Systematic design method of embedded type module cross test

Abstract

The invention relates to a systematic design method of an embedded type module cross test. The method includes the following steps that firstly, all modules in an embedded type system are classified or sorted, and a complete inter-module cross correlation matrix is built; secondly, according to the first step cross test use case reduction strategy, specific modules involving the in-pair cross and module categories are analyzed and determined to obtain an original cross test use case set; then, according to the second step cross test use case reduction strategy, analysis and reduction are further conducted on the original cross test use case set, and a preliminary cross test use case set is obtained; finally, the cross depth and intensity of each cross test use case in the preliminary cross test use case set is analyzed and selected, and a final cross test use case set is obtained. The good systematic test design guarantees the quality of the specific test implementation in the next stage, also reduces the workload of the final test implementation, and makes the 'fast and good' advantages of the cross test embodied.

Description

technical field [0001] The invention relates to a systematic design method for cross-testing of embedded modules. Background technique [0002] With the growing size and complexity of embedded software, the quality of embedded software accounts for an increasing proportion of the quality of the entire embedded system. Sex is also agreed upon. However, due to the characteristics of strong real-time performance, limited storage and computing resources, and close correlation with hardware, the traditional software testing theory may not be effective when directly applied to embedded software testing. For example, when system testing methods such as stress testing, which have been proven effective in traditional testing theory, are applied to embedded software testing, problems such as long testing time periods, low testing efficiency, and high testing costs (such as labor costs) often arise. , in the embedded software testing industry, a new testing method has emerged, which ...

AI Technical Summary

Problems solved by technology

At its root, the main reason is that poor cross-test design limits the maximum power that this method can exert, and the effect is limited

There are two specific reasons: 1. Bad cross-test design (mainly referred to as "over-design") has carried out too many cross-combination designs without analysis, resulting in a surge in the number of use cases and a large workload for specific test implementation. The test can't be "fast"; 2. Poor cross-test design (mainly refers to "random design") is likely to miss important cross-combinations, which makes the use case design incomplete. Execute, but fail to find bugs that could be found, making cross-testing look less "good" and less effective in terms of bug detection

Images