Semantic-based resource discovery, composition and substitution in IEEE 802.11 mobile ad hoc networks

Abstract

We present a general framework for resource discovery, composition and substitution in mobile ad-hoc networks, exploiting knowledge representation techniques. Key points of the proposed approach are: (1) reuse of discovery information at network layer in order to build a fully unified semantic-based discovery and routing framework; (2) use of semantic annotations in order to perform the orchestration of elementary resources for building personalized services adopting a concept covering procedure, and to allow the automatic substitution of no more suitable/available components. Using ns-2 simulator, we evaluated performances of the proposed framework with reference to a disaster recovery scenario. In particular, the impact of the number of available services and active clients has been investigated in various mobility conditions and for several service covering threshold levels. Obtained results show that: (1) the proposed framework is highly scalable, given that its overall performance is improved by increasing the number of active clients; (2) the traffic load due to clients is negligible; (3) also for a very small number of available service providers very high hit ratios can be reached; (4) increasing the number of providers can lead to hit ratios very close to 100% at the expense of an increased traffic load. Finally, the effectiveness of cross-layer interaction between routing and resource discovery protocols has been also evaluated and discussed.

An agent based network resource planner for workflow applications

“…In the experiment, we have demonstrated that the QoSAWF schema is capable of describing the interaction scenarios required by the CineGrid use case. Compared to the network resource discovery in the Mobile Ad-hoc NETworks (MANETs) [44], our approach more focuses on the circuit network and on the large data movements, although both work share changing network routings behaviour as the common objective.…”

A Framework for Query Processing over Compressed Knowledge Bases

“…In some approaches [2], [4], resources are pre-selected by matching only the reference ontology identifier and ancillary extra-logical, data-oriented attributes w.r.t. the request.…”

“…Therefore, we deem that avoiding full decompression is important to improve efficiency of on-the-fly query processing. Finally, a suitable communication protocol (e.g., extending the ones in [2], [4]) is needed to support semantic requests and replies among MANET hosts, but that is outside the scope of the paper.…”

“…The example task is to find a unit equipped with an "aid kit" and "sensors" able to measure "weather" conditions. According to a modified version of the ontology developed in [4] in order to satisfy the simple TBox condition (not reported here due to lack of …”

IT and Optical Network Orchestration Framework: An Industrial Research Project

“…Given a network goal and a set of available services/resources both described using a subset of logic-endowed languages, it can be possible to carry out discovery and composition simple components covering as much as possible the needs of the system. Moreover, to increase the fault tolerance, the composition protocol integrates an automatic substitutability among resource components [9]. In this context one of main goal of the project is to prove that it is possible to apply in real systems the above described technologies, improving at the same time obtained performances.…”