Embedded System Program Installation Using Pre-Generated Optimization Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In process automation technology, embedded systems require unique program codes for execution, leading to redundant development and inefficiencies due to limited computing and memory resources, which also result in high energy consumption and longer runtime.
Innovation Solution
A method for installing a program on embedded systems that involves reading a first program code and additional information, using the latter for optimized binary translation, allowing execution on multiple resource-limited systems with reduced energy consumption and faster runtime.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If optimized binary translation is performed to reduce program runtime, then execution speed is improved, but computing power and memory requirements exceed embedded system resources
Solution Approach 1:
The patent applies preliminary action by performing program analysis and generating optimization information on an external system before the program is installed on the embedded system. This pre-processing step identifies optimization opportunities and creates instructions that guide the translation process, allowing the resource-constrained embedded system to execute optimizations without performing complex analysis itself, thus reducing runtime while maintaining limited resources.
Solution Approach 2:
The patent introduces an intermediary optimization information structure that mediates between the source program and the translation process. This optimization information acts as a guide that contains pre-analyzed optimization opportunities, enabling the binary translator to apply optimizations efficiently without requiring the embedded system to perform resource-intensive analysis, thereby resolving the contradiction between execution speed and resource constraints.
2Reliability
If unique program codes are created for each embedded system to ensure proper execution, then system reliability is improved, but development time and cost increase due to redundant work
Solution Approach 1:
The patent applies universality by creating a standardized optimization information format that can be generated once for a source program and reused across multiple target embedded systems. This allows a single program analysis to produce optimization instructions that guide translation to various architectures, eliminating redundant analysis work while ensuring each system receives appropriate optimizations for correct execution.
Solution Approach 2:
The patent uses copying by generating optimization information that can be replicated and applied to multiple embedded systems. Instead of performing unique analysis for each target system, the same optimization information structure is copied and adapted for different architectures, significantly reducing development time while maintaining execution correctness through architecture-specific translation rules.
3Manufacturing precision
If embedded systems are equipped with higher computing power and memory for program analysis, then translation quality is improved, but energy consumption increases
Solution Approach 1:
The patent applies preliminary action by moving the computationally intensive program analysis to an external system before deployment. The embedded system only needs to execute the translation using pre-generated optimization information, which maintains high translation quality while avoiding the energy cost of running complex analysis on resource-constrained hardware.
Data Source
AI summary
The present disclosure relates to a method for installing a program on an embedded system of process automation technology. The method includes steps of reading a first program code into a data memory of the embedded system via a data interface of the embedded system and reading additional information specific to the first program code into the data memory of the embedded system via the data interface. The method additionally includes a step of translating the first program code by means of the embedded system into a second program code executable by the embedded system. The program-specific additional information is used for the translation so that the second program code can be executed more efficiently by the embedded system than a third program code translated from the first program code without the program-specific additional information.