Instruction Execution Circuit Dual Memory Power Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Instruction execution circuits in electronic devices that handle digital sound data face high power consumption due to prolonged latency times during decoding, especially when dealing with varying data frame sizes, leading to inefficient energy usage in devices like portable music players and IC recorders.

Innovation Solution

The implementation of a dual-memory system where loop processing instructions are stored in a register with lower power consumption, allowing only the register to be accessed during repetitive 'NOP' and 'JUMP WAIT' executions, thereby reducing overall power usage without halting the clock generation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the processor repeatedly executes NOP and JUMP WAIT instructions during latency time, then the processor can wait for reproduction completion, but power consumption increases due to continuous memory access

Engineering Contradiction:
Improveprocessing correctnessVSAvoidpower consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent segments the program storage into two distinct memory elements: a first memory element storing the main program and a second memory element storing loop processing instructions. This segmentation allows the system to access only the small second memory element during latency periods, avoiding continuous access to the larger first memory element and thereby reducing power consumption while maintaining processing correctness.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements partial action by storing only the essential loop processing instructions (NOP and JUMP WAIT) in the second memory element rather than the entire program. This partial storage is sufficient to handle latency waiting, eliminating unnecessary memory access operations and reducing power consumption during idle periods.

Inventive Principle:
Principle #16Partial or excessive action

2Manufacturing precision

If the processor waits for reproduction completion before next decoding, then data processing accuracy is maintained, but latency time increases causing inefficient energy usage

Engineering Contradiction:
Improvedata processing accuracyVSAvoidlatency time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-storing the loop processing instructions in the second memory element before they are needed. This allows the processor to immediately execute the waiting loop without accessing the main memory during latency periods, thereby reducing latency time while maintaining data processing accuracy through proper synchronization.

Inventive Principle:
Principle #10Preliminary action

3Ease of operation

If the processor accesses the first memory element continuously, then all instructions can be retrieved, but power consumption increases during loop processing

Engineering Contradiction:
Improveinstruction retrievalVSAvoidpower consumption
Core Design Contradiction:
Ease of operationVSUse of energy by moving object

Solution Approach 1:

The patent extracts the frequently accessed loop processing instructions from the first memory element and places them in the smaller second memory element. This extraction eliminates the need for continuous access to the power-hungry first memory element during loop execution, significantly reducing power consumption while maintaining ease of instruction retrieval for the critical waiting operations.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS9075620B2Instruction execution circuit
Publication Date: 2015.07.07 SEMICON COMPONENTS IND LLC
  • US9075620B2 patent drawing
  • US9075620B2 patent drawing
  • US9075620B2 patent drawing

AI summary

An instruction execution circuit includes: a memory circuit including a first memory element and a second memory element configured to require less power than the first memory element; a processor; and an address decoder configured to output an enable signal to either one, storing an instruction, of the first memory element and the second memory element when an address is outputted from the processor, the enable signal corresponding to a signal to output the instruction stored at the address, one portion of a program stored in the first memory element corresponding to a portion in which processing other than loop processing is described, the loop processing causing the processor to execute the specific instruction in a repetitive manner, the other portion of a program stored in the second memory element corresponding to a portion in which the loop processing is described.