Abstract:Multiplications have been commonly conducted in quantized CNNs, filters, and reconfigurable cores, and so on, which are widely deployed in mobile and embedded applications. Most multipliers are designed to perform multiplications with symmetric bit-widths, i.e., n - by n -bit multiplication. Such features would cause extra area overhead and performance loss when m - by n -bit multiplications ( m > n ) are deployed in the same hardware design, resulting in inefficient multiplication operations. It is highly desired and challenging to propose a reconfigurable multiplier design to accommodate operands with both symmetric and asymmetric bit-widths. In this work, we propose a reconfigurable approximate multiplier to support multiplications at various precisions, i.e., bit-widths. Unlike prior works of approximate adders assuming a uniform weight distribution with bit-wise independence, scenarios like a quantized CNN may have a centralized weight distribution and hence follow a Gaussian-like distribution with correlated adjacent bits. Thus, a new block-based approximate adder is also proposed as part of the multiplier to ensure energy-efficient operation with an awareness of the bit-wise correlation. Our experimental results show that the proposed approximate adder significantly reduces the error rate by 76% to 98% over a state-of-the-art approximate adder for Gaussian-like distribution scenarios. Evaluation results show that the proposed multiplier is 19% faster and 22% more power saving than a Xilinx multiplier IP at the same bit precision and achieves a 23.94-dB peak signal-to-noise ratio, which is comparable to the accurate one of 24.10 dB when deployed in a Gaussian filter for image processing tasks.

Study of 8-Bit Booth Asynchronous Multiplier Based on Two-Phase Handshake Protocol with Bounded Bundle Data

High Quality Multiplier Design Based on Improved Circuits

A 64X64-Bit Modified Booth Multiplier Utilizing Multiplexer-Select Booth Encoder

A Reconfigurable Multiplier for Signed Multiplications with Asymmetric Bit-Widths.

Design of a fast multiplier used in 32Kb/s ADPCM

A Hybrid Multiplier Architecture Using Partially Redundant Booth Algorithm

Asic Implementation of 12-Bit Radix-8 Booth Multiplier

Booth Multiplier Design Using Modified Square Root Carry-Select-Adder

Implementation of Booth Multiplier Algorithm using Radix-4 in FPGA

Design of low-power and high-speed multiplier

Implementation and Comparison of Radix-8 Booth Multiplier by using 32-bit Parallel prefix adders for High Speed Arithmetic Applications

Design of Radix16 Booth Pipeline Multiplier

Approximate Radix-8 Booth Multipliers for Low-Power and High-Performance Operation

Algorithm and Circuit Design for Secondary Booth Encoded Multiplier

Design and Implementation of Radix 8 Booth Encoding Multiplier for Low Area and High-Speed Applications

VHDL implementation of 16x16 multiplier using pipelined 16x8 modified Radix-4 booth multiplier

Montgomery Multiplier Based on Secondary Booth Encoded Algorithm

Multiplier Design Addressing Area-Delay Trade-offs by using DSP and Logic resources on FPGAs

AxBMs: Approximate Radix-8 Booth Multipliers for High-Performance FPGA-Based Accelerators