Small Logarithmic Floating-Point Multiplier Based on FPGA and Its Application on MobileNet
- 17 October 2022
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 69 (12) , 5119-5123
Abstract
The small floating-point (SFP) multiplier proposed by Xilinx is utilized to implement the convolution neural networks (CNNs). This scheme can balance the resource usage of look-up tables (LUTs) and digital signal processing blocks (DSPs) so that high compute density is achieved on Field Programmable Gate Arrays (FPGAs). In addition, this scheme can quantize the CNNs with several simple scaling operations rather than a lengthy compute intensive retraining process. However, the mantissa field of SFP multiplier is required to be less than or equal to 3-bit, thus significantly restricts the application of this scheme. To figure out this issue, we implement the SFP multiplier in the logarithmic domain such that the multiplication is performed by the addition with the aid of logarithmic and anti-logarithmic converters that is referred to as small logarithmic floating-point (SLFP) multiplier. Compared to SFP multiplier (3-bit mantissa), the proposed SLFP multiplier can support multiple accuracy levels (3∼5-bit mantissa) with a relatively low overhead (0∼3× LUT6s). Moreover, we utilize the look-ahead carry chain to reduce the delay of addition so that the proposed SLFP multiplier can operate at 650MHz. As a result, the latency (1.5ns, 1× clock cycle) and throughput (650MOPS) of proposed SLFP multiplier (3∼5-bit mantissa) are the same as the SFP multiplier (3-bit mantissa). In the end, the implementation of MobileNet proves that the accuracy level of SFP multiplier (3-bit mantissa) is not sufficient, which can be solved by the proposed SLFP multipliers (5-bit mantissa).Keywords
Funding Information
- Fundamental Research Funds for the Central Universities (DUT20RC(3)058)
This publication has 13 references indexed in Scilit:
- Utilize the shift‐and‐add architecture to approximate multiple nonlinear functionsInternational Journal of Circuit Theory and Applications, 2021
- Circuits and Architectures for In-Memory Computing-Based Machine Learning AcceleratorsIEEE Micro, 2020
- Universal Deep Neural Network CompressionIEEE Journal of Selected Topics in Signal Processing, 2020
- Automatic Generation of Multi-Precision Multi-Arithmetic CNN Accelerators for FPGAs2019
- Retrain-Less Weight Quantization for Multiplier-Less Convolutional Neural NetworksIEEE Transactions on Circuits and Systems I: Regular Papers, 2019
- Iterative Deep Neural Network Quantization With Lipschitz ConstraintIEEE Transactions on Multimedia, 2019
- DeMAS: An efficient design methodology for building approximate adders for FPGA-based systems2018
- Double Logarithmic Arithmetic Technique for Low-Power 3-D Graphics ApplicationsIEEE Transactions on Very Large Scale Integration (VLSI) Systems, 2017
- An Improved Two-Step Binary Logarithmic Converter for FPGAsIEEE Transactions on Circuits and Systems II: Express Briefs, 2014
- A 231-MHz, 2.18-mW 32-bit Logarithmic Arithmetic Unit for Fixed-Point 3-D Graphics SystemIEEE Journal of Solid-State Circuits, 2006