Bernard Cousin

Light-hierarchy: the optimal structure for multicast routing in WDM mesh networks

Based on the false assumption that multicast incapable (MI) nodes could not be traversed twice on the same wavelength, the light-tree structure was always thought to be optimal for multicast routing in sparse splitting Wavelength Division Multiplexing (WDM) networks. In fact, for establishing a multicast session, an MI node could be crosswise visited more than once to switch a light signal towards several destinations with only one wavelength through different input and output pairs. This is called Cross Pair Switching (CPS). Thus, a new multicast routing structure light-hierarchy is proposed for alloptical multicast routing, which permits the cycles introduced by the CPS capability of MI nodes. We proved that the optimal structure for minimizing the cost of multicast routing is a set of light-hierarchies rather than the light-trees in sparse splitting WDM networks. Integer linear programming (ILP) formulations are developed to search the optimal light-hierarchies. Numerical results verified that the light-hierarchy structure could save more cost than the light-tree structure.

Distance priority based multicast routing in WDM networks considering sparse light splitting

As we know, the member-only algorithm in provides the best links stress and wavelength usage for the construction of multicast light-trees in WDM networks with sparse splitting. However, the diameter of tree is too big and the average delay is also too large, which are intolerant for QoS required multimedia applications. In this paper, a distance priority based algorithm is proposed to build light-trees for multicast routing, where the Candidate Destinations and the Candidate Connectors are introduced. Simulations show the proposed algorithm is able to greatly reduce the diameter and average delay of the multicast tree (up to 51% and 50% respectively), while keep the same or get a slightly better link stress as well as the wavelength usage than the famous Member-Only algorithm.

Graph Partitioning for Survivability in Multi-Domain Optical Networks

As optical networking deployments increase, multi-domain provisioning and survivability are becoming major concerns. A key challenge in multi-domain survivability is the scalability problem. To address this concern, various solutions have already been proposed based upon topology aggregation schemes. However, these mechanisms do not scale well with increasing domain counts and further investigation is required to develop more scalable alternatives. Along these lines, in this paper we propose to use graph partitioning techniques to solve the scalability problem in multi-domain optical networks. To demonstrate the efficiency of our method, we also extend the p-cycle concept to multi-domain settings. Overall simulation results show the efficiency of our proposed solution in terms of resource utilization and the number of p-cycle structures.

Modeling Uncertainties in Proactive Routing Protocols for AdHoc Networks

In this paper we introduce a new model for ad hoc networks. Our model aims at reproducing the states alternation of links and nodes led by dynamic and random topology of ad hoc networks as well as random delays of packets delivery. Furthermore, we study the phenomenon of uncertainties in routing operation due to the unpredictable topological changes. Unlike mobility models which reflect only the impact of mobility on routing protocols, our proposal can demonstrates the impact of several identified hazards on the performances of proactive routing protocols. We show through various simulations the performances of our model within the framework of a proactive routing.

Evaluation of gateway-based shaping methods for HTTP Adaptive Streaming

HTTP Adaptive Streaming (HAS) is a streaming video technique commonly employed over best-effort networks. However, it is characterized by some issues that harm its quality of experience (QoE) in cases of daily use. The main use case of the present investigation involves HAS clients competing for bandwidth inside the same home network. Based on related work, we found that one of the most convenient solutions for this use case is to define a bandwidth manager, on the gateway side, that divides the available home bandwidth between HAS clients. Two main methods have previously been proposed to shape the HAS streams in accordance with the bandwidth managers direction and are referred to as gateway-based shaping methods: a highly renowned method, Hierarchical Token Bucket Method (HTBM), that uses the hierarchical token bucket queuing discipline, and another method, Receive Window Tuning Method (RWTM), that employs TCP flow control by handling only acknowledgment TCP packets. In this paper, we compare these two shaping methods. Results indicate that RWTM improves the QoE better than HTBM and does not add queuing delay. Results were validated through experimentation and objective QoE analytical criteria.

