Design of efficient DSP algorithms
Oscar Gustafsson and Prof. Lars Wanhammar
The aim of this project is to develop cost models for the hardware implementation at the algorithmic level that are applicable to common DSP algorithms with emphasis on digital filters, fast transforms and filter banks. These models will support development, optimization, and evaluation of algorithms that may be efficiently implemented with respect to power consumption, chip area, throughput, and to some extent design effort.
Another aim is to develop efficient digital filter algorithms based on these models, as well as adapt appropriate design tools for domain-specific applications, e.g., filters for interpolation and decimation, frequency selective filters, and Orthogonal Frequency Division Multiplex (OFDM) filter banks.
This project is funded by TFR
, Swedish Research Council for Engineering Sciences.
- O. Gustafsson and L. Wanhammar, "Implementation of maximally fast ladder wave digital filters using a numerically equivalent state-space representation," in Proc. IEEE Int. Symp. Circuits Syst., Orlando, USA, June 1999, vol. 3, pp. 419422.
- P. Löwenborg, O. Gustafsson, and L. Wanhammar, "Filter design using MATLAB," in Proc. RVK'99, Karlskrona, Sweden, June 1999, pp. 374378.
- C. Larsson, O. Gustafsson, M. Vesterbacka, and L. Wanhammar, "A tool for manual scheduling of DSP algorithms implemented in Java," in Proc. RVK'99, Karlskrona, Sweden, June 1999, pp. 367369.
- O. Gustafsson and L. Wanhammar, "Maximally fast scheduling of bit-serial lattice wave digtial filters using constrained third-order sections," in Proc. IEEE Int. Conf. Electronics Circuits Syst., Paphos, Cyprus, Sep. 1999, pp. 729732.
- O. Gustafsson, H. Johansson, and L. Wanhammar, "Design and efficient implementation of high-speed narrow-band recursive digital filters using single filter frequency masking techniques," in Proc. IEEE Int. Symp. Circuits Syst., Geneva, Switzerland, May 2000.
- O. Gustafsson, H. Johansson, and L. Wanhammar, "Design and
efficient implementation of narrow-band single frequency masking FIR
filters," in Proc. European Signal Processing Conf., Tampere,
Finland, Sept. 2000.