Article Preview
TopIntroduction
The technology of wireless sensor networks has completed a decade or so. In these years researchers, deployers, application programmers have put tremendous efforts to make these new technologies reach to the beneficiaries. The benefits of sensor networks are well realized and many deployments are taking place to study the serious problems of environmental monitoring, habitat monitoring, healthcare, production system, inventory management etc. However, for the solutions to some gross problems like global warming or pollution etc. we need to collect, understand the data as well as its patterns in totality and wide scale. This will not be possible unless diverse physical WSN collaborate. The cloud computing comes as a possible answer to the problem. To achieve this task there is an urgent need to rethink about the present architectures orientation and seclusion of wireless networks. Further, there is a need to change the perception of WSN organisation.
Our earlier work described in this paper, prepared the background in the direction of this new structure. In this paper the problem pertaining to the current state of WSNs architecture is identified, the need of new architecture supporting cloud computing on WSN is emphasized and a solution comprising of a framework having Service stack divided in four major layers, each layer supporting various services and repositories, is proposed. Functioning of the architecture and interaction sequences between major players like WSN, Subscribers and Other clouds are also described. The new architecture uses the features of cloud like virtualisation and Service oriented architecture to deal with heterogeneity.
The paper is structured as follows: related work on various architectures that focus on data sharing and limited WSN collaboration and limitations of current WSN architectures are presented. The path toward collaborative model is also introduced. The model with service stack for WSN cloud is described and the need of SOA approach for achieving heterogeneity is stated. The functioning of WSN cloud is also shown. Finally the need of WSN cloud to solve the global problem is emphasized.