KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Settings

Access type: UnknownSubject: search.filters.subject.Diffusion strategy

Search Results

Now showing 1 - 5 of 5
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Multitask Diffusion Adaptive Tuning Affine Projection Algorithm for Distributed Model
    (2025-01-01)
    Sitjongsataporn, Suchada
    ;
    Wiangtong, Theerayod
    This paper presents combine-then-adapt diffusion model on affine projection adaptive algorithm with adaptive tuning mechanism. An adaptive affine projection algorithm is suggested in the distributed network. Diffusion strategy involves with the combination stage and adaptation stage. By adding the extra-stage in the combination stage, the combination between nodes and inside nodes are built with the global cost function for multitask model to eliminate unknown data. As dealt with fast convergence, the adaptive tuning parameters are assigned in terms of step-size value. Experimental results reveal that this proposed algorithm is able to contribute the approving results for system identifications.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Diffusion recursive least squares algorithm based on triangular decomposition
    (2023-10-01)
    Prongnuch, Sethakarn
    ;
    Sitjongsataporn, Suchada
    ;
    Wiangtong, Theerayod
    In this paper, diffusion strategies used by QR-decomposition based on recursive least squares algorithm (DQR-RLS) and the sign version of DQR-RLS algorithm (DQR-sRLS) are introduced for distributed networks. In terms of the QR-decomposition method and Cholesky factorization, a modified Kalman vector is given adaptively with the help of unitary rotation that can decrease the complexity from inverse autocorrelation matrix to vector. According to the diffusion strategies, combine-then-adapt (CTA) and adapt-then-combine (ATC) based on DQR-RLS and DQR-sRLS algorithms are proposed with the combination and adaptation steps. To minimize the cost function, diffused versions of CTA-DQR-RLS, ATC-DQR-RLS, CTA-DQR-sRLS and ATC-DiQR-sRLS algorithms are compared. Simulation results depict that the proposed DQR-RLS-based and DQR-sRLS-based algorithms can clearly achieve the better performance than the standard combine-then-adapt-diffusion RLS (CTA-DRLS) and ATC-DRLS mechanisms.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Diffusion Spline Adaptive Filtering with Adaptive Step-size Normalised Least Mean Square Algorithm
    (2023-01-01)
    Sitjongsataporn, Suchada
    ;
    Prongnuch, Sethakarn
    ;
    Wiangtong, Theerayod
    Diffusion adaptation on spline adaptive filtering (SAF)is presented with combine-then-adapt (CTA) strategy for each node. SAF consists of an adaptive linear filtering and a spline interpolation function. Normalised least mean square algorithm is furnished in the adaptive linear filtering part. An adaptive averaging step-size mechanism is applied for both tap-weight vector of linear and nonlinear filtering parts to provide the fast convergence with low computation complexity. Statistical results testify that the proposed diffusion algorithm is able to provide promising and competitive results to the conventional diffusion strategy algorithm.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Diffusion Mixed-Node Recursive Least Square-based Algorithm over Adaptive Network
    (2022-01-01)
    Sitjongsataporn, Suchada
    ;
    Prongnuch, Sethakarn
    ;
    Wiangtong, Theerayod
    This paper proposes a diffusion framework related on the mixed-node constraint based on recursive least squares (RLS) algorithm with adaptive network. The proposed Combine-Then-Adapt and Adapt-Then-Combine diffusion algorithms based on RLS algorithm are derived shortly by minimising the mixed-node least square-based criterion for distributed estimation over the adaptive network. The mixed-node cost function is described and illustrated briefly. Experimental results depict that these proposed algorithms is able to achieve the fast convergence better than the conventional algorithms.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Diffusion Affine Projection Sign Algorithm based on QR-Decomposition
    (2021-03-10)
    Sitjongsataporn, Suchada
    ;
    Prongnuch, Sethakarn
    ;
    Wiangtong, Theerayod
    In this paper, an Adapt-Then-Combine diffusion strategy on affine projection sign algorithm is presented. Based on the QR-decomposition method, the modified inverse autocorrelation matrix is shortly brief to decrease the computational complexity. The Adapt-Then-Combine diffusion strategy is applied on the low complexity of sign algorithm version of inverse square-root affine projection algorithm including with the adaptation and combination steps. By applying an idea of diffusion adaptation algorithm, we derive a diffused version of an affine projection sign algorithm based on QR-Decomposition, denoted as DiQR-APSA. Experimental results show that the proposed DiQR-APSA can achieve the good performance in comparison with the standard affine projection sign algorithm.