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.

A 32-Bit Delay-Insensitive Asynchronous Pipelined Multiplier

High Quality Multiplier Design Based on Improved Circuits

A Low Latency High Throughput Multiply-accumulator Unit for Float Point and Integer

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

Design of low-power and high-speed multiplier

8×8 Bit Pipelined Dadda Multiplier in CMOS

A standard cell automated layout for a CMOS 50 MHz 8*8 bit pipelined parallel Dadda multiplier

Energy-delay Efficient Asynchronous-Logic 16&Amp;#x00d7;16-Bit Pipelined Multiplier Based on Sense Amplifier-Based Pass Transistor Logic

An 8-Bit Radix-4 Non-Volatile Parallel Multiplier

Design of a CPL-Based Pipeline Multiplier Accumulator for 32-Bit CPU

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

Design of Radix16 Booth Pipeline Multiplier

Research on Highly Effective Multiplier/divider

Research of High-Speed Pipelined Floating-Point Multipfier Design

Integrated Algorithmic Logical and Physical Design of Integer Multiplier

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

A Fully Pipelined Reconfigurable Montgomery Modular Multiplier Supporting Variable Bit-Widths

A 0.35v 1.3pj/cycle 20mhz 8-Bit 8-Tap Fir Core Based on Wide-Pulsed-Latch Pipelines

Low-Power, High-Speed, and Area-Efficient Multiplier Based On the PTL Logic Style

High Speed Multiplier Algorithm Based on Four-Tree Structure