Cooperative Evolution of Services in Ubiquitous Computing Environments

摘要

As the number and capabilities of mobile devices is rapidly increasing, new challenges arise in the way services are currently designed. Users are seeking for more complex and advanced functionalities able to satisfy their increasing requirements. As a consequence, in ubiquitous environments, a different way to design services has to be introduced in order to guarantee services always up to date in a transparent and efficient way. In this paper, we present and analyze a framework for distributed cooperative service evolution in a wireless nomadic environment. In particular, we assume a disconnected network architecture, where users' mobility is exploited to achieve a scalable behavior, and communication is based on localized peer-to-peer interactions among neighboring nodes. Service management is achieved by introducing autonomic services, whose operations are based on a distributed evolution process, which draws tools and concepts from evolutionary computation (and genetic algorithms in particular). The latter relies on the concept of recombination, i.e., the exchange of information among service users, which collaborate to enhance their fitness, denned as the ability of the actual service to fulfill user's requirements. We introduce a general framework for analyzing service recombination policies and exploit results from martingales theory to study their convergence properties.