Catálogo de publicaciones - libros

Many research works address detection and identification of network anomalies using traffic analysis. This paper considers large topologies, such as those of an ISP, with traffic analysis performed on multiple links simultan-eously. This is made possible by using a combination of simple online traffic parameters and specific data from headers of selective packets. Even though large networks may have many network links and a lot of traffic, the analysis is simplified with the usage of Principal Component Analysis (PCA) subspace method. The proposed method proves that aggregation of such traffic profiles on large topologies allows identification of a certain set of anomalies with high level of certainty.

In NGN, many transport-related technologies were considered. One technology discussed is ATM (Asynchronous Transfer Mode) based MPLS (Multi-protocol Label Switching) for its provisioning QOS (Quality Of Service) commitment, traffic management aspects and smooth migration to the BcN in Korea. In this paper, we present preventive congestion control mechanism that uses network signaling information for detecting and fuzzy control to a SC(Slim Chance) of LSP (Labeled Switched Path) in ATM based MPLS system. Proposed control can handle 208% call processing and more than 147% success call, than those without control. It can handle 187% BHCA (Busy Hour Call Attempts) with 100 times less use of exchange memory.

This paper proposes a DiffServ-over-MPLS Traffic Engineering (TE) architecture and describes the implementation of its functional blocks on Intel IXP2400 Network Processor using Intel IXA SDK 4.1 framework. We propose fast and scalable 6-tuple range-match classifier, which allows traffic policing procedures to operate on per-flow level, and a scalable low-jitter Deficit Round Robin (DRR) scheduler that can provide bandwidth guarantees on LSP level. The proposed DiffServ-over-MPLS TE functional blocks have been implemented on Intel IXDP2400 platform for up to 4,096 flows mapped to L-LSPs, and can handle an aggregated traffic rate of 2.4Gbps.

Rapid evolution in ubiquitous systems is complicating its management. In the realm of ubiquitous computing, the domain specific ontologies are constructed manually by experts, which is quite a toiling task. We proposed to extract ontologies dynamically for ubiquitous systems. We also argue that the components of the software middleware are the specialized research areas with different subject orientation e.g. context reasoning, service discovery etc[1] and following an evolvable component oriented approach would let the components of the software evolve independently of other software components, while making them interoperable irrespective of versions and vendors.

This paper proposes a dynamic shared-memory packet buffer management algorithm for a protocol processor in a network terminal. The protocol processor is located in a network interface card (NIC). In general, two types of packet buffer management algorithms, static and dynamic, can be used in a NIC; the dynamic buffer management algorithms work better than the static ones for reducing the packet loss ratio. However, conventional dynamic buffer management algorithms do not provide even packet losses to all the applica-tions. Therefore, we propose an algorithm to enhance even packet losses and the proposed algorithm improves the packet loss ratio by 11.80% to 16.70% compared to other conventional dynamic algorithms.

The explosive growth of new multimedia services over the Internet necessitates efficient network management. Improved network management systems are expected to simultaneously provide diverse multimedia traffic services and enhance network performance. In this paper, we propose a new online network management algorithm that implements adaptation, reservation, and call-admission strategies. Our online approach to network management exhibits dynamic adaptability, flexibility, and responsiveness to the current traffic conditions in multimedia networks.

In this paper, an efficient neighbor AR (Access Router) discovery scheme and handoff procedure using neighbor information are proposed. It allows each AR and Mobility Anchor Point (MAP) to know neighboring ARs and MAPs, and therefore, a mobile node (MN) can perform the handoff process in advance, using the proposed handoff mechanism. It is important to note that the Inter-MAP domain handoff improvement of the proposed scheme is up to about 57% and 33% for handoff latency in comparison of the Hierarchical Mobile IPv6 (HMIPv6) and the Hierarchical Mobile IPv6 with Fast handover (F-HMIPv6), respectively. The proposed scheme is approximately two times better than the existing one in terms of the total signaling costs through performance analysis. Therefore, it is sufficiently satisfied with the requirements of real-time applications, and seamless communication is expected.

For an efficient role based access control using attribute certificate in highly distributed computing environments, we use a technique of structuring role specification certificates. The roles are grouped and made them into the relation tree. It can reduce management cost and overhead incurred when changing the specification of the role. Further we use caching of frequently used role specification certificate for better performance in case applying the role. And for the global space reduction, we also consider the issue of tree normalization. In order to be scalable distribution of the role specification certificate, we use multicasting packets. In the experimental section, it is shown that our proposed method is secure and efficient.

In this paper, for the combination of DiffServ and RSVP, we propose traffic conditioning algorithm. Through the proposal, we expect the service quality improvement of real time application, especially voice traffic, by separation of other data traffic.

Over the past few years we have witnessed steady progress toward convergence in communications. We believe we are now witnessing convergence beyond just media transport. Further, abstract application level signaling technologies are poised for takeoff and will serve as the unifying fabric for multi-dimensional convergence. Our research shows that such a signaling mechanism known as “presence”, combined with advances in namespace and security technology, will enable ultimate service innovations.