Sustaining Negotiated QoS in Connection Admission Control for ATM Networks Using Fuzzy Logic Techniques

Article excerpt

Introduction

The major objective of ATM is to integrate real-time information such as voice and video with non-real-time computer data, within the same transmission and switching medium ("TechArena Community", 2007). Data requires very low Bit Error Rate (BER) but can tolerate large propagation delays (seconds) (Valcourt, 1997). Voice and video require small propagation delays (milliseconds) but can tolerate some errors or small losses of information.

The objectives of ATM traffic management are to deliver Quality of Service (QoS) guarantee for the multimedia applications and provide overall optimization of network resources (Esaki, Iwamura, Kodama, & Fukuda, 1994). The control of ATM traffic is complicated due to ATM's high link speed and small cell size, the diverse service requirements of ATM applications, and the diverse characteristics of ATM traffic (Suda, 1998). The environment also has a significant impact on the choice of control mechanism, either local or wide area.

Connection Admission Control (CAC) is a procedure responsible for determining whether a connection request is admitted or denied. The procedure is based on resource allocation schemes applied to each link and switching unit (Esaki et al., 1994). Admission control decision is made using a traffic descriptor that specifies traffic characteristics alongside the QoS requirements. These traffic characteristics include: Peak Cell Rate (PCR); Sustainable Cell Rate (SCR); and Maximum Burst Size (MBS).

The statistical CAC takes advantage of the variable bit rate bursty nature of traffic with the hope that not all sources will need their peak rate at the same time thus balancing the peaks and valleys of the bit rates (Esaki et al., 1994). With this, allocation is made on the network which increases network utilization leading to network efficiency. Statistical gain can also be significant.

Fuzzy logic is multi-valued, dealing in degrees of membership or truth within the set (Kaehler, 1998; Kosko, 2003). Fuzzy logic us(x) describes the membership function of S, or the degree to which x is a member of the set S, this is known as the degree of truth.

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

With fuzzy logic, this transition at the borders of sets is gradual, thus for the allowance for membership in both sets (Kaehler, 1998).

The fuzzy logic analysis and control method is, therefore: * Receiving one of or a large number of measurements or other assessment of conditions existing in the control system.

* Processing all these inputs according to human based, fuzzy "If-Then" rules, which can be expressed in plain language words, in combination with traditional non-fuzzy processing.

* Averaging and weighting the resulting outputs from all the individual rules into one single output decision or signal which decides what to do or tells the controlled system what to do.

The output signal eventually arrived at is a precise appearing, defuzzified, "crisp" value as shown in Figure 1.

[FIGURE 1 OMITTED]

Background Study

For ATM networks to integrate all types of data and multimedia user traffic, CAC functions must provide guaranteed, or at least differentiated, quality of service levels. This must occur at the packet level to meet rate and delay specifications, and also occur at the connection level to give differentiated access to shared resources (Valcourt, 1997).

Much of the existing work on CAC specifies the QoS parameters as fixed values (e.g., traffic with peak 10 Mbps, and deadline 30 milliseconds) and does not exploit the dynamic fluctuations in resource availability (Esaki et al., 1994; Tian, 1998). A connection can be viewed as a contract between an application and the connection management system. A real-time connection is additionally characterized by stringent deadline constraints imposed on its packet delivery time (Devalla et. …