Extending node protection concept of p-cycles for an efficient resource utilization in multicast traffic

In this paper, we consider link-and-node failure recovery in dynamic multicast traffic in WDM networks. We extend the node protection concept of the p-cycle approach to achieve more efficient resource utilization. Then, we propose a novel algorithm that integrates our concept for the node protection, named node-and-link protecting p-cycle based algorithm (NPC). We also propose a second algorithm, named node-and-link protecting candidate p-cycle based algorithm (NPCC). This algorithm deploys our concept for node protection and relies on a candidate p-cycle set to speed up the computational time. We compare our proposed algorithms to the ESHN algorithm, which is reported to be the most efficient algorithm for protecting dynamic multicast sessions. Extensive simulations show that the NPC algorithm achieves the lowest blocking probability, but has the highest computational time among the NPCC and ESHN algorithms. The NPCC algorithm outperforms the ESHN algorithm in terms of resource utilization efficiency and computational time.

Heuristics for Joint Optimization of Monitor Location and Network Anomaly Detection

Achieving cost-effective systems for network performance monitoring has been the subject of many research works over the last few years. Most of them adopt a two-step approach. The first step assigns optimal locations to monitors, whereas the second step selects a minimal set of paths to be monitored. However, such an approach does not consider the trade-off between the optimization objectives of each step, and hence may lead to sub-optimal usage of network resources and biased measurements. In this paper, we propose to evaluate and reduce this trade-off. Toward this end, we come up with two ILP formulations for a novel monitoring cost model that apply for both passive and active monitoring. The aim is to jointly minimize the monitor location cost and the anomaly detection cost, thereby obtaining a monitoring solution that minimizes the total monitoring cost. Simulation results illustrate the interplay between the optimization objectives and evaluate the quality of the obtained monitoring solution.

Hybrid decision algorithm for access selection in multi-operator networks

In this paper, we propose a hybrid decision algorithm for the selection of the access in multi-operator networks environment, where competing operators share their radio access networks to meet traffic and data rate demands. The proposed algorithm guarantees the user satisfaction and a global gain for all cooperating operators. Simulation results prove the efficiency of the proposed scheme and show that the cooperation between operators achieves benefits to both users and operators; user acceptance as well as the operator resource utilization and the operator revenue increase.

Joint optimization of monitor location and network anomaly detection

Achieving cost-effective systems for network performance monitoring has been the subject of many research works over the last few years. Most of them adopt a two-step approach. The first step assigns optimal locations to monitors, whereas the second step selects a minimal set of paths to be monitored. However, such an approach does not consider the trade-off between the optimization objectives of each step, and hence may lead to sub-optimal usage of network resources and biased measurements. In this paper, we propose to evaluate and reduce this trade-off. Toward this end, we come up with two ILP formulations for a novel monitoring cost model that apply for both passive and active monitoring. The aim is to jointly minimize the monitor location cost and the anomaly detection cost, thereby obtaining a monitoring solution that minimizes the total monitoring cost. Simulation results illustrate the interplay between the optimization objectives and evaluate the quality of the obtained monitoring solution.

Avoidance of multicast incapable branching nodes for multicast routing in WDM networks

Although many multicast routing algorithms have been proposed in order to reduce the total cost in WDM optical networks, the link stress and delay are two parameters which are not always taken into consideration. This paper proposes a novel wavelength routing algorithm, which tries to avoid the multicast incapable branching nodes (MIB, branching nodes without splitting capability) to diminish the link stress for the shortest path based multicast tree and maintains good parts of the shortest path tree to reduce the end-to-end delay. Firstly a DijkstraPro algorithm with priority assignment and node adoption is introduced to produce a shortest path tree with up to 38% fewer MIB nodes, and then critical articulation and deepest branch heuristics are used to process the MIB nodes. Finally distance based reconnection algorithm is proposed to create the multicast tree or forest.