Catálogo de publicaciones - libros

This paper describes a novel method of achieving packet scheduling in several routers of network, in order to optimize the end to end delay. We use a multi-agent system to model this problem, where each agent of this system tries to optimize the local scheduling and through a communication with each other, attempts to make global coordination in order to optimize the total scheduling. The communication between agents is done by mobile agents like ants colony. A pheromone-Q learning approach is presented in this paper, which consists to applying the standard Q-learning technique adapted to our architecture with a synthetic pheromone that acts as a communication medium speeding up the learning process of cooperating agents.

This paper provides a summary of development process for the conceptual design of an integrated network management system (NMS) platform for incumbent with a huge legacy system. Based on proven international standards such as Enhanced Telecom Operation Map (eTOM) [1] and New Generation Operations System and Software (NGOSS) [2] framework an approach has been develop to address the current situation of incumbent as well as to make provision for NGN. In other world, the solution has been developed in a way that a relatively smooth transformation/ development of the existing fragmented management landscape towards an integrated NMS is possible. A service centric concept has been deployed for our conceptual design.

This paper relates to a method for performing Stateful Packet Inspection(SPI) in real time using a session table management scheme that allows more efficient generation of session state information. SPI is an important technique to reduce false positive alerts in network intrusion detection system(NIDS). As the number of session increases, this technique requires a higher processing speed, thereby causing performance problems. However, existing software-based solutions cannot perform real-time packet inspection ensuring the wire speed. To guarantee both performance and functionality with respect to statefulness, we designed and implemented SPI-based intrusion detection module in a FPGA to help alleviating a bottleneck in network intrusion detection systems in this paper.

Currently ISPs use tools, such as Interior Gateway Protocol Weight Optimizer (IGP-WO), to tune link weights of routing protocols so that offered traffic load is balanced evenly for network’s links. IGP-WO uses a sequential algorithm that prevents efficient use of parallel computing resources. We introduce a parallel link weight optimization architecture and compare its performance to IGP-WO. We find that our system achieves good load balancing much faster than IGP-WO.

Currently, network management systems on Internet are lots of islands floating around. They seldom get connected with each other, especially among those of different networks. Under such situation, overlapping implementation of network management functions in the same network domain commonly exists, and end-to-end management can’t be smoothly implemented across different network domain. To solve these problems, a novel model of network management architecture is proposed, by name, Internet Management Network (IMN). IMN is an overlay network of NMSes. The structure of IMN is almost the same like Internet itself. A prototype system based on SNMP is also introduced with an interesting application, multi-domain search engine of network management information.

In a popular video broadcasting method, two major technical issues are focused on both to reduce the viewer’s waiting time maintaining a given bandwidth allocation and the client’s buffer requirements. In this paper, we propose a pagoda staggered broadcasting (PSB) method which has a simple structure and substantially improves broadcasting efficiency in given bandwidth. The proposed method performs significantly better than the existing broadcasting methods in the viewer’s waiting time and the buffer requirement with simplicity. The numerical results demonstrate that the viewer’s waiting time of the PSB method is nearly equal to the existing pagoda broadcasting methods. However the complexity of this method decreases up to 60% if dividing coefficient is 1 by adjusting the short front part of a video.

Congestion control is an important part of IP QoS. The traditional congestion control algorithms judge the network congestion status and make drop decisions mainly according to the congestion information of local buffer resources. This causes the bandwidth-wasting problem when data flow congested. The paper analyzes the problem theoretically and proposes a new congestion control algorithm (CC-AMR) based on awareness of the congestion status of multistage resources. CC-AMR accounts the congestion status of resources in remote forward engines and their ports synthetically so that more reasonable congestion control decisions can be made. The CC-AMR has been implemented and tested on a core router which is implemented basing on network processor. Results show that this algorithm can enhance the total throughput of router effectively during periods of congestion.

To support various devices for multimedia services, it needs to some adaptation technologies. A transcoding is one of major fields in adaptive multimedia delivery. A transcoding proxy is usually necessary to provide adapting multimedia object to various mobile devices by not only transcoding multimedia objects to meet different needs on demand, but also caching them for later use. In this paper, we propose a caching algorithm in transcoding proxy that can reduce the startup time, improve caching ratio with segment-based proxy management. Our simulation results demonstrate that the proposed algorithms outperform the competing algorithms.