Catálogo de publicaciones - libros

In this paper, a novel algorithm optimizing the utilization of backup path resources for survivable optical networks, based on graph vertex-coloring approach, is introduced. To the best of our knowledge, this is the first optimization technique, dedicated to WDM grooming networks, such that does not increase the length of backup paths and thus provides fast service restoration. Due to NP-completeness of the proposed Integer Linear Programming model, the respective heuristic algorithm has been developed. The concept was evaluated for the U.S. Long-Distance Network, European COST 239 Network and Polish PIONIER Network. The results show that with only a little capacity utilization degradation, fast restoration can be achieved and the resource utilization kept at low level. The observed reduction in restoration time values is significant (up to 40%), compared to the typical a priori approach.

In this paper, we tackle the question of how anycast communication can improve survivability of connection-oriented networks. The focus on network survivability arises naturally in the context of growing role of computer networks in almost all aspects of human life. We concentrate on connection-oriented networks e.g. MPLS, ATM for the reason that these techniques can provide effective and reliable services in transport network. We take a detailed view on anycast communication from the perspective of network survivability. A new optimization problem is formulated that is equivalent to the problem of joint unicast and anycast flows restoration in connection-oriented networks. Next, we propose a heuristic algorithm solving that problem. Finally, we present results of exhaustive numerical experiments and a comprehensive discussion on the application of anycast communication in the context of network survivability.

In an all-IP environment, the concept of context transfer is used to provide seamless secure handovers between different administrative domains. However, the utilization of context transfer arises some privacy issues concerning the location and movement of users roaming between domains. In this paper we elaborate on these privacy issues and propose an alternative context transfer protocol that protects user’ location privacy as well. In addition, assuming that the context carries a user identity in the form of a Network Access Identifier (NAI), we show how the employment of temporary NAIs can further increase the privacy of our scheme.

Current multipath protocols for Multi-Hop Wireless Networks (MWNs) use hop-count as the default route selection criteria. Route selection should also consider network and link conditions. We propose a network-environment-aware trust-based route selection framework for MWNs that makes informed and adaptive route-selection decisions. A node quantifies trust values for its neighboring nodes and for the routes that pass through it. The trust metric adjusts to varying network conditions and quick convergence of the protocol implies it works well in mobility scenarios. Glomosim simulations demonstrate throughput improvement over conventional multipath protocols under congestion, link failure and route unreliability scenarios.

The lack of control inherent to digital content has been put on the spotlight by copyright infringement coupled with massive content online distribution (e.g., Peer-to-Peer). Digital Rights Management seems to be the solution to counter this problem advocating the use of cryptography and other related security mechanisms to protect digital content and to associate rights with it which determine how, when and by whom it can be consumed. The rapid growth of digital multimedia technologies brings tremendous attention to the field of digital watermarking. Watermarking embeds a secret message into a cover multimedia data. In media watermarking the secret is usually a copyright notice and the cover a digital image. In digital watermarking, robustness is still a challenging problem if different sets of attacks needed to be tolerated simultaneously. In this paper we present an original spatial watermarking technique for video and images. Our approach modifies blocks of the image or frames by a spatial watermark insertion. Spatial mask of suitable size is used to hide data with less visual impairments. Watermark insertion process exploits average color of the homogeneity regions of the cover image. We took a frame-based approach to video watermarking. From video we extract a certain number of key-frames: the first, the middle, and last key-frame. The first step is decoding: transformation of mpeg to jpeg sequences, after that we select three frames that will be process by applying the watermark mask. In the reverse process of encoding we take the marked frames.

Enforcement of copyright laws of software products is primarily managed in legal way, as the available technological solutions are not strong enough to prevent illegal distribution and use of software. This situation is particularly dangerous on the growing market of mobile software products: the lack of protection may be the most important setback of the expansion of this market, although we have all the necessary components to assure a trusted environment on handsets. In this paper we present our copy protection scheme called Swotector targeting primarily mobile phone applications, which combines techniques of obfuscation and software-watermarking to provide a solution which is purely technical, however still provides the necessary flexibility. Our test results have shown that the solution is efficient enough to overcome current challenges of software copy protection, above all the ability of the operating system to run both protected and non-protected applications.