Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Accelerometer signal denoising method based on particle filtering and wavelet transform

A particle filter and accelerometer technology, which is used in navigation, mapping and navigation, and navigation calculation tools through velocity/acceleration measurement. The effect of increasing the number of particles

Inactive Publication Date: 2018-08-24
TIANJIN UNIV
View PDF5 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, under the inaccurate signal model, Kalman filtering will bring large errors or even divergence, and it is only suitable for linear systems and standard noise conditions with zero mean and Gaussian distribution

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Accelerometer signal denoising method based on particle filtering and wavelet transform
  • Accelerometer signal denoising method based on particle filtering and wavelet transform
  • Accelerometer signal denoising method based on particle filtering and wavelet transform

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0047] Based on particle filter and wavelet transform, an embodiment of the present invention proposes a method for denoising an accelerometer signal based on particle filter and wavelet transform. Because high-precision accelerometers are expensive and not suitable for popular applications, it is a feasible solution to optimize the noise reduction of low-cost accelerometer signals with poor precision. The embodiment of the present invention adopts the idea of ​​combining particle filter and wavelet transform to achieve the purpose of optimizing and denoising the signal, see figure 1 , see the description below:

[0048] 101: Initialize the particles, apply the particle filter to the preprocessing part of acceleration signal denoising, retain the low entropy of wavelet denoising, describe the non-stationary characteristics of acceleration signals, and perform good denoising on nonlinear non-Gaussian systems ;

[0049] 102: According to the relationship between the observed v...

Embodiment 2

[0054] Combine below figure 1 , the specific calculation formula, the scheme in embodiment 1 is further introduced, see the following description for details:

[0055] 201: Based on the Bayesian criterion, the accelerometer can be described as a discrete dynamic system including state equations and observation equations as follows;

[0056] x k =f(x k-1 )+v k-1 (3)

[0057] z k = h k (x k )+w k (4)

[0058] Among them, formula (3) is the state equation, which is used to express the change of the state of the discrete dynamic system with time; formula (4) is the observation equation, which describes the relationship between the state and the observed quantity at a certain moment. k is the index, x k and z k are state variables and measured variables, respectively, v k and w k are process noise and measurement noise, respectively. f(·) state transition equation, h(·) is the observation equation.

[0059] 202: Initialize the particles;

[0060] sample particle ...

Embodiment 3

[0090] Combine below Figure 3-Figure 4 , and Table 1-Table 2 verifies the feasibility of the scheme in Examples 1 and 2, see the following description for details:

[0091] Quantitatively, Table 1 shows that as the number of particles increases, the root mean square error of denoising acceleration signals by this method gradually decreases. And the denoising effect of choosing the default threshold is better than that of soft threshold and hard threshold. Table 2 shows the error results of positioning the instance whose real trajectory length is 3.8m by double integrating the acceleration signal denoised by particle filter, wavelet transform and this method. The positioning error results show that the positioning accuracy of the denoising acceleration signal by this method is obviously better than other denoising methods.

[0092] When using particle filter to preprocess the acceleration signal, in order to obtain the best denoising effect, the number of particles is select...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an accelerometer signal denoising method based on particle filtering and wavelet transform. The accelerometer signal denoising method based on particle filtering and wavelet transform comprises the following steps: initializing particles, applying particle filter to a pretreatment portion of acceleration signal denoising, and retaining low entropy of wavelet denoising; calculating the weight of each particle according to the relation between an observation value and a pre-measured value, preliminary estimating the states of the particles according to the normalized weight, abandoning the particles with small weights, and overcoming degradation phenomena of the particles; transforming initial state estimation of acceleration signals to a wavelet domain, selecting a suitable wavelet basis and a suitable wavelet decomposition layer number, and carrying out wavelet decomposition on obtained initial state estimation: carrying out threshold quantifying on high-frequency coefficients of wavelet decomposition, and reconstructing state signals by various layers of coefficients of wavelet decomposition, wherein the reconstructed signals are denoised acceleration signals. By the method, noise distribution conditions of an accelerometer do not need to be considered, and the denoising degree is increased by increasing the number of particles of a particle filter device.

Description

technical field [0001] The invention relates to the fields of signal processing and autonomous positioning, in particular to an accelerometer signal denoising method based on particle filter and wavelet transform. Background technique [0002] With the improvement of people's living standards, the demand for services based on location information such as navigation and route planning is increasing year by year. Unlike outdoor positioning, there is currently no widely accepted and applied solution for indoor positioning such as GPS. Compared with technologies such as RFID that require the deployment of expensive hardware devices, inertial sensor technology that is inexpensive and easy to maintain is currently gaining more and more attention. However, the measurement error of the accelerometer signal caused by noise will increase rapidly over time, often resulting in unsatisfactory positioning accuracy, so the denoising of the accelerometer signal is essential. The accelerat...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G01C21/16G01C21/20
CPCG01C21/16G01C21/206
Inventor 徐岩于航张家赫
Owner TIANJIN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com