∫1-Norm Constrained Minimum Eror Entropy Algorithm
Rajni Yadav1, Chandra Shekhar Rai2, Kanika Agarwal3
1Rajni Yadav*, Department of Electronics and Communication, Maharaja Agrasen Institute of Technology, Guru Gobind Singh Indraprastha University, Delhi, India.
2Chandra Shekhar Rai,, University School of Information, & Communication Technology, Guru Gobind Singh Indraprastha University, Delhi, India.
3Kanika Agarwal, Department of Electronics and Communication, Maharaja Agrasen Institute of Technology, Guru Gobind Singh Indraprastha University, Delhi, India.
Manuscript received on January 26, 2020. | Revised Manuscript received on February 05, 2020. | Manuscript published on February 30, 2020. | PP: 2350-2355 | Volume-9 Issue-3, February 2020. | Retrieval Number: C5721029320/2020©BEIESP | DOI: 10.35940/ijeat.C5721.029320
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: This work proposes a linear phase sparse minimum error entropy adaptive filtering algorithm. The linear phase condition is obtained by considering symmetry or anti symmetry condition onto the system coefficients. The proposed work integrates linear constraint based on linear phase of the system and -norm for sparseness into minimum error entropy adaptive algorithm. The proposed -norm linear constrained minimum error entropy criterion ( -CMEE) algorithm makes use of high-order statistics, hence worthy for non-Gaussian channel noise. The experimental results obtained for linear phase sparse system identification in the presence of non-Gaussian channel noise reveal that the proposed algorithm has lower steady state error and higher convergence rate than other existing MEE variants.
Keywords: Constrained adaptive filtering, Information theory, non-Gaussian noise, sparse system