PPU Adaptive LMS Algorithm, a Hardware-Efficient Approach; a Review on
Keywords:
Adaptive signal processing, Convergence speed, Partial-selective weight update, Resource usage.
Abstract
The periodic partial updates (PPU) method is evaluated. Cost, performance, portability and physical size considerations compel serious resource limitations on adaptive signal processing systems. Partial-Selective weight updates techniques can be used to decrease the resource usage and hardware complication in practical fields at the likely price of greater steady-state MSE error and more lower convergence speed. Periodic partial update (PPU) is one of partial-selective weight update techniques by which, on behalf of updating the all weights (coefficients), a limit number of the weights are updated at each epoch. The performance, convergence speed and MSE error of periodic partial updates is evaluated in the presence of white and colored Gaussian input. It is concluded that in some practical fields, this method could be considered on behalf of full-update algorithm with some loses on quality.References
[1] O. Sharifi-Tehrani, “Novel hardware-efficient design of LMS-based adaptive FIR filter utilizing finite state machine and Block-RAM,” PRZEGLAD ELEKTROTECHNICZNY (Electrical Review) J., vol. 2011, no. 7, pp. 240-244, July 2011.
[2] O. Sharifi-Tehrani, M. Ashourian, “An FPGA-based implementation of ADALINE neural network with low resource utilization and fast convergence,” PRZEGLAD ELEKTROTECHNICZNY (Electrical Review) J., vol. 2010, no. 12, pp. 288-292, December 2010.
[3] O. Sharifi-Tehrani, M. Ashourian, P. Moallem, “An FPGA-based implementation of fixed-point standard-LMS algorithm with low resource utilization and fast convergence,” Inter. Rev. on Comp. and Soft. (IReCOS) J., vol. 5, no. 4, pp. 436-444, July 2010.
[4] E. Ghafarioun, M. Ashourian, H. Mahdavi-nasab, O. Sharifi-Tehrani, “Partial-update adaptive LMS algorithms; design, analysis and comparison,” Inter. Rev. on Comp. and Soft. (IReCOS) J., vol. 6, no. 3, May 2011.
[5] K. Dogancay, “Partial Update Adaptive Signal Processing,” Elsevier, Oxford, 2008.
[6] T. Adali, S. Haykin, “Adaptive Signal Processing Next-Generation Solutions,” New York : Wiley, 2010.
[2] O. Sharifi-Tehrani, M. Ashourian, “An FPGA-based implementation of ADALINE neural network with low resource utilization and fast convergence,” PRZEGLAD ELEKTROTECHNICZNY (Electrical Review) J., vol. 2010, no. 12, pp. 288-292, December 2010.
[3] O. Sharifi-Tehrani, M. Ashourian, P. Moallem, “An FPGA-based implementation of fixed-point standard-LMS algorithm with low resource utilization and fast convergence,” Inter. Rev. on Comp. and Soft. (IReCOS) J., vol. 5, no. 4, pp. 436-444, July 2010.
[4] E. Ghafarioun, M. Ashourian, H. Mahdavi-nasab, O. Sharifi-Tehrani, “Partial-update adaptive LMS algorithms; design, analysis and comparison,” Inter. Rev. on Comp. and Soft. (IReCOS) J., vol. 6, no. 3, May 2011.
[5] K. Dogancay, “Partial Update Adaptive Signal Processing,” Elsevier, Oxford, 2008.
[6] T. Adali, S. Haykin, “Adaptive Signal Processing Next-Generation Solutions,” New York : Wiley, 2010.
Published
2017-06-15
How to Cite
Sharifi-Tehrani, O., & Talati, S. (2017). PPU Adaptive LMS Algorithm, a Hardware-Efficient Approach; a Review on. Majlesi Journal of Mechatronic Systems, 6(1). Retrieved from https://ms.majlesi.info/index.php/ms/article/view/312
Issue
Section
Articles
