This paper proposes a novel VLSI architecture for an FIR filter chip providing variable-length taps. To change the number of taps, we propose two special features called a data-reuse structure and a recurrent-coefficient scheme. These features consist of several MUXs and registers and reduce the number of gates over 20% compared with existing chips using an address generation unit and a module unit. Since parallel multipliers occupy a large VLSI area, a filter chip using bit-serial multipliers meeting the real-time requirement specification can dramatically save the area. We propose a modified bit-serial multiplication algorithm to compute two partial products in parallel, and thus, the proposed filter can be twice faster than previous filters using bit-serial multipliers. We developed VHDL models and performed logic synthesis using the SYNOPSYS/sup TM/ CAD tool with the 0.8 /spl mu/m SOG cell library (HSG30042). The chip has only 9,507 gates, was fabricated, and is running at 77 MHz.
[1]
Sanjit K. Mitra,et al.
Handbook for Digital Signal Processing
,
1993
.
[2]
Israel Koren.
Computer arithmetic algorithms
,
1993
.
[3]
T. J. Sullivan,et al.
The design of a bit-serial coprocessor to perform multiplication and division on a massively parallel architecture
,
1988,
Proceedings., 2nd Symposium on the Frontiers of Massively Parallel Computation.
[4]
Earl E. Swartzlander.
The Quasi-Serial Multiplier
,
1973,
IEEE Transactions on Computers.
[5]
J.B.G. Roberts,et al.
Figures of merit for VLSI implementations of digital signal processing algorithms
,
1984
.
[6]
Woo Jin Oh,et al.
Implementation of programmable multiplierless FIR filters with powers-of-two coefficients
,
1995
.
[7]
Ratan K. Guha,et al.
The Two's Complement Quasi-Serial Multiplier
,
1975,
IEEE Transactions on Computers.