φ T Φ such that Φ is invertible. Then, using (7) and noticing y k ∈ R p , we have…”

“…where γ k is determined by ( 8) and (7). For the proposed MMSE state estimator, we easily see that we only need to prove (34) and (36) where we easily obtain (34) by using (20) and (22).…”

“…Event-triggered scheme is an effective means to reduce sensorto-estimator communication cost since communication is not permitted unless a pre-defined triggered condition is satisfied. Previous studies have shown that event-triggered scheme can strike a proper balance of trade-offs between communication cost and estimation performance [7]− [10].…”

Generalized algebraic connectivity for asymmetric networks

“…However, if L has a pair of complex conjugate dominant eigenvalues, only the subspace sequence {W k } k∈N converges asymptotically to the two-dimensional subspace W * := span{v n−1 ( L), v n ( L)} spanned by the two right eigenvectors associated with the pair of complex conjugate dominant eigenvalues of L, while the subspace sequence {V k } k∈N is not convergent. By defining matrices Řk and Rk as the projections of L onto the subspaces V k and W k , respectively, it can be demonstrated that the Euclidean norm of sequence { Řk } k∈N converges to the magnitude of the real dominant eigenvalue of L, while that of sequence { Rk } k∈N converges to the magnitude of the complex conjugate dominant eigenvalues of L. Using (9) and the obtained sequence of the magnitude of the dominant eigenvalue of L, the sequence { λk } k∈N is computed which is proved to asymptotically converge to the network's GAC.…”

“…In parameter estimation algorithms, particularly, a set of unknown parameters are estimated using sensor data corrupted by measurement noise [5], [8]. Estimation algorithms performing distributed computations have significant advantages compared to methods based on a centralized fusion scheme in terms of scalability and resilience to node failure [9]. The connectivity degree of the network can significantly affect the convergence speed of cooperative algorithms used for various objectives over ad-hoc networks [4], [7].…”