An Approach Based on Situation Analysis and Separation of Environment Perception to Implement Adaptive Strategies

Abstract

In the existing research, most software adaptation is triggered by changes in the environment. However, the environment of a software system is characterized by uncertainty and complexity. For a software system, the same situation is reflected by different kinds of data in different environments. Unfortunately, traditional adaptive strategies can only be applied to specific environments. When the environment changes, the adaptive strategy also changes accordingly. At the same time, the existing adaptive strategies lack intuition. Therefore, the problem to be solved is how to simplify the specification and implementation of adaptive logic, while ensuring that the software system can adapt to a variety of deployment environments, and also tolerate the uncertainty in the environment. To solve the above problems, we propose our method, an adaptive strategy implementation method based on situation analysis and separation of environment perception. Our method is implemented by the MAPE-K loop, we consider the similar environmental states as one kind of situation in general, and then the adaptive strategy in this situation is defined. It separates the specific environmental data from the adaptive logic of how the software adapts to different situations. Besides, this paper also provides a case study - Intelligent Medical Assistance System to evaluate the feasibility and effectiveness of our method.