Loading...

Efficiency Analysis of Principal Soft Computing Techniques for Noise Cancellation

Ebrahimi, Naeim | 2016

500 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 48311 (45)
  4. University: Sharif University of Technology
  5. Department: Aerospace Engineering
  6. Advisor(s): Banazadeh, Afshin
  7. Abstract:
  8. In this thesis, LMS, NLMS, RLS, ADALINE and ANFIS as five adaptive noise removal methods have been studied with a software application approach. The aim of this study is to obtain the best method to remove noise in terms of signal to noise ratio (SNR) improvement as well as the speed of convergence. In this regard, firstly by applying a frequency sweep input, which is contaminated with white noise, the performance of these algorithms are investigated. Then, an audio signal is utilized as the target input, and the performance of the corresponding algorithms have been analyzed in aspect of filter order and learning coefficient. The results show that by increasing the order of filter and reducing the learning coefficient, the SNR improves generally while the speed of convergence of the algorithm declines. Therefore, to select the best algorithm, it requires to compromise between the SNR improvement and the learning coefficient. According to the simulation results, the LMS algorithm is the best in aspect of the speed of convergence whereas the RLS algorithm is the best in aspect of SNR improvement. Also, by considering the result of the sensitivity analysis, the RLS is chosen as the best algorithm for the current study
  9. Keywords:
  10. Noise ; Adaptive Filters ; Signal to Noise Ratio ; Neural Networks ; Noise Removing ; Efficiency

 Digital Object List

 Bookmark