Finite Precision Error Analysis of Neural Network Hardware Implementations
IEEE Transactions on Computers
A multiplier generator for Xilinx FPGAs
VLSID '96 Proceedings of the 9th International Conference on VLSI Design: VLSI in Mobile Communication
International Journal of Reconfigurable Computing
Hi-index | 0.00 |
FPGA implementation of artificial neural networks calls for multipliers of various word length. In this paper, a new algorithm for generating variable word length multipliers for FPGA implementation is presented. The multipliers generated are based on a Booth Encoded optimized Wallace tree architecture. Several features of FPGA architecture are used to generate fast and efficient multipliers. These multipliers are shown to be 20% faster than existing FPGA multiplier implementations.