Performance evaluation of bursty traffic using neural networks

Mehrvar, H.R.; Le‐Ngoc, Tho; Huang, Jun

doi:10.1109/ccece.1996.548312

Cited by 6 publications

(5 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…1) gives the Hurst parameter from equation 1. As shown from the curve, the Fractal effect [23] calms down, but does not disappear, for this reason we call it persistent Hurst phenomenon [24]. …”

Section: Review Of Hurst Model and Fractional Analysismentioning

confidence: 95%

Definition of Complex Hurst and Fractional Analysis for Stock Market Fluctuation

Zou

Huang

2015

Lecture Notes in Electrical Engineering

View full text Add to dashboard Cite

Abstract.Hurst model has received significant interest in recent years and are being increasingly used to explain the stochastic phenomenon with long term dependency such as stock market fluctuations. Different from existing methods in traditional integer dimension construction, this paper proposes a novel fractional dimension derivation along with the exact algorithm involving the fractional norm definition and the fractional center moment extension, which ends up a complex Hurst parameter. The proposed algorithm provides more granularities in terms of the norm and the center moment that eventually leading to the detail revealing of the underline stock marketing drivers from both real and imaginary angles. The calculation results demonstrate that the complex model is able to distinguish the subtle difference between the stock market performances for the same field that the real model may overlook. We take the e-commence cluster of online related companies as an example.

show abstract

Section: Review Of Hurst Model and Fractional Analysismentioning

confidence: 95%

Definition of Complex Hurst and Fractional Analysis for Stock Market Fluctuation

Zou

Huang

2015

Lecture Notes in Electrical Engineering

View full text Add to dashboard Cite

show abstract

“…However, the bandwidth will still drop dramatically after more than 10 hops in an ad-hoc network, which is not enough to support any good video quality. We noticed that the ad-hoc mesh network constructed by Strix [11] is the best found in the industry so far in this aspect. But at their best data rate, a GenieView [6] camera on such system would need to switch from video to image mode.…”

Section: IImentioning

confidence: 98%

“…The Hurst parameter is very useful here to capture importance of the peak to average ratio of compressed video, and much more, As shown from the curve, the Fractal effect calms down, but does not disappear. For this reason we call it persistent Hurst phenomenon [11] …”

Section: Fig 2: Pdf Of Video Tracementioning

confidence: 99%

See 1 more Smart Citation

Performance Tradeoff Study of Streaming Video among Vehicle

Lawal

Huang

Yang

2010

2010 IEEE 72nd Vehicular Technology Conference - Fall

View full text Add to dashboard Cite

Video and image communications can assist in highway traffic control, accident avoidance and management as well as road congestion control and coordination. This paper studies video streaming over wireless medium among a cluster of vehicles in an ad-hoc network. Through intense tradeoff study, we investigate the issues encountered in building a robust system that would allow automobiles to effectively transmit multimedia information in an ad-hoc network with constantly changing wireless environments. Based on our OPNET simulation results, we offer some design guidelines for transmitting streaming video over a vehicular adhoc network. I. INTRODUCTIONAs the number of vehicles on the road increased over the years, the need for safety on our roads cannot be over emphasized. VANET (Vehicular Ad-hoc NETwork) is a cluster of transporters equipped to communicate with each other either in an ad-hoc way (without any existing infrastructure) or with a nearby Road Side Unit (RSU) [2]. VANET has become an increasingly interesting focus of researchers and the automobile industry because it can enhance safety on the roadways by allowing vehicles to communicate with one another to prevent unwanted situations, thus enhancing comfort and safety of road users.As a part of the AUTO21 research group, our aim is to enhance safety on the road by using video and image communication to assist drivers without taking driving control from them [1]. VANET faces some difficulties to handle adhoc communication both among cars [2] and with Road Side Unit (RSU). The constantly changing vehicular environment and ad-hoc nature (governed mostly by the physical topology of the road and the proximity of nearby vehicles) of VANET poses the need for a careful study. Furthermore, information being sent in this network has to be real-time. Retransmissions may not be allowed as the delay could cause a lot of loss (which unfortunately is expected for a mobile adhoc network as route discovery time is inevitable. The study of tradeoffs among different routing algorithms in WiMAX Ad hoc networks is shown in [12]. To obtain reasonable performance, RSUs have to be much simpler than WiMAX or LTE base stations, and every tier in the OSI model has to be optimized for this purpose. One must also understand the traffic patterns in order to find an effective distributed transport layer protocol suite for this purpose. Therefore, we shall first focus on the video and image traffic, not only for the reasons given earlier,

show abstract

“…El Modelo Wavelet Multifractal (MWM) es un modelo de tráfico [16], [17] que se basa en la captura de las principales características estadísticas del tráfico en redes con gran eficiencia computacional [5], [6], [18], la herramienta principal de análisis de este modelo es la transformada wavelet discreta, la cual representa a una señal real unidimensional [19] construida en términos del cambio de desplazamiento y la versión dilatada de una función wavelet pasa bandas [20]; también se considera el cambio de una función de escala pasa bajos para algunas funciones wavelet y de escala seleccionadas, las versiones dilatadas y desplazadas y su base ortogonal así la señal puede ser representada [16], [17].…”

Section: Algunas Técnicas De Estimación De Tráficounclassified

Caracterización del tráfico en redes móviles ad-hoc basada en series de tiempo y redes neuronales recurrentes

Aponte

Fonseca

2014

View full text Add to dashboard Cite

El objetivo de este artículo es implementar una metodología de análisis y estimación del comportamiento del tráfico de paquetes en redes móviles Ad-hoc. Se construyeron procesos de estimación de tráfico en redes móviles con base en la implementación de redes neuronales recurrentes y series de tiempo; para ello se trabajaron 50 muestras de diferentes escenarios de caracterización de variables de entorno de forma aleatoria, simulados en NS-2 (Network Simulator Two), luego estas trazas fueron procesadas con la ayuda de herramientas como AWK y Perl. Se consiguieron históricos de 1300 y 3000 eventos para su posterior validación frente al estimador. Los estimadores y las caracterizaciones de tráfico implementadas para las redes neuronales, estan basadas en la regla Back Propagation y modelos de regresión lineal como ARIMA, SARIMA y FARIMA, para las series de tiempo.

show abstract

Performance evaluation of bursty traffic using neural networks

Cited by 6 publications

References 6 publications

Definition of Complex Hurst and Fractional Analysis for Stock Market Fluctuation

Definition of Complex Hurst and Fractional Analysis for Stock Market Fluctuation

Performance Tradeoff Study of Streaming Video among Vehicle

Caracterización del tráfico en redes móviles ad-hoc basada en series de tiempo y redes neuronales recurrentes

Contact Info

Product

Resources

About