文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > 工程科技 > 信息与通信 > Integrated Application MAC Modeling for Cross-layer Optimized Wireless Video
侵权投诉

Integrated Application MAC Modeling for Cross-layer Optimized Wireless Video

802.11跨层优化

Integrated Application MAC Modeling for Cross-layer Optimized Wireless Video

Integrated Application MAC Modeling for Cross-Layer Optimized Wireless Video

Raymond S Wong1 Sai Shankar N2 Mihaela van der Schaar1

1

Department of ECE, University of California, Davis, CA, USA

2

Philips Research USA, 345 Scarborough Road, Briarcliff Manor, 10510, NY, USA

Abstract— Cross-layer optimization has been recently proposed for improving the performance of real-time video transmission over 802.11 WLANs. To perform cross-layer optimization, realistic analytical models are necessary that are able to accurately predict the interactions between layers, such that the parameters of the physical, MAC, transport and application layers are correctly identified. In this paper, we propose to modify existing IEEE 802.11 WLAN MAC models to explicitly consider how they impact and are impacted by the application layer in real-time video transmission systems. Specifically, we adapt existing MAC Distributed Coordination Function (DCF) models to consider the actual arrival rates of video flow packets, as well as provide adaptive retransmission limit for video packets with different priorities. Conversely, the video application can adjust the maximum retransmission limit per packet and the video rate based on realistic 802.11 MAC models rather than on simplistic two-state Markov models for packet-losses. Consequently, these integrated MAC-application layer models can be used to assist the cross-layer optimization for real-time wireless transmission by realistically predicting the interactions between layers.

Index Terms— MAC models, wireless multimedia, cross-layer adaptation, multimedia adaptation, robust video transmission, IEEE 802.11e.

I. INTRODUCTION

T

o enable efficient and robust transmission of multimedia over 802.11 wireless LAN (WLAN) networks [1], a formal understanding of the various transmission algorithms and protocols and their interactions is necessary. For this, realistic analytical models need to be build that are able to predict the performance impact of different cross protocol layer adaptation strategies under varying channel conditions, application requirements, and content characteristics. These models may then be used to drive the cross-layer optimization for optimal wireless multimedia transmission at each station as well as centralized or distributed resource management across stations. Currently, the performance of various OSI layer protocols is modeled in a “layered” manner that does not consider interactions between different layers. For instance, in [2], a model has been constructed to capture the features (such as back-off process etc.) and predict the performance of the 802.11 MAC Distributed Coordination Function in terms of throughput, packet-loss and delay, in a “layered” setting. Similarly, numerous multimedia traffic models have been proposed at application-layer, but these do not consider any interaction with the lower layers.

In this paper, we propose a new integrated framework for modeling the performance of MAC and applications layers involved in wireless multimedia transmission that can be used for cross-layer optimization. Constructing such “cross-layer models” is very challenging because the adaptation strategies at the various layers operate on different units of the multimedia traffic and take as input different types of information. For instance, the MAC layer is concerned with individual packet transmission and its performance depends heavily on the channel characteristics, while the application layer is concerned with multimedia flows, that consider the distortion dependencies between packets, and its performance depends heavily on the multimedia content characteristics. Moreover, while the MAC layer is delay-constraint agnostic and will attempt to forward all packets present in the MAC queue, the application layer will transmit only those packets that can arrive within the stringent delay constraints of video streaming applications. Consequently, rather than jointly modeling the MAC-Application layers, we propose to adapt existing MAC and application layer models to explicitly consider the impact of the other layer.

Previous research in the area of 802.11 WLAN MAC modeling [2][3][6] assumes saturated network conditions. However, in reality, the network condition will depend on the application layer transmission bit-rates. In this paper, we modify the analytical approach of [2],[3], and [4] and incorporate the arrival rate of the application explicitly in the MAC-layer modeling. By doing so, we capture the transmission characteristics of the MAC accurately, and in return, can accurately model the interaction between MAC and application layers. This integrated model is than used to assist the cross-layer optimization determining the optimal retransmission limits for the various video packets that leads to the maximum video performance (i.e. PSNR).

The rest of the paper is organized as follows. In section 2 we introduce the modified Markov model for the backoff process of 802.11 MAC that considers application layer characteristics. We also derive the optimal MAC backoff stage given application Quality of Service (QoS) constraints of delay and packet loss. In section 3, simulations results are shown using our model approach. Section 4 summarizes the conclusions.

第1页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email