Service Discovery Protocol Interoperability in the Mobile Environment

Abstract

Abstract. The emergence of portable computers and wireless technologies has introduced new challenges for middleware. Mobility brings new requirements and is becoming a key characteristic. Mobile devices may move around different areas and have to interact with different types of networks, services and may be exposed to new communication paradigms. Thus, mobile distributed systems need to dynamically detect and adapt their interaction protocols to interoperate with services available in the environment. As a result, middleware for mobile devices must overcome two heterogeneity issues to provide interoperability in the mobile environment, i.e, heterogeneity of discovery protocols and of interaction protocols between services. Whereas adaptation techniques from reflective middleware are suitable for the latter, it is more problematic for the former if both issues are addressed concurrently. Specifically, reflective mechanisms consume too many resources like bandwidth, memory and CPU, which are limited on the mobile devices. This paper first highlights why current solutions to interoperability fail to realize service discovery protocol interoperability with both high performance and low resource consumption. Second, this paper addresses this open issue by using software architecture concepts enhanced with event-based parsing techniques to provide efficient, lightweight and flexible mechanisms to bring full service discovery interoperability to any existing mobile platform.

Semantics-Aware Services for the Mobile Computing Environment

“…Specifically, we have studied service discovery protocol interoperability in the open mobile environment [12]. This solution employs a mapping of several standard service discovery protocols (SLP [7], UPnP 11 , Jini [8]) on semantic events.…”

An SCA-Based Middleware Platform for Mobile Devices

“…Several middleware approaches have been proposed to deal with pervasive environments. In [2] authors propose a solution to support SDP interoperability using event-based parsing. Each SDP has associated a parser, that translates the messages in events, and a composer, that makes the reverse process.…”

MaLM: Machine Learning Middleware to Tackle Ontology Heterogeneity

“…The few works that specifically tackle the issue can be classified in one of two categories: static mapping and dynamic mapping. Approaches belonging to the former category (e.g., [5]) make the assumption that the various ontologies are known a priori, so that static mappings between the various ontologies can be created off-line and injected into the device middleware; no dynamic learning occurs so that, if a device is encountered that speaks a language other than those already known, it is inevitably excluded from interactions. Off-line translations between the new language and each of the other languages must be created and injected into the middleware, inevitably leading to a scalability problem.…”