Giuseppe Ficili

A comparative analysis of fuzzy versus conventional policing mechanisms for ATM networks

In ATM networks, usage parameter control is required in order to ensure that each source conforms to its negotiated parameters. To this purpose, several policing methods, such as leaky bucket and window mechanisms, have been introduced in literature. However, traditional methods have proved to be inefficient in coping with the conflicting requirements of ideal policing, that is, a low false alarm probability and high responsiveness. This led us to explore alternative solutions based on artificial intelligence techniques, specifically, in the field of fuzzy systems. We propose a policing mechanism based on fuzzy logic that aims at detecting violations of the parameters negotiated. The main characteristics of the proposed fuzzy policer are simplicity and the capacity to combine a high degree of responsiveness with a selectivity close to that of an ideal policer. Moreover, it can easily be implemented in hardware, thus, enhancing both cost and processing performance. The reported simulation results show that the performance of our fuzzy policer is much better than that of conventional policing mechanisms.

Using fuzzy logic in ATM source traffic control: lessons and perspectives

Due to its capacity to capture human expertise and to formalize approximate reasoning processes, fuzzy logic can be a good answer to the many challenges of congestion control in ATM networks. The authors deal with the application of fuzzy logic to problems of usage parameter control and propose a simple mechanism which, avoiding complex mathematical calculations, guarantees low response times while remaining effective. The flexibility of the fuzzy control proposed is discussed with respect to the probability of facing various types of traffic sources, ranging from bursty to MPEG video.

A fuzzy buffer management scheme for ATM and IP networks

We address the relevant issue of managing traffic flows with different priorities in packet switched networks, namely ATM and IP networks. We consider a reference model in which two traffic flows with different priorities are multiplexed within the buffer of a cell-based switch. The solution we propose, based on the fuzzy system theory, is able to guarantee the QoS requirements of high-priority traffic flow, allowing at the same time the exploitation of unused buffer resources to accommodate low-priority traffic flow in order to maximize the total throughput. The performance assessment of the fuzzy scheme demonstrates that our solution outperforms other popular mechanisms, based on conventional logic, such as threshold and push out mechanisms.

A fuzzy expert system for usage parameter control in ATM networks

In ATM networks usage parameter control is required in order to ensure that each source conforms to its negotiated parameters. However, traditional methods, such as leaky bucket and window mechanisms, revealed themselves to be inefficient in coping with the conflicting requirements of the ideal policing. We propose a fuzzy expert system for usage parameter control. The control strategy is described through a set of fuzzy inferences which emulate the knowledge base that is typical of human expertize. The simulation results shown in the paper demonstrate that our fuzzy policer outperforms conventional policing mechanisms.

Performance analysis of a fuzzy system in the policing of packetized voice sources

In this paper we analyze the performance of a policing mechanism based on Fuzzy Logic for the control of packetized voice sources. The results obtained show an excellent selectivity, close to that of an ideal policer, and a responsiveness, assessed by the combined measures of reaction time and rise time, which is decidedly better than that of the Leaky Bucket. Analysis by simulation also shows that the fuzzy policer is efficient in the combined control of two traffic descriptors, the Sustainable Cell Rate and the long-term Average Cell Rate.

A fuzzy logic based approach to multipriority control in ATM networks

The priority control function in ATM networks allows users to generate traffic flows with different priorities by using the Cell Loss Priority (CLP) bit. The implementation of this function, however, represents a hard task as the network has to guarantee the Quality of Service (QoS) of the higher priority traffic regardless of the presence of low-priority traffic and, at the same time, to reach an efficient use of network resources. The conventional schemes proposed so far suffer from some limitations either in performance (threshold mechanisms) or in feasible implementation (push-out scheme). We propose a queue management strategy based on fuzzy logic, which combines high performance, guaranteeing the QoS requirements and increasing the throughput, with a low implementation cost. In the paper, we assess the fuzzy mechanism by considering several traffic scenarios ranging from bursty traffic to MPEG video traffic. The presented results show that our mechanism is robust and outperforms conventional mechanisms.

On the impact of traffic control algorithms on resource management in ATM networks

Resource allocation algorithms in ATM networks operate on the basis of the value of certain source traffic descriptors and the extent to which the network resources are occupied. The different types of algorithms exhibit varying degrees of efficiency in estimating the resource to be allocated. However, the values any of them estimate will only be reliable if the traffic descriptors values remain, throughout the lifetime of the connections, the same as those negotiated in the call set-up phase. This can be guaranteed by a source traffic policing function which forces the traffic of any violating sources to respect the values negotiated. Unfortunately, a certain amount of excess traffic escapes the control of real policing mechanisms. This overloads the ATM switch for access to the network and consequently may cause an increase in the cell loss ratio. The article estimates the correction required by the CAC algorithm to avoid this risk. We compare two different policing mechanisms, one based on conventional logic and another on fuzzy logic, assessing the influence of their degree of selectivity on the additional bandwidth the CAC algorithm needs to reserve in order to guarantee the QoS requirements of all connections.

Performance evaluation of a fuzzy policer for MPEG video traffic control

The MPEG coding scheme is expected to be the compression algorithm for the first ATM video applications. On account of its wide variability and the type of coding used, usage parameter control (UPC) of MPEG video traffic is critical. As the widespread leaky bucket policing algorithm has proved to be unable to control the MPEG traffic, we investigate the performance of a new mechanism based on fuzzy logic as a policer for MPEG video transmission. The simulation results given demonstrate that the fuzzy policer performs very well in reducing both cell losses due to UPC and their effects.

An assessment of resource exploitation using artificial intelligence based traffic control strategies

We assess the application of artificial intelligence techniques to the complex problem of traffic control in ATM networks. The paper deals with the close link between call admission control and usage parameter control and proposes a simulation-based analysis to demonstrate how inefficiency on the part of policing affects bandwidth allocation. To take this into account, the paper proposes a framework for traffic control in which the CAC and policing functions are both based on artificial intelligence techniques, i.e. neural networks and fuzzy logic. In this way it is possible to train the neural network in such a way as to take into account the real behavior of the policer. As the results obtained show this allows us to implement traffic management strategies which can improve the exploitation of network resources.

An integrated framework for traffic control in ATM networks based on soft-computing techniques

Abstract In this paper we assess the application of artificial intelligence techniques to the complex problem of traffic control in Asynchronous Transfer Mode (ATM) networks. The paper deals with the close link between Call Admission Control (CAC) and Usage Parameter Control (UPC) and proposes a simulation-based analysis to demonstrate how inefficiency on the part of policing affects bandwidth allocation. To take this into account, the paper proposes a framework for traffic control in which the CAC and policing functions are both based on artificial intelligence techniques, i.e., neural networks and fuzzy logic. A measurement-based CAC mechanism is then proposed and implemented by means of neural networks trained in such a way as to take into account the real behaviour of the policer. As the results obtained show, this allows us to implement traffic management strategies that can improve the exploitation of network resources.