Single-Multiplier Carry Compensation for Multi-Precision Products

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing multiplier designs require combining multiple low-precision multipliers to achieve high-precision multiplication operations, leading to significant hardware resource consumption and area overhead.

Innovation Solution

An operation method and apparatus that determine a carry compensation term for low-order input data sets, incorporating it into a target partial product array to cancel out carries during accumulation, allowing for accurate high-order product operations using a single multiplier, thereby reducing hardware resource consumption and area.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple low-precision multipliers are combined to achieve high-precision multiplication operations, then multiplication precision is improved, but hardware resource consumption and area overhead increase significantly

Engineering Contradiction:
Improvemultiplication precisionVSAvoidhardware area
Core Design Contradiction:
Measurement precisionVSArea of stationary object

Solution Approach 1:

The patent segments the multiplication operation into multiple precision levels (low-precision and high-precision modes) that can be selectively executed by a single multiplier. The multiplier is divided into functional components that can operate independently or in combination, allowing one multiplier to replace multiple multipliers through intelligent segmentation of the computation process.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the operational parameters of the multiplier by dynamically adjusting the precision mode based on computational requirements. By modifying parameters such as the number of bits processed and the accumulation strategy, a single multiplier can adapt to perform both low-precision and high-precision multiplication operations, eliminating the need for multiple dedicated multipliers.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If multiple low-precision multipliers are combined to achieve high-precision multiplication operations, then multiplication precision is improved, but hardware resource consumption increases

Engineering Contradiction:
Improvemultiplication precisionVSAvoidhardware resource consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent makes the multiplier universal by enabling it to perform multiple functions: low-precision multiplication, high-precision multiplication, and carry compensation operations. This multi-functionality allows a single multiplier to replace multiple specialized multipliers, reducing overall hardware resource consumption while maintaining the capability to achieve high-precision results through coordinated operation.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent merges multiple multiplication operations into a single unified computation process. By combining low-precision and high-precision operations within one multiplier, along with integrating carry compensation mechanisms, the system achieves high-precision multiplication without requiring separate hardware resources for each operation type, thereby reducing total hardware consumption.

Inventive Principle:
Principle #5Merging (Combining)

3Area of stationary object

If a single multiplier is used with carry compensation to achieve multiple precision operations, then hardware area is reduced, but computational complexity increases

Engineering Contradiction:
Improvehardware areaVSAvoidcomputational complexity
Core Design Contradiction:
Area of stationary objectVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by pre-calculating and storing carry compensation values in lookup tables before the main multiplication operation. This allows the complex carry compensation process to be simplified during execution, as the compensation values are already prepared and can be directly applied without performing complex calculations in real-time, thus reducing the perceived computational complexity during operation.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces carry compensation terms as intermediary elements that mediate between low-precision and high-precision multiplication results. These intermediaries simplify the overall computational process by breaking down the complex high-precision multiplication into manageable steps: low-precision multiplication, carry calculation, compensation term application, and final result synthesis, making the computational complexity more tractable.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4455859A1Operation method of multiplier, operation apparatus, electronic device, and storage medium
Publication Date: 2024.10.30 BEIJING HORIZON INFORMATION TECH CO LTD
  • EP4455859A1 patent drawingFigure 1~3
  • EP4455859A1 patent drawingFigure 4
  • EP4455859A1 patent drawingFigure 5

AI summary

Disclosed are an operation method of multiplier, operation apparatus, electronic device, and storage medium. The method includes: determining a plurality of input data sets of the multiplier and an encoding manner for the multiplier; determining at least one low-order input data set in the plurality of input data sets; determining a carry compensation term corresponding to the at least one low-order input data set based on the at least one low-order input data set and the encoding manner; determining a target partial product array based on the carry compensation term corresponding to the at least one low-order input data set and the plurality of input data sets; and determining a product operation result for each input data set based on the target partial product array. According to this disclosure, multiplication operations with multiple precision may be implemented by using one multiplier, thereby reducing hardware resource consumption and hardware area.