Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

On Efficient Polyphase Network Implementation Using Successive Vector Approximation

Luiz F. da S. Coelho, Didier Le Ruyet, Paulo S. R. Diniz

TL;DR

This work targets energy-efficient multiplierless implementation of the FBMC polyphase network for OQAM-FBMC systems by introducing SOPOT-based vector approximations. It develops two greedy algorithms, Signed Digit Loading (SDL) and Matching Pursuits with Generalized Bit Planes (MPGBP), and provides a detailed comparison against the canonical signed-digit (CSD) approach. The results show vector-based SOPOT approximations yield significantly lower mean-squared error and interference, with substantially better OOB radiation characteristics at similar computational costs, especially for PHYDYAS prototypes. Practically, the methods enable low-cost, low-energy transceivers suitable for wideband, power-constrained deployments while maintaining BER performance across modulation scenarios.

Abstract

In this work, we explore an energy-efficient implementation of the polyphase network for a filter bank multicarrier (FBMC) system. The network is approximated using a greedy algorithm based on matching pursuits (MP) that converts the numerical representation directly from floating point to sum of signed powers of two (SOPOT), which is key for a multiplierless implementation. We compare this technique with other state-of-the-art methods for designing multiplierless hardware, and show that our technique achieves superior performance with similar computational complexity.

On Efficient Polyphase Network Implementation Using Successive Vector Approximation

TL;DR

This work targets energy-efficient multiplierless implementation of the FBMC polyphase network for OQAM-FBMC systems by introducing SOPOT-based vector approximations. It develops two greedy algorithms, Signed Digit Loading (SDL) and Matching Pursuits with Generalized Bit Planes (MPGBP), and provides a detailed comparison against the canonical signed-digit (CSD) approach. The results show vector-based SOPOT approximations yield significantly lower mean-squared error and interference, with substantially better OOB radiation characteristics at similar computational costs, especially for PHYDYAS prototypes. Practically, the methods enable low-cost, low-energy transceivers suitable for wideband, power-constrained deployments while maintaining BER performance across modulation scenarios.

Abstract

In this work, we explore an energy-efficient implementation of the polyphase network for a filter bank multicarrier (FBMC) system. The network is approximated using a greedy algorithm based on matching pursuits (MP) that converts the numerical representation directly from floating point to sum of signed powers of two (SOPOT), which is key for a multiplierless implementation. We compare this technique with other state-of-the-art methods for designing multiplierless hardware, and show that our technique achieves superior performance with similar computational complexity.
Paper Structure (11 sections, 20 equations, 7 figures, 3 algorithms)

This paper contains 11 sections, 20 equations, 7 figures, 3 algorithms.

Table of Contents

  1. Introduction
  2. System Model
  3. SOPOT Approximation
  4. Canonical Sign Digit
  5. Signed Digit Loading
  6. Matching Pursuits with Generalized Bit Planes
  7. Bit Depth Selection
  8. Codeword Selection
  9. Stopping Criteria
  10. Simulation Results
  11. Conclusion

Figures (7)

  • Figure 1: Illustration of the bounding region (in pink) for the bit plane depth allocation.
  • Figure 2: MSE for the PHYDYAS filter approximations, using the CSD, SDL, and MPGBP algorithms in different complexity levels.
  • Figure 3: Interference MSE for the PHYDYAS filter approximations, using the CSD, SDL, and MPGBP algorithms in different complexity levels.
  • Figure 4: PSD of FBMC/OQAM systems with similar complexity; $4$-bit CSD obtained using the CSD algorithm, $1.8~\textrm{SPT/Coeff.}$ obtained using the SDL algorithm, and Original as an estimate for the system using a non-approximated prototype filter.
  • Figure 5: PSD of FBMC/OQAM systems with different approximation levels, using the SDL algorithm.
  • ...and 2 more figures