Video streaming is becoming one of the most important services deployed over telecommunication networks such as the Internet and triple-play operators’ networks. This service differs from the rest in being loss sensitive and highly delay sensitive and requiring a considerable amount of bandwidth in order to offer a smooth transmission of packets through the network. While upgrading network elements with quality of service and multicast transmission capabilities becomes prohibitive for most network operators, peer-to-peer (P2P) architectures appear to be smart and efficient solutions to the previous issues. Many different P2P systems have been proposed and deployed to offer reliable video streaming services. These approaches address issues such as multicast transmission, quality of service enablement, mobility robustness, and video distribution according to network and user device capabilities. This chapter gives an overview of the different issues related to performance and reliability in multicast streaming over wireless networks and presents several alternatives facing them, including amendments to the already existing multicast mechanism of data distribution, video scalability and how peer-to-peer networking can provide a cost-effective solution to such problems.
TopIntroduction
Media streaming services over IP networks such as videoconference and video-on-demand have enormously increased in popularity in the last decade. Providers of such services find multicast an enormously advantageous system for distributing video and audio data to their customers as it can increase the bandwidth occupation efficiency, although its deployment implies great investments due to the need for updates to legacy hardware. In addition, IEEE 802.11 and 3G networks have spread and evolved in such a way that they can offer access to these services through the widely available wireless infrastructure.
However, multicast transmissions are prone to considerably diminish the quality-of-experience of such services when deployed in wireless environments. It is widely known that wireless networks are characterized by signal strength fluctuation and vulnerability to interference. Moreover, in multicast sessions, the lack of effective proactive and reactive mechanisms intended to avoid transmission errors causes a noticeably poor performance in media streaming applications, which are quite sensitive to packet loss. Additionally, currently available communication devices have many different capabilities regarding network connectivity, display size and resolution, power and battery durability, data storage and processing features, thus, producing a highly heterogeneous network.
In order to alleviate such problems and provide a suitable quality-of-service (QoS) to all users, several video and audio compression algorithms and network protocols improvements have been developed. Thus, QoS in such services can be achieved by considering many factors, such as bandwidth management, multicast traffic acknowledgment, video scalability and peer-to-peer networking, amongst others.
Basic solutions to overcome multicast performance issues include packet duplication, usage of different media codecs and copying multicast traffic into unicast streams targeted to client devices. However, such solutions do not scale well or do not perform as expected in wireless environments, so specific approaches have to be considered. Such solutions include implementations of leader-based acknowledgment protocols (Li & Herfet, 2008) (Ding et al., 2004) (Miroll et al., 2010), NACK-based protocols such as NACK Oriented Reliable Multicast (NORM) (NORM) and more robust transmissions of data with additional Forward Error Correction (FEC) protection like those included in the File Delivery over Unidirectional Transport (FLUTE) protocol (Paila et al., 2004).
Furthermore, scalable video provides a number of benefits in terms of reliability, these mainly related to the heterogeneity of networks and devices where video services are being deployed. With Scalable Video Coding (SVC) and Multiple Description Coding (MDC), content can be offered to a huge amount of receivers with different computational, display and network access capabilities, while these can choose which enhancement layers receive and process them in order to play the video. Thus, SVC and MDC can be used to deploy new QoS-enabled video services to a many different devices.
As previously stated, multicast distribution of media is considered a scalable technique when compared with unicast and simulcast methods. Despite this, current networks do no fully support multicast transport given how expensive the network devices upgrade is. This situation has made network operators search for a cheaper, more scalable and reliable network transmission architecture without the drawbacks of multicast systems. As a result, the first considered approach was Content Delivery Networking (CDN), which reduces the probability of network congestion in streaming services by distributing contents through different servers along the network, then placing the source servers closer to the end users. However, this solution faces a number of problems. CDNs are not scalable since a large number of simultaneous connection requests can lead to servers' overload. Additionally, CDNs are based in the client-server model, so if the link between the two end points is broken there is no alternative to receive the video streaming data. Thus, deploying a reliable CDN platform for video streaming services is still prohibitive for most network operators and does not provide with important advantages when compared with multicast techniques.