دانلود رایگان مقاله سیستم سایبر فیزیکی برای موبایل های فرصت طلب شبکه در پروکسی میتی (MNP)

The pervasive presence of mobile personal devices like smartphones, tablets, and similar smart devices, together with the massive use of online social networking services (e.g., Facebook, Twitter, Google Plus, LinkedIn, etc.) are increasingly creating a cyberphysical space where users can interact exploiting and generating information. Enriched with several sensing capabilities and networking interfaces, today’s portable devices are enabling new ways of communication including Mobile Networking in Proximity (MNP) [1]. This network mode complements the classic scenario with Internet coverage by enlarging the range of functionalities of these devices through short-range communications (e.g., Bluetooth, Wi-Fi Direct, etc.). Even with the wide coverage of the Internet, there are still some situations where the traditional network is overloaded, unavailable, or too expensive. In such situations, when two devices are in proximity, they could still share information by just exploiting the meeting opportunities and activating an opportunistic hop-by-hop forwarding towards the destination.

Mobility modeling and management

Node mobility is fundamental for MNP: the more people move, the more the messages spread through the networks. As such, mobility modeling and management are the basis of dissemination choices. Mobility modeling in particular, requires accurate knowledge of human behavior. Recent models [4] consider, for example, both (online and offline) social and spatial factors. However, the techniques used to capture such factors, are usually able to capture either positions, or contacts, with a limited accuracy. In “SOUK: Spatial Observation of hUman Kinetics”, Killijian et al. propose a method to gather positions and orientation of mobile nodes even in a crowd, with great accuracy. The developed open-source software pipeline is able to extract several metrics on movement like position and orientation, and social contacts allowing the analysis of their relationships and with the long-term goal of refining mobility models or deriving new ones. For demonstrating the accuracy and the validity of such an approach, the motion of 50 individuals was traced through the SOUK platform and compared against random way-point models having the same global characteristics. Results demonstrate that group- and interaction-based models are required to fully exploit the benefits of MNP since human kinetics cannot be accurately predicted by a random way point model