Abstract-Data replication and in-network storage are two basic principles of the Information Centric Networking (ICN) framework in which caches spread out in the network can be used to store the most popular contents. This work shows how one of the ICN architectures, the Named Data Networking (NDN), with content pre-fetching can maximize the probability that a user retrieves the desired content in a Vehicle-to-Infrastructure scenario. We give an ILP formulation of the problem of optimally distributing content in the network nodes while accounting for the available storage capacity and the available link capacity. The optimization framework is then leveraged to evaluate the impact on content retrievability of topology-and network-related parameters as the number and mobility models of moving users, the size of the content catalog and the location of the available caches. Moreover, we show how the proposed model can be modified to find the minimum storage occupancy to achieve a given content retrievability level. The results obtained from the optimization model are finally validated against a Name Data Networking architecture through simulations in ndnSIM.

No abstract

Abstract-The Information Centric Networking (ICN) model relies on the ubiquitous use of caching to improve performance and reduce bandwidth requirements. ICN also makes it possible for routers to fetch content from downstream nodes, such as when a content for a home user is fetched from a neighbor's home router, with significant performance improvement. This paper shows how an attacker using compromised hosts can easily gather a massive amount of low-cost, low-latency storage for malware, junk, and other attacker-controlled content. We conclude by considering a possible countermeasure, a blacklist fed by a honeypot, which we show to be effective.

ricevuto il 17 Gennaio 1994; approvato il 15 Settembre 1994)Summary. --Stephany's treatment is not correct since he neglects the friction force proportional to the velocity, which we show to be universal, i.e. present in any mechanism of electric conduction. Moreover his predicted power spectral density is in some cases three orders of magnitude larger than the observed one. Finally, his noise cannot be the true 1/f noise because it should predict a low cut-off of 5.10 .2 Hz due to the transit time of the electron-linked lattice (ELL) carriers through a distance of 20 ~rn while no cut-off has been observed down to 10 -3 Hz.PACS 72.70 -Noise processes and phenomena.

A tradeoff between data collection needs and user privacy is of paramount importance in the Smart Grid. This paper proposes a pseudonymization protocol for data gathered by the Smart Metres, which relies on a network infrastructure and a dedicated set of nodes, called privacy preserving nodes. The network privacy is enforced by a separation of duties; the privacy preserving nodes perform data pseudonymization without having access to the measurements, which are masked by means of a secret sharing scheme, while the entities accessing the data recover and relate the plain measurements generated by the same metre along a time window of finite duration but have no access to the metre identities. The paper also provides an evaluation of the security and of the performance of the protocol, comparing it to the two alternative encryption techniques, which mask the measurements by means of the Chaum mixing scheme or of an identity‐based proxy re‐encryption scheme. Copyright © 2013 John Wiley & Sons, Ltd.