Abstract

Part 1: Traffic Engineering and Quality-of-Service; International audience; In current network infrastructures, several management tasks often require significant human intervention and can be of high complexity, having to consider several inputs to attain efficient configurations. In this perspective, this work presents an optimization framework able to automatically provide network administrators with efficient and robust routing configurations. The proposed optimization tool resorts to techniques from the field of Evolutionary Computation, where Evolutionary Algorithms (EAs) are used as optimization engines to solve the envisaged NP-hard problems. The devised methods focus on versatile and resilient aware Traffic Engineering (TE) approaches, which are integrated into an autonomous optimization framework able to assist network administrators. Some examples of the supported TE optimization methods are presented, including preventive, reactive and multi-topology solutions, taking advantage of the EAs optimization capabilities. Document type: Part of book or chapter of book

Abstract

Segment Routing (SR) combines the simplicity of Link-State routing protocols with the flexibility of Multiprotocol Label Switching (MPLS). By decomposing forwarding paths into segments, identified by labels, SR improves Traffic Engineering (TE) and enables new solutions for the optimization of network resources utilization. This work proposes an Evolutionary Computation approach that enables Path Computation Element (PCE) or Software-defined Network (SDN) controllers to optimize label switching paths for congestion avoidance while using at the most three labels to configure each label switching path. This work has been supported by COMPETE: POCI-01-0145-FEDER-007043 and FCT Fundac¸˜ao para a Ciˆencia e Tecnologia within the Project Scope: UID/CEC/00319/2013. info:eu-repo/semantics/publishedVersion

Abstract

In this work, the main goal is to develop and evaluate a number of optimization algorithms in the task of improving Quality of Service levels in TCP/IP based networks, by configuring the routing weights of link-state protocols such as OSPF. Since this is a complex problem, some meta-heuristics from the Evolutionary Computation arena were considered, working over a mathematical model that allows for flexible cost functions, taking into account several measures of the network behavior such as network congestion and end-to-end delays. A number of experiments were performed, resorting to a large set of network topologies, where Evolutionary Algorithms (EAs), Differential Evolution and some common heuristic methods including local search were compared. EAs make the most promising alternative leading to solutions with an effective network performance even under unfavorable scenarios. Document type: Part of book or chapter of book

Abstract

This paper proposes a novel traffic engineering framework able to automatically provide near-optimal OSPF routing configurations for QoS constrained scenarios. Within this purpose, this work defines a mathematical model able to measure the QoS compliance in a class-based networking domain. Based on such model, the NP-hard optimization problem of OSPF weight setting is faced resorting to Evolutionary Algorithms. The presented results show that, independently of other QoS aware mechanisms that might be in place, the proposed framework is able to improve the QoS level of a given domain only taking into account the direct influence of the routing component of the network. The devised optimization tool is able to optimize OSPF weight configurations in scenarios either considering a single level of link weights or using multiple levels of weights (one for each class) in multi-topology routing scenarios. Document type: Part of book or chapter of book

Abstract

Electric vehicle fleets and smart grids are two growing technologies. These technologies provided new possibilities to reduce pollution and increase energy efficiency. In this sense, electric vehicles are used as mobile loads in the power grid. A distributed charging prioritization methodology is proposed in this paper. The solution is based on the concept of virtual power plants and the usage of evolutionary computation algorithms. Additionally, the comparison of several evolutionary algorithms, genetic algorithm, genetic algorithm with evolution control, particle swarm optimization, and hybrid solution are shown in order to evaluate the proposed architecture. The proposed solution is presented to prevent the overload of the power grid.

Abstract

International audience; This paper presents a new method for 3D cutting of geometrical space with application to airspace sectoring. This problem comes from the air traffic management but the proposed method may be applied to many other areas. This problem consists in finding a cutting of a 3D volume into sectors in order to balance the weights of sectors and which minimizes the flow cut on sector boundaries. A mathematical modeling of this problem has been proposed for which state space, objective functions and constraints are defined. The complexity of such problem being NP_Hard, stochastic optimization have been used to address it. An Evolutionary Algorithm has been implemented for which chromosome coding and operators have been developed. Realistic problem instances have been tested on this algorithm for which the solutions produced fulfill our objective.

Abstract

In the SESAR traffic growth predictions, traffic complexity will become an issue that the current Air Traffic Management organization is not able to handle. The 4D trajectory concept offers new perspectives for deconflicting the traffic by ground-holding aircraft before they take-off. This paper studies the possible complexity reduction achievable by optimizing the aircraft take-off times. Therefore a simple model is introduced to detect pairwise 3D possible conflicts and define conflicting take-off time differences. Two resolution algorithms are tested on a real traffic data sample collected in the French airspace. The first one is based on a Constraint Programming model of the problem and ensures the optimality of the maximum delay required to solve every conflict. The second one uses an evolutionary computation algorithm to minimize the mean delay among the aircraft population. A sliding window model is introduced to reduce the size of the problem and to regularly update the current situation. Experimental results performed in the French airspace with fast time simulation show that with perfect 4D trajectory, every conflict over flight level 290 can be solved by delaying less than a quarter of the traffic within a range of delays varying from 1 to 90 minutes and a mean delay of 4 minutes. The Constraint Programming approach gives better results than the evolutionary computation approach. Adding uncertainty around 4D trajectories dramatically degrades the results.