article · IEEE Signal Processing Letters · 2025
Optimized Multistage Decimation Based on Optimal Factorization of Decimation Ratio
Cited by 0Abstract and details only
Abstract
In this paper, we analyze a method for designing an optimized multistage decimation filter chain focused on minimizing computational load. Decimation ratios for each of the K stages are determined to achieve the optimal number of filter multiplications per second. To enable the optimal factorization of non-integer or prime decimation ratios, the first stage of the decimation chain is designed to perform non-integer decimation using a transposed modified Farrow structure. We evaluate the performance of the optimized multistage filter chain with varying numbers of stages and compare it to an optimal FIR realization. The results demonstrate that the Farrow structure significantly reduces the computational load for prime and non-integer decimation cases and benefits non-prime cases.
Cite this paper
Babić, V., & Babić, D. (2025). Optimized Multistage Decimation Based on Optimal Factorization of Decimation Ratio. IEEE Signal Processing Letters, 32, 3142–3146. https://doi.org/10.1109/lsp.2025.3595512
Babić, Vesna, and Djordje Babić. “Optimized Multistage Decimation Based on Optimal Factorization of Decimation Ratio.” IEEE Signal Processing Letters, vol. 32, 2025, pp. 3142–46, https://doi.org/10.1109/lsp.2025.3595512.
Babić, Vesna, and Djordje Babić. 2025. “Optimized Multistage Decimation Based on Optimal Factorization of Decimation Ratio.” IEEE Signal Processing Letters 32: 3142–46. https://doi.org/10.1109/lsp.2025.3595512.
Babić, V. and Babić, D. (2025) ‘Optimized Multistage Decimation Based on Optimal Factorization of Decimation Ratio’, IEEE Signal Processing Letters, 32, pp. 3142–3146. Available at: https://doi.org/10.1109/lsp.2025.3595512.
@article{babic2025optimized,
title = {{Optimized Multistage Decimation Based on Optimal Factorization of Decimation Ratio}},
author = {Babić, Vesna and Babić, Djordje},
journal = {IEEE Signal Processing Letters},
volume = {32},
pages = {3142--3146},
year = {2025},
doi = {10.1109/lsp.2025.3595512}
}Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.