- All Journals
- Al-Qadisiyah University
- Basrah University
- Kirkuk University
- Mosul University
- The Iraqi Borad for Medical Specialization
- University of Anbar
- University of Technology
- Al Mustafa University
- Al-Zahraa University for Women
- Alnoor University
- Kunooz University
- Misan University
- Ninevah University
- Tikrit University
- University of Al-Hamdaniya
- University of Kerbala
A New Signal De-Noising Method UsingAdaptive Wavelet Threshold based on PSO Algorithm and Kurtosis Measuring for Residual Noise | ||
| Journal of University of Babylon | ||
| Article 1, Volume 25, Issue 1, 2017, Pages 8-19 | ||
| Author | ||
| Sameer Abdul Kadhim | ||
| Abstract | ||
| The signal de-noising based on waveletthresholdingis subjected to the value of threshold and how the way selection for it. Thismadea threshold value acts as an oracle which distinguishes between noise and signal.To date, there havebeen several methodsdeveloped to predictthe value of thresholddepending on statistical calculationsfor the noisy signalassuming that there is some priori knowledge for original signal and noise distributions.In fact, in any practical issues, only theobserved noisy signal that wehold.Therefore, in this work, an intelligent modelis developed to estimate the value of thresholdwithout any priority of knowledge for these distributions. This is done by implementingthe Particle Swarm Optimization (PSO) algorithm for kurtosis measuring of the residual noise signal to find an optimum threshold value at which the kurtosis function be maximum. These residual noise signal can be estimated by applying an inverse threshold function to the detail coefficients of the DWT. This model has been validated by comparison the results with the statisticalmodels and shows strong agreement for the obtained threshold parameter. Computer simulation for proposed model was implemented using MATLAB2011 as follows:At first, the kurtosis measuring for residual noise was analyzedusing three different signals with four SNR levels. Through this step we found that: there's a single value for the threshold that maximizes the kurtosis of residual noise. Then, the PSO algorithm was implemented to find this optimumvalue.At the same time, it's noticed that the PSO algorithm with ten swarmprovidesa convergence speed about (20~ 30)iterationfor any signal distribution at any SNR level and for each decomposition level.Finally, the mean square error (MSE) was used to evaluate the performance oftwo and five decomposition levels for each tested signal. | ||
| Keywords | ||
| Signal de; Noising; inverse threshold; residual noise; PSO; kurtosis | ||
|
Statistics Article View: 90 PDF Download: 36 |
||