An efficient statistical approach for time synchronization in wireless sensor networks

Abstract

In this paper, we propose a powerful method of estimating the model parameters for time synchronization in wireless sensor networks (WSNs). Joint estimation of clock offset and clock skew has been proposed in the literature using the standard regression framework. Here, we claim that simple regression poorly estimates the parameters because of the inherent correlation among successive time readings between two sensors. We propose an alternative autoregressive model and use generalized least squares for estimating the relative offset and skew parameters. A computationally efficient Bayesian approach is also proposed for the parameter estimation considering correlated readings between two sensors. The effectiveness of the proposed approach compared with the earlier approach has been investigated through extensive simulation studies.

A unified approach of simultaneous state estimation and anomalous node detection in distributed wireless sensor networks

“…We propose an alternative Bayesian approach for the state estimation and detection of the anomalous nodes. Bayesian approach has been previously used in the context of WSNs . In the context of time synchronization of WSNs, Chatterjee et al have shown that Bayesian approach provides computationally more efficient and faster algorithm for parameter estimation.…”

“…For complex models, implementation of the traditional methods like ML, KF is really challenging and computationally expensive. However, MCMC can easily handle complex models and estimate the model parameters iteratively in a relatively less time . Hence we also propose an alternative Bayesian model for state estimation and anomaly detection.…”

Study of Wireless Sensor Network Location Algorithm and Data Collection Protocol

“…Additionally, the position information of sink is only updated in the responsibility zone which avoids energy consumption for forwarding due to the overlapping of multiple sink position information. In MSGPFMR, all the sinks move freely, which will lead to the non-uniform distribution of sinks and a portion of sink neighboring load is too large, increasing the energy consumption [11][12][13]. The application and study of static wireless sensor network are based on the assumption that the nodes do not move.…”

Wireless Sensor Network Multi-Hop Positioning Algorithm Based on Continuous Regression

“…Therefore, after the ridge parameter is selected, computation overhead is o(m 3 ). If the method in reference [2] is adopted, the computation overhead can be reduced to o(m 2 log m). The computation overhead of the ML-CR method proposed in this paper is similar to that of the RP method.…”

“…As for the defects of multi-hop positioning, on the basis of the common multi-hop positioning method and partial least squares regression algorithm, the multi-hop positioning algorithm based on continuous regression is proposed in this paper [2]. The ML-CR method takes the continuous regression model as the optimal relational model between actual distance and node hop count, and it predicts the distance between an unknown node and a beacon node.…”