This work aims at proposing a transfer learning (TL) based framework to enhance system scalability of fingerprint-based indoor localization by reducing offline training overhead without jeopardizing the localization accuracy. The basic principle is to reshape data distributions in the target domain based on the transferred knowledge from the source domains, so that those data belonging to the same cluster will be logically closer to each other, whereas others will be further apart from each other. Specifically, the TL-based framework consists of two parts: metric learning and metric transfer, which are used to learn the distance metrics from source domains and identify the most suitable metric for the target domain, respectively. Furthermore, this work implements a prototype of the fingerprint based indoor localization system with the proposed TL-based framework embedded. Finally, extensive real-world experiments are conducted to demonstrate the effectiveness and the generality of the TL-based framework.

Efficient data dissemination is one of the fundamental requirements to enable emerging applications in vehicular cyber-physical systems. In this paper, we present the first study on real-time data services via roadside-to-vehicle communication by considering both the time constraint of data dissemination and the freshness of data items. Passing vehicles can submit their requests to the server, and the server disseminates data items accordingly to serve the vehicles within its coverage. Data items maintained in the database are periodically updated to keep the information up-to-date. We present the system model and analyze challenges on data dissemination by considering both application requirements and communication characteristics. On this basis, we formulate the temporal data dissemination (TDD) problem by introducing the snapshot consistency requirement on serving real-time requests for temporal data items. We prove that TDD is NP-hard by constructing a polynomial-time reduction from the Clique problem. Based on the analysis of the time bound on serving requests, we propose a heuristic scheduling algorithm, which considers the request characteristics of productivity, status, and urgency in scheduling. An extensive performance evaluation demonstrates that the proposed algorithm is able to effectively exploit the broadcast effect, improve the bandwidth efficiency, and enhance the request service chance.

This article examines a new problem of k -anonymity with respect to a reference dataset in privacy-aware location data publishing: given a user dataset and a sensitive event dataset, we want to generalize the user dataset such that by joining it with the event dataset through location, each event is covered by at least k users. Existing k -anonymity algorithms generalize every k user locations to the same vague value, regardless of the events. Therefore, they tend to overprotect against the privacy compromise and make the published data less useful. In this article, we propose a new generalization paradigm called local enlargement , as opposed to conventional hierarchy- or partition-based generalization. Local enlargement guarantees that user locations are enlarged just enough to cover all events k times, and thus maximize the usefulness of the published data. We develop an O ( H _n )-approximate algorithm under the local enlargement paradigm, where n is the maximum number of events a user could possibly cover and H _n is the Harmonic number of n . With strong pruning techniques and mathematical analysis, we show that it runs efficiently and that the generalized user locations are up to several orders of magnitude smaller than those by the existing algorithms. In addition, it is robust enough to protect against various privacy attacks.