دانلود رایگان مقاله آگاهی از وضعیت در شبکه فرصت طلب برای حمایت انتقال داده برنامه سنجش شهری

Abstract

Smart cities can be seen as large-scale Cyber-Physical Systems with sensors monitoring cyber and physical indicators and with actuators dynamically changing the complex urban environment in some way. In this context, urban sensing is a new paradigm that exploits human-carried or vehicle-mounted sensors to ubiquitously collect data to provide a holistic view of the city. A challenge in this scenario is the transmission of sensed data in situations where the networking infrastructure is intermittent or unavailable. This paper outlines our research into an engine that uses opportunistic networks to support the data transmission of urban sensing applications. It applies situation awareness and computational intelligence approaches to perform routing, adaptation, and decision-making procedures. We carried out simulations within a simulated environment that showed our engine had 12% less overhead than other compared approaches.

6. Conclusion and suggestions for future work

In this paper, we have described our attempt to build an engine that employs the opportunistic networks paradigm to transmit sensed data in situations where the networking infrastructure is intermittent or unavailable. Additionally, our experiments applied situation awareness and computational intelligence approaches to make decisions about the routing of messages and adaptation decisions. The proposed engine will be used as a basis for the data transmission in the Communication component of a large-scale architecture called UrboSenti. We have also outlined our design models for the software modules and their internal components. Currently, we are working on the incorporation of new dynamic rules in Situation Manager. The preliminary results obtained from a statistical situation set are acceptable. We believe that the performance of engine will be improved when such implementation is finished. The experiments also showed that ESN is a reliable technique for prediction. It achieved an impressive predictive performance and has a low computational cost compared with all the other approaches that we had previously adopted. In addition, the results revealed that some popular opportunistic networks initiatives cannot be used “as is” in the area of urban sensing applications. Finally, the proposed engine is able to fill the gap of data transmission that was outlined in our initial problem-scenario. Moreover, this should encourage us to conduct further research into the multidisciplinary area of smart cities with the aim of improving services and applications for urban sensing. In future work, we are seeking alternative means of constructing fuzzy sets and rules “on the fly”, depending on the situation in which the node is embedded and intend to explore the application of a Deep Belief Network (DBN) or Restricted Boltzmann machines (RBMs) for the purposes of prediction.