Hardware Acceleration of Kalman Filter for Leak Detection in Water Pipeline Systems using Wireless Sensor Network

Karray, Fatma; Maalaoui, Melek; Obeid, Abdulfattah M.; García-Ortiz, Alberto; Abid, Mohamed

doi:10.1109/hpcs48598.2019.9188156

Cited by 4 publications

(6 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…

Q

R

, and

Z(t)={\left[{p}_x(t),{p}_y(t)\right]}^T

are defined as process noise variance, observation noise variance, and detection result, then the calculation flow of the t th Kalman filter is as follows 17 :

$$ \left\{\begin{array}{c}{X}_{\mathrm{pre}}(t)={FX}_{\mathrm{KF}}\left(t-1\right)\\ {}{P}_{\mathrm{pre}}(t)= FP(t){F}^T+Q\\ {}K(t)...

…”

Section: Methodsmentioning

confidence: 99%

“…For the real‐time tracking of space dim targets, it is necessary to complete the general tracking calculation acceleration in response to different space areas and complex backgrounds, which puts forward higher requirements for the engineering implementation architecture 16 . The previous design mainly relies on the acceleration design of hardware and software, uses high performance CPU to complete the control of complex processes, realizes matrix multiplication and other operations in the logical part, or adopts the simplification for some special applications to reduce resource and time consumption 17 . Some literatures also started to optimize algorithms, combining sequential fusion and square root methods to reduce the gain calculation process matrix and reverse to improve real‐time performance, but such methods require the measurement noise co‐variance to be a constant matrix or diagonal matrix, with strict constraints 18 .…”

Section: Introductionmentioning

confidence: 99%

“…16 The previous design mainly relies on the acceleration design of hardware and software, uses high performance CPU to complete the control of complex processes, realizes matrix multiplication and other operations in the logical part, or adopts the simplification for some special applications to reduce resource and time consumption. 17 Some literatures also started to optimize algorithms, combining sequential fusion and square root methods to reduce the gain calculation process matrix and reverse to improve real-time performance, but such methods require the measurement noise co-variance to be a constant matrix or diagonal matrix, with strict constraints. 18 Some literatures also adopt the method based on block inversion for acceleration, but the block matrix in the process of gain matrix inversion cannot guarantee that the inverse can be obtained in all cases.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Parallel accelerated computing architecture for dim target tracking on‐board

Yu,

Huang,

et al. 2023

Computational Intelligence

View full text Add to dashboard Cite

The real‐time tracking process of dim targets in space is mainly achieved through the correlation and prediction of dots after the detection and calculation process. The on‐board calculation of the tracking needs to be completed in milliseconds, and it needs to reach the microsecond level at high frame rates. For real‐time tracking of dim targets in space, it is necessary to achieve universal tracking calculation acceleration in response to different space regions and complex backgrounds, which poses high requirements for engineering implementation architecture. This paper designs a Kalman filter calculation based on digital logic parallel acceleration architecture for real‐time solution of dim target tracking on‐board. A unified architecture of Vector Processing Element (VPE) was established for the calculation of Kalman filtering matrix, and an array computing structure based on VPE was designed to decompose the entire filtering process and form a parallel pipelined data stream. The prediction errors under different fixed point bit widths were analyzed and deduced, and the guidance methods for selecting the optimal bit width based on the statistical results were provided. The entire design was engineered based on Xilinx's XC7K325T, resulting in an energy efficiency improvement compared to previous designs. The single iteration calculation time does not exceed 0.7 microseconds, which can meet the current high frame rate target tracking requirements. The effectiveness of this design has been verified through simulation of random trajectory data, which is consistent with the theoretical calculation error.

show abstract

“…

Q

R

, and

Z(t)={\left[{p}_x(t),{p}_y(t)\right]}^T

are defined as process noise variance, observation noise variance, and detection result, then the calculation flow of the t th Kalman filter is as follows 17 :

$$ \left\{\begin{array}{c}{X}_{\mathrm{pre}}(t)={FX}_{\mathrm{KF}}\left(t-1\right)\\ {}{P}_{\mathrm{pre}}(t)= FP(t){F}^T+Q\\ {}K(t)...

…”

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Parallel accelerated computing architecture for dim target tracking on‐board

Yu,

Huang,

et al. 2023

Computational Intelligence

View full text Add to dashboard Cite

show abstract

“…The proposed [21], Kalman Filter (KF), lessens the burden on the environment, we've developed and deployed a hardware acceleration of the KF. Time to completion, power usage, and other metrics have been compared between the software and hardware versions of the method and space requirements.…”

Section: Ijici Issn: 2791-2868 Rmentioning

confidence: 99%

Increase Efficiency And Lifetime In Wsns Through Data Reduction

Alsawad

2022

IJICI

View full text Add to dashboard Cite

The mining and processing of data for wireless sensor networks (WSNs) have been a significant research area in a number of computer science disciplines, including distributed applications, database management systems, and information gathering. The main goal of implementing applications based on WSNs is to make real-time choices, which have proven very difficult due to the limitations of computing, communication, and data mining capacity for WSNs. Thus, due to the nature and unique characteristics of sensor data and the limitations of WSNs, traditional data mining methods cannot be easily applied to them. This paper presents a new approach, the Kalman Filter (KF) with K-Nearest Neighbors (KNN) named (KF-KNN) which is proposed for data classification and collection as well as noise elimination in WSNs, increasing the network efficiency and lifetime. The proposed technique is compared to KNN to illustrate the effectiveness of the recommended ways in increasing energy consumption and prolonging network lifetime.

show abstract

“…• The use of cosine similarity in sensor node creation further simplifies the process based on the data's density and similarity. [30] • Kalman Filter…”

Section: Introductionmentioning

confidence: 99%

A New Method Based on Machine Learning to Increase Efficiency in Wireless Sensor Networks

Alkenani

Kheerallah

2023

IJCAI

View full text Add to dashboard Cite

Wireless sensor networks (WSNs) contain many sensor nodes, and this network is used for many applications such as military, medical, and others. Accurate data aggregation and routing are critical in hostile environments, where sensors' energy consumption must be carefully monitored. There is, nevertheless, a substantial probability of duplicate data due to ambient circumstances and shortdistance sensors. Large datasets include a variety of information, some of which is useful, while others are completely superfluous. This redundancy degrades performance in terms of computing cost and redundant transmission. Data aggregation, on the other hand, may eliminate redundant data in a network. In this paper new method called Kalman filter with Support vector machine (KF-SVM) is introduced to classify and data aggregate and get rid of noise in WSNs, which enhances network efficiency and extends its lifetime. Povzetek: V prispevku je opisana izvirna metoda za agregiranje podatkov v senzorskih omrežjih, ki dela na osnovi Kalmanovega filtra in SVM.

show abstract

Hardware Acceleration of Kalman Filter for Leak Detection in Water Pipeline Systems using Wireless Sensor Network

Cited by 4 publications

References 21 publications

Parallel accelerated computing architecture for dim target tracking on‐board

Parallel accelerated computing architecture for dim target tracking on‐board

Increase Efficiency And Lifetime In Wsns Through Data Reduction

A New Method Based on Machine Learning to Increase Efficiency in Wireless Sensor Networks

Contact Info

Product

Resources

About