Data fusion in wireless sensor networks: a new application for Kalman filtering

Honarbacht, A.; Rauschert, P.; Kummert, Anton

doi:10.1109/tencon.2004.1414469

Cited by 6 publications

(1 citation statement)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An overview of these methods can be found in [1]- [5]. Recent research has focused on some applications of the Kalman filter in sensor networks, like location algorithms, extended Kalman for energy efficient wireless sensor networks, and data fusion in sensor networks [6]- [8]. Kalman filter has presented the most complex and precise algorithm to approach the radar tracking problem.…”

Section: Introductionmentioning

confidence: 99%

A Fully-Pipelined Parallel Architecture for Kalman Tracking Filter

Merhi¹,

Ghantous²,

Elgamel³

et al. 2006

2006 International Workshop on Computer Architecture for Machine Perception and Sensing

View full text Add to dashboard Cite

The Kalmanfilter is a set ofmathematical equations that provides an efficient computational (recursive) mean to estimate the state of a process, in a way that minimizes the mean of the squared error. This filter is very powerful in several aspects. it provides estimations ofpast, present, and future states, and it can do so when the precise nature of the modeled system is unknown, and even with the presence of measurement and process noise. Moreover, Kalman filter for linear estimate is the most complex and precise algorithm used for target tracking. However, using Kalman filter algorithms in software for multitarget tracking (MTT) radar system would result in a very long computational time which may not be suitable for today's warfare constraints, or real-time processing. Consequently, a hardware alternative has to be developed which may result in big area overhead which is not suitable for today's area constraints such as sensor nodes in a sensor network. In this paper, we break the arrays into their scalar forms, and develop fully-pipelined hardware architecture for the radar tracking Kalman filter, with time division multiplex blocks to decrease the silicon area.. The proposed architecture contains 6 multipliers, 2 dividers, 9 adders, 5 subtractors, one control unit, and some registers and multiplexers for pipeline and control.Simulation results show that the loss in accuracy between the exact track and the estimated is found to be only 4.9/

show abstract

Section: Introductionmentioning

confidence: 99%

A Fully-Pipelined Parallel Architecture for Kalman Tracking Filter

Merhi¹,

Ghantous²,

Elgamel³

et al. 2006

2006 International Workshop on Computer Architecture for Machine Perception and Sensing

View full text Add to dashboard Cite

show abstract

Optimization of Kalman Filter for Target Tracking Applications

Kumar

Duraiswamy

2020

Lecture Notes in Mechanical Engineering

View full text Add to dashboard Cite

An Algorithm of Mobile Sensors Data Fusion Tracking for Wireless Sensor Networks

Chen

2009

Wireless Pers Commun

View full text Add to dashboard Cite

In this paper an algorithm of data fusion to track both of non-maneuvering and maneuvering targets with mobile sensors deployed in an WSN (wireless sensor network) is proposed and investigated. In addition, the "GATING" technique is also applied to solve the problem of MSDFT (mobile-sensor data fusion tracking) for targets, i.e., the research is constrained in an "limited area" for the reason of solving the complicated situation and reducing the computation burden happened when the multiple sensing events exist in a WSN environments. On the other hand, a simple approach which is implemented with an adaptive filter (Kalman filter) consists of a data association technique denoted as 1-step CML (conditional maximum likelihood). Furthermore, a variable structured model is established as an adaptive maneuvering compensator simultaneously to solve data association and targets maneuvering problems, i.e., a detection algorithm for tracking targets with multiple mobile sensors deployed in an WSN is investigated too. Moreover, for the purpose of improving the tracking performance, a multiple sensor fusion algorithm is suggested in this research. Finally, in order to validate the fact that such proposed tracking system is really to arrive at the goal, some simulations of multi-mobile sensor tracking based on the proposed method are conducted. Computer simulation results indicate that the approach successfully tracks the targets with multiple mobile sensors and illustrates precisely the outcome also. By the way, a piece of the developed simulating script with m-file (@Matlab) is appended for referring.Keywords CML (conditional maximum likelihood) · GATING · MSDFT (mobile sensor data fusion tracking) · Kalman filter · WSN (wireless sensor networks)

show abstract

Data fusion in wireless sensor networks: a new application for Kalman filtering

Cited by 6 publications

References 5 publications

A Fully-Pipelined Parallel Architecture for Kalman Tracking Filter

A Fully-Pipelined Parallel Architecture for Kalman Tracking Filter

Optimization of Kalman Filter for Target Tracking Applications

An Algorithm of Mobile Sensors Data Fusion Tracking for Wireless Sensor Networks

Contact Info

Product

Resources

About