Cross-Layer Optimization for Video Transmission over WLAN: Cross-Layer Prioritization

Cross-Layer Optimization for Video Transmission over WLAN: Cross-Layer Prioritization

Chih-Yu Wang (National Taiwan University, Taiwan), Yin-Cheng Huang (National Taiwan University, Taiwan), Cheng-Han Mai (National Taiwan University, Taiwan), Fu-Wang Chang (National Taiwan University, Taiwan) and Hung-Yu Wei (National Taiwan University, Taiwan)
Copyright: © 2012 |Pages: 27
DOI: 10.4018/978-1-4666-0960-0.ch020
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

As IEEE 802.11 wireless devices have become increasingly widespread, providing Quality of Service in the context of H.264/AVC, the video coding standard for future multimedia networking, has become an important issue in the fields of communication and networking. Cross-Layer Adaptive Video Prioritization (CAVP) is a cross-layer framework that prioritizes video frame transmission according to the application-layer information and the MAC layer transmission condition. In this chapter, a Peak Signal-to-Noise Ratio (PSNR) estimation method is proposed to sort out different priorities of H.264/AVC (Advanced Video Coding) video frames at the application layer to provide user-centric media quality estimation. Compared to previous heuristic algorithms, the authors also investigate a theoretic access delay estimator to monitor the wireless medium access delay at the MAC layer. In addition, an admission control is employed to serve the delay-sensitive video application and to give higher priority to those critical video frames. Video packets are dynamically classified into different 802.11e access categories according to the level of wireless medium access delay and the priority of the video frames. The myths of naïvely prioritizing video packets based on I/P/B types as well as naïvely assign packets to high priority access categories in 802.11e are resolved. Rather than creating complex scheme that is unable to be implemented in practical scenarios, the authors design the proposed scheme with practical implementability in mind. The proposed scheme is implemented with Click kernel module and the MadWifi WLAN driver. The performance of proposed CAVP design is evaluated by both NS-2 simulations and real testbed experiments, and results show that it enhances receiving video quality in error-prone wireless networking environments.
Chapter Preview
Top

Background

IEEE 802.11 WLAN

The basic medium access mechanism of the IEEE 802.11 Distributed Coordination Function (DCF) is Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA). An 802.11 wireless station senses the wireless medium before transmitting any packet. If the medium is busy, it applies a random backoff mechanism to defer the wireless medium access. The backoff counter is a random value between 0 and the current Contention Window (CW) value, which is initialized to the minimum window size. When the 802.11 station senses that the medium is free for at least the time duration DIFS (DCF Interframe Space), it begins to decrease the backoff counter. When the backoff counter value is 0 and the medium is still idle, the station can begin transmitting packets. After the packet transmission is done, the station which successfully received the data should send an Acknowledgement (ACK) frame back to the sending station. A Short Interframe Space (SIFS) waiting time is applied before replying the ACK message. The whole transmission procedure ends when the sending station successfully receives the ACK frame. If a data transmission is detected failed (no ACK is received), the backoff counter CW is doubled, until it reaches the maximum contention window size (CWmax). The CW value is reset to CWmin after successful data transmission.

In basic IEEE 802.11 operation, no quality of service is supported. All traffic contends for channel access with the same DIFS, CW, CWmax, and CWmin. Therefore, there is no difference between each packet. IEEE 802.11e (IEEE, 2005) is developed for QoS support in wireless LAN. The Enhanced Distributed Channel Access (EDCA) provides differentiated QoS support (IEEE, 2005; Mangold, et al., 2003). In EDCA, there are four Access Categories (AC): AC_VO for voice transmission, AC_VI for video transmission, AC_BE for best effort traffic, and AC_BK for background traffic. Each AC has different CWmax, CWmin, Arbitration Interframe Space (AIFS), which is similar to DIFS, and duration of Transmission Opportunity (TxOP). For high priority AC, smaller contention window parameters and AIFS value are applied.

To date, IEEE 802.11e has become a fundamental part of 802.11 protocols. It has been merged with IEEE 802.11a,b,g as a new standard of 802.11 (IEEE, 2007). It is also widely supported by most commercial products. Most subsequent 802.11 family members, including the high-throughput 802.11n (IEEE, 2009), retain the support of EDCA for compatibility. Thus, a cross-layer design based on the popular EDCA mechanism can provide high compatibility to both existing and future 802.11 infrastructures and mobile devices.

Complete Chapter List

Search this Book:
Reset