SPIRE: Scalable and Unified Platform for Real World IoT Services with Feature Interaction
Susumu Takeuchi, Michiharu Takemoto, Masato Matsuo · 2016
The real world services (RWSs) which affect the real world through the deployed actuators are essential in the IoT environment. These services have possibility to occur a conflict with other services, i.e., feature interaction (FI), because they have to share the same actual device and/or same field. Smart Shadow System (S3) had been proposed to provide RWSs and dynamic detection of FIs by introducing the distributed agents of real world elements. However, the S3 model lacks the scalability required to handle a lot of RWSs to satisfy the various needs in the real world. Moreover, the responsibility of each agent, the roles of stakeholders and the relationships between RWS and FI are ambiguous. Therefore, a scalable platform for intelligent and reciprocal environment (SPIRE) framework, which clarifies the roles and responsibilities of each agent and divides part of agents in the parent-child structure, is proposed in this paper. The experimental system was developed with the actual RWSs in a smart house. The preliminary experimental results show that the proposed framework is feasible and the load for RWS could be suppressed within the responsible agent that is inadequate to achieve the scalability.