Estimation of Traffic Matrix from Links Load using Genetic Algorithm

Pachuau, Joseph L.; Krishna, Gopal; Saha, Anish Kumar

doi:10.12694/scpe.v22i1.1834

Cited by 4 publications

(2 citation statements)

References 32 publications

(34 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They also propose a suitable re-calibration strategy after fixed time intervals to maintain adequate performance. In [27], tomogravity is used to provide an initial TM estimate, which is subsequently improved via Genetic Algorithm (GA) optimization using an objective function based on (1). Last but not least, Ephraim et al [28] extend NT by assuming that the distribution of the counts of each OD flow is a (continuous or discrete) mixture of Poisson distributions, and they estimate the mean traffic rate for every OD pair by solving a second-order moment matching system using least squares and the minimum I-divergence iterative procedure.…”

Section: Related Workmentioning

confidence: 99%

Generative Deep Learning Techniques for Traffic Matrix Estimation From Link Load Measurements

Kakkavas,

Fryganiotis,

Karyotis

et al. 2024

IEEE Open J. Commun. Soc.

View full text Add to dashboard Cite

Traffic matrices (TMs) contain crucial information for managing networks, optimizing traffic flow, and detecting anomalies. However, directly measuring traffic to construct a TM is resource-intensive and computationally expensive. A more practical approach involves estimating the TM from readily available link load measurements, which falls under the category of inferential network monitoring based on indirect measurements known as network tomography. This paper focuses on solving the problem of estimating the traffic matrix from link loads by utilizing deep generative models. The proposed models are trained using historical data-specifically, previously observed TMs-and are then leveraged to transform traffic matrix estimation (TME) into a simpler minimization problem in a lower-dimensional latent space. This transformed problem can be efficiently solved using a gradient-based optimizer. Our work aims to examine and test different model architectures and optimization approaches. The performance of the proposed methods is comparatively evaluated over a comprehensive set of suitable metrics on two publicly available datasets comprising actual traffic matrices obtained from real backbone networks. In addition, we compare our approach with a state-of-the-art method previously published in the literature.

show abstract

Section: Related Workmentioning

confidence: 99%

Generative Deep Learning Techniques for Traffic Matrix Estimation From Link Load Measurements

Kakkavas,

Fryganiotis,

Karyotis

et al. 2024

IEEE Open J. Commun. Soc.

View full text Add to dashboard Cite

show abstract

“…In this section, we use the evolutionary algorithm (EA) approach NSGA-II [2] (Non-dominated sorting genetic algorithm II) to deal with the equivalent crop planning problem as a multi-objective programming problem (12), instead of using the deterministic algorithms (see [15], [16] and references therein) or interactive methods (see [17], [21]). The reason is that EAs simultaneously find many approximate solutions (see [1], [11]) without analysis of objective functions and have speedy performance when compared with classical methods. Decision makers can easily choose the optimal solution from this approximated solution set, for instance, by adding a sub-criteria as optimizing a function over the finite set of approximated solutions.…”

Section: The Multi-objective Evolutionary Algorithm Solving Crop Plan...mentioning

confidence: 99%

A Multi-criteria Fuzzy Random Crop Planning Problem using Evolutionary Optimization

Hoang¹,

Xuan²,

Thang³

2022

Proceedings of the Sixth International Conference on Research in Intelligent and Computing

View full text Add to dashboard Cite

In this paper, we deal with fuzzy random objectives in a multi-criteria crop planning problem considered as a multiobjective linear programming problem. These fuzzy random factors are related to decision making processes in practice, especially the uncertainty and synthesized objectives of experts. The problem is transformed into a multi-objective nonlinear programming problem by a step of evaluating expectation value. Instead of using classical methods, we use a multi-objective evolutionary algorithms called NSGA-II to solve the equivalent problem. This helps finding many approximate solutions concurrently with a low time consumption. In computational experiments, we create a specific fuzzy random crop planning problem with the data synthesized from government's reports and show convergence of the algorithm for proposed model.

show abstract