Gunars Lauks

Fuzzy-CAC driven MPLS-TE realization

The Multiprotocol Label Switching - Traffic Engineering Extension (MPLS-TE) was introduced by the Internet Engineering Task Force (IETF) to ensure traffic engineering over MPLS. However, an actual Connection Admission Control (CAC) implementation inside the resource reservation protocol - traffic engineering extension (RSVP-TE) in MPLS-TE networks does not provide the ability of effective decision making, since the applied threshold CAC lacks of the capability to consider QoS policies on MPLS-TE network nodes. This prevents an effective end-to-end QoS control in a fully dynamic, application driven Label Switched Path (LSP) setup scenario. Current paper presents the practical fuzzy-CAC driven MPLS-TE realization, which is based on the specific implementation of fuzzy-CAC algorithm over an RSVP-TE agent. Fuzzy-CAC implementation is applied to a testbed where a client application requests a real-time data transfer through the MPLS-TE network, which results in dynamic LSP setup and exclusion. The admission control is performed upon service request based on QoS class requirements and network resource availability. The differentiated traffic treatment on per-flow basis is realized through employment of IF-THEN rule based expert knowledge. Effective traffic management is achieved in a best and worst case scenarios and thus, it validates fuzzy-CAC as a candidate for RSVP-TE protocol enhancement for application driven QoS provisioning in MPLS-TE networks and Generalized Multiprotocol Label Switching (GMPLS) networks in the nearest future.

Fuzzy approach for QoS aware application driven traffic control in GMPLS networks

The ITU-T defined next generation network (NGN) architecture designates the resource admission control function to perform the application-driven Quality of Service (QoS) control across access and core networks. The Generalized Multiprotocol Label Switching (GMPLS) was introduced by the Internet Engineering Task Force (IETF) to cope with new traffic engineering challenges in fast optical networks and provide them with reliable end-to-end QoS mechanisms. However, an actual Connection Admission Control (CAC) implementation inside the resource reservation protocol – traffic engineering extension (RSVP-TE) in GMPLS networks does not provide the ability of effective decision making, since the applied threshold CAC lacks of the capability to consider QoS policies on GMPLS network nodes. This prevents an effective end-to-end QoS control in a fully dynamic, application driven Label Switched Path (LSP) setup scenario. This work presents a specific implementation of fuzzy-CAC operating over an RSVP-TE agent in GMPLS network domain. This fuzzy-CAC implementation is applied to a testbed where a client application requests a real-time data transfer through a GMPLS network, which results in dynamic LSP setup and exclusion. The admission control is performed upon service request based on QoS class requirements and network resource availability. The differentiated traffic treatment on per-flow basis is realized through employment of IF-THEN rule based expert knowledge. Effective traffic differentiation is achieved in a multi-service network scenario and thus it validates fuzzy-CAC as candidate for RSVP-TE protocol enhancement for application driven QoS provisioning in GMPLS networks.

Network traffic classification for anomaly detection fuzzy clustering based approach

In this paper we develop network traffic classification and anomaly detection methods based on traffic time series analysis using fuzzy clustering technique. The effectiveness of fuzzy and possibilistic algorithms is compared on generated traffic data with and without traffic attack components.

Comprehensive analysis of AggSessAC method for revenue maximization using OMNeT

The growing demand for Internet connection of various devices with an ability to provide smarter online services and the rapid growth of mobile applications significantly increases the number of processed data flows. All the generated flows require selective and priority-based flow admission strategy. Network operators are interested in effective utilization of their infrastructure as well as in minimizing rejection probability of higher priority flows while maximizing their revenue, especially in peak hours. The existing connection admission control (CAC) schemes are largely based on serialized processing strategies of new flows without any comparison among consequent requests. However, evolution of Internet and present performance capabilities of routers allows us to offer a new approach for admission control - our developed Aggregated Session Admission Control (AggSessAC). We propose to handle service requests using a new operation paradigm of CAC, where requests are temporarily collected and processed using mutually comparisons among them, thus facilitating selectivity and ensuring network revenue maximization as well as operator gain. In order to evaluate the proposed algorithm, OMNeT++ simulation platform with the INET Framework was used and a new output queue of router has been developed including all relevant entities of proposed admission control. Simulation results are compared with conventional threshold admission control method, which only uses available link bandwidth for decision-making process and serialized flow processing strategy. The proposed method shows that selective and comparative flow control allows maximizing the number of accepted higher priority flows and is able to significantly increase the total network revenues in peak hours, compared to the standard threshold based approach. We assume that AggSessAC can be effectively used as the potential admission control mechanism in Next Generation Networks (NGN).

Fuzzy Logic Based Network Bandwidth Allocation: Decision Making, Simulation and Analysis

We present a fuzzy logic based methodology of decision making on bandwidth allocation in a substrate network with DaVinci architecture, according to which the physical substrate network is divided into virtual networks. This methodology describes a fuzzy modification of the adaptive bandwidth allocation mechanism introduced in order to optimize decision making under uncertain network conditions by using fuzzification and defuzzification principles and the expert knowledge database of fuzzy rules. The system of fuzzy rules is optimized using inequalities for fuzzy values of linguistic variables. The effectiveness of this methodology is evaluated on the link level for two traffic types within simulation experiments realized by using Coloured Petri Nets Tools.

A fuzzy approach for network bandwidth management

The paper deals with the problem of resource allocation in a substrate network with DaVinci architecture. We present a methodology of decision making on network bandwidth allocation based on fuzzification and defuzzification principles and the expert knowledge database of fuzzy rules and we describe a fuzzy logic based modification of the adaptive bandwidth allocation mechanism to optimize decision making under uncertain network conditions. We describe a simulation scheme used for the estimation of performance of this modification for two nodes network topology.