Applying Behaviour Impact Analysis to Predict Complexity Variations in Embedded Systems Using Open-Source Software

Abstract

This paper investigates the application of Behaviour Impact Analysis (BIA) to predict complexity variations in embedded systems resulting from the integration of open-source software (OSS). Embedded systems are characterized by resource constraints, making OSS integration a double-edged sword—offering enhanced functionality but potentially increasing system complexity. The study employs empirical data collection and preprocessing to establish baseline metrics before OSS integration. Behavioural metrics such as execution time and memory usage are analyzed using BIA methodologies to quantify their impact on system complexity. Predictive models are developed using machine learning techniques to forecast complexity variations post-integration. The findings validate the efficacy of BIA in anticipating changes and provide insights for developers to manage system complexities effectively.