Analog Multiply-Accumulate Unit With Limited Input Range

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Analog systems performing multiply-accumulate operations face challenges in detecting operation results with high accuracy due to limitations in power consumption and signal processing precision.

Innovation Solution

An arithmetic logic unit with input lines and a multiply-accumulate operation device that generates charges based on pulse signals, where at least one of the input or weight values is limited to a predetermined range, allowing for accurate detection of operation results by adjusting the sum of multiplication values.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Use of energy by moving object

If analog system is used for multiply-accumulate operation, then power consumption is reduced, but measurement precision of operation result deteriorates

Engineering Contradiction:
Improvepower consumptionVSAvoiddetection accuracy of operation result
Core Design Contradiction:
Use of energy by moving objectVSMeasurement precision

Solution Approach 1:

The patent limits the values of input signals and weight signals to specific ranges (e.g., input signals between -1 and 1, weight signals between 0 and 1). This parameter constraint ensures that the sum of multiplication values remains within a detectable range, improving measurement precision while maintaining the low power consumption benefits of analog computation.

Inventive Principle:
Principle #35Parameter changes

2Adaptability or versatility

If value range of input or weight is expanded, then computational flexibility is improved, but detection accuracy of operation result deteriorates

Engineering Contradiction:
Improvecomputational flexibilityVSAvoiddetection accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent establishes specific value ranges for input signals (-1 to 1) and weight signals (0 to 1), creating an optimal balance between computational flexibility and detection accuracy. This parameter optimization allows the system to handle various computational tasks while ensuring the sum of multiplication values remains within the high-precision detection range of the analog circuit.

Inventive Principle:
Principle #35Parameter changes

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach enables high-accuracy detection of operation results in analog systems by reducing the sum of multiplication values and improving detection precision, thereby enhancing the accuracy of multiply-accumulate operations.

Implementation Method 1

a plurality of multiplication units that generates, on the basis of pulse signals input to each of the plurality of input lines, charges corresponding to multiplication values obtained by multiplying the input values by weight values

Methodology Applied
Scientific EffectCharge generation through multiplication operation:

Implementation Method 2

an output unit that outputs a multiply-accumulate signal representing a sum of the multiplication values by accumulating the charges corresponding to the multiplication values generated by each of the plurality of multiplication units

Methodology Applied
Scientific EffectCharge accumulation: Capacitance

Data Source

PatentUS12061877B2Arithmetic logic unit, multiply-accumulate operation device, multiply-accumulate operation system, and multiply-accumulate operation method
Publication Date: 2024.08.13 SONY GROUP CORP
  • US12061877B2 patent drawing
  • US12061877B2 patent drawing
  • US12061877B2 patent drawing

AI summary

An arithmetic logic unit according to an embodiment of the present technology includes: a plurality of input lines; and a multiply-accumulate operation device. Pulse signals corresponding to input values are input to the plurality of input lines. The multiply-accumulate operation device includes a plurality of multiplication units that generates, on the basis of the pulse signals input to each of the plurality of input lines, charges corresponding to multiplication values obtained by multiplying the input values by weight values, and an output unit that outputs a multiply-accumulate signal representing a sum of the multiplication values by accumulating the charges corresponding to the multiplication values generated by each of the plurality of multiplication units. A value of at least one of the input value or the weight value is limited.