Diffusion-Probabilistic Least Mean Square Algorithm
- PDF / 2,601,063 Bytes
- 19 Pages / 439.37 x 666.142 pts Page_size
- 41 Downloads / 207 Views
Diffusion-Probabilistic Least Mean Square Algorithm Sihai Guan1 · Chun Meng1 · Bharat Biswal1,2 Received: 17 October 2019 / Revised: 24 May 2020 / Accepted: 26 May 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract In this paper, a novel diffusion estimation algorithm is proposed from a probabilistic perspective by combining the diffusion strategy and the probabilistic least mean square (LMS) at all distributed network nodes. The proposed method, namely diffusionprobabilistic LMS (DPLMS), is more robust to the input signal and impulsive noise than previous algorithms like the diffusion sign-error LMS (DSE-LMS), diffusion robust variable step-size LMS (DRVSSLMS), and diffusion least logarithmic absolute difference (DLLAD) algorithms. Instead of minimizing the estimation error, the DPLMS algorithm is based on approximating the posterior distribution with an isotropic Gaussian distribution. In this paper, the stability of the mean estimation error and the computational complexity of the DPLMS algorithm are analyzed theoretically. Simulation experiments are conducted to explore the mean estimation error for the DPLMS algorithm with varied conditions for input signals and impulsive interferences, compared to the DSE-LMS, DRVSSLMS, and DLLAD algorithms. Both results from the theoretical analysis and simulation suggest that the DPLMS algorithm has superior performance than the DSE-LMS, DRVSSLMS, and DLLAD algorithms when estimating the unknown linear system under the changeable impulsive noise environments. Keywords Distributed network · Impulsive noise · Input signals · Probabilistic
B B B
Sihai Guan [email protected] Chun Meng [email protected] Bharat Biswal [email protected]
1
The Clinical Hospital of Chengdu Brain Science Institute, MOE Key Laboratory for Neuroinformation, Center for Information in Medicine, School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, China
2
Department of Biomedical Engineering, New Jersey Institute of Technology (NJIT), Newark, NJ, USA
Circuits, Systems, and Signal Processing
1 Introduction Over the last decade, distributed estimation has received increased attention in the multi-task network and single-task network [3, 11, 15, 16, 21, 29], especially for the parameter estimation of linear frequency modulation (LFM) signals [24], as well as heat and mass transfer [34]. In a single-task network, the same target parameters are collaboratively estimated for all network nodes, while in a multi-task network, target parameters of each network node need to be estimated separately [20]. This can be improved since diffusion is important for distributed estimation. Diffusion is a common physical phenomenon in the complex transport process. For example, Yang et al. proposed some novel methods to deal with the analytical solution in heat and mass transfer, such as fractional derivatives [33, 37] and kernel functions [35, 36]. Local cooperation and data processing are two important components of distributed d
Data Loading...
