The policing function in an ATM network

Denissen, F.; Desmet, E.; Petit, Guido H.

doi:10.1109/digcom.1990.129366

Cited by 18 publications

(3 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Se requiere de un conocimiento bastante bueno del comportamiento y de las estadísticas que lo soportan [17], [18]. Parámetros como velocidad de bit pico, velocidad de bit promedio y longitud de la ráfaga (explicada como la duración de la velocidad de bit pico), son parámetros usados para caracterizar el trafico multimedia [18], [19]. Modelos matemáticos como Cadenas de Markov de tiempo continuo (CTMC), procesos modulados semi-markov (SMMP), y procesos de poisson modulados Markov (MMPP) [20] son usados para caracterizar y modelar el trafico.…”

Section: Voz En Redes Atm a Traves De Re-des Neuronales (Nn)unclassified

Caracterizacion Voz Y Video

Parra

Cuellar²,

Correa³

2011

redes ing.

View full text Add to dashboard Cite

ResumenLa motivación para caracterizar el tráfico de voz y video está en la necesidad de las empresas proveedoras de servicio en mantener redes de transporte de información con capacidades acordes a los requerimientos de los usuarios. Es importante poder determinar en forma oportuna como los elementos técnicos que hacen parte de las redes afectan su desempeño, teniendo en cuenta que cada tipo de servicio es afectado en mayor o menor medida por dichos elementos dentro de los que tenemos el jitter, las demoras y las pérdidas de paquetes entre otros. El presente trabajo muestra varios casos de caracterización de tráfico tanto de voz como de video en las que se utilizan una diversidad de técnicas para diferentes tipos deservicio. AbstractThe motivation behind characterizing voice and video traffic is represented in the necessity of the service provider companies to maintain information of the transport networks within their capacities according to the users' requirements. It is important to determine the way technical elements affect the performance of the networks in a timely manner, considering that services are affected (to a greater or lesser extent) by elements such as jitter, delays and loss of packages. The following article states various cases of traffic characterization, involving voice and video, where a number of techniques are used for different types of services.

show abstract

Section: Voz En Redes Atm a Traves De Re-des Neuronales (Nn)unclassified

Caracterizacion Voz Y Video

Parra

Cuellar²,

Correa³

2011

redes ing.

View full text Add to dashboard Cite

show abstract

“…The proposed congestion control algorithm consists of a critic part and a NN controller part as shown in figure (2). The system to be controlled (the environment) is composed of the input traffic sources and the input multiplexer buffer at the access node t o the network (UNI).…”

Section: Reinforcement Learning Methodsmentioning

confidence: 99%

“…A major shortcoming of the currently available queuing models [2], [3] is that only steady-state results are tractable and consequently, any control algorithm tailored on the basis of such models can ensure optimal performance only under steady-state conditions. However, performance control methods that dynamically regulate traffic flows according t o changing network conditions require understanding of its dynamics, which is almost impossible in ATM networks.…”

Section: Introductionmentioning

confidence: 99%

Reinforcement learning-based neural network congestion controller for ATM networks

Tarraf

Habib

Saadawi

Proceedings of MILCOM '95

View full text Add to dashboard Cite

This paper presents a new approach to the problem of congestion control arising at the User-toNetwork Interface (UNI) of the ATM-based Broadband Integrated Services Digital Networks (B-ISDN). Our approach employs an adaptive rate-based feedback control algorithm using reinforcement learning Neural Networks ("s).The reinforcement learning NN controller provides an adaptive optimal control solution. This is achieved via the formulation of a performance measure function (cost function) that is used to, adaptively, tune the weights of the NN. The cost function is defined in terms of two main objectives: 1) to minimize the Cell Loss Rate (CLR), i.e., control congestion and 2) to preserve the quality of the voice/video traffic via maintaining the original coding rate of the multimedia sources. The results show that the NN control system is adaptive in the sense that it is applicable to any type of multimedia traffic. Also, the control signal is optimal in the sense that it maximizes the performance of the system which is defined in terms of its performance measure function. Hence, our novel approach is very effective in controlling congestion of the multimedia traffic in ATM networks.

show abstract

The use of hop-delay for policing in ATM networks

Lam

Turner

1996

Journal of the Franklin Institute

View full text Add to dashboard Cite

The policing function in an ATM network

Cited by 18 publications

References 3 publications

Caracterizacion Voz Y Video

Caracterizacion Voz Y Video

Reinforcement learning-based neural network congestion controller for ATM networks

The use of hop-delay for policing in ATM networks

Contact Info

Product

Resources

About