Envirommental noise classification for context-aware applications

Context-awareness is essential to the development of adaptive information systems. Much work has been done on developing technologies and systems that are aware of absolute location in space and time; other aspects of context have been relatively neglected. We describe our approach to automatically sensing and recognising environmental noise as a contextual cue for context-aware applications. Environmental noise provides much valuable information about a user's current context. This paper describes an approach to classifying the noise context in the typical environments of our daily life, such as the office, car and city street. In this paper we present our hidden Markov model based noise classifier. We describe the architecture of our system, the experimental results, and discuss the open issues in environmental noise classification for mobile computing.