Reliable multicast transmission

Motivation

As today users require more bandwidth and the frequency spectrum is a limited resource, it is necessary to search for underexploited features of current wireless technologies that could optimize the usage of spectrum efficiency. Modern wireless standards provide wireless multicast access, which represents a good mechanism for reducing the bandwidth. The utilization of multicast networks allows to send content to different users using a single transmission operation. Thus, depending on the service, multicasting results more efficient than unicasting, which generates more traffic in the network. In this way, multicast networks are highly recommendable when there is sufficient number of users interested in receiving the same contents. Multicast networks are used to broadcast both video streaming and files. In this sense, the Quality of Experience (QoE) perceived by the users is an important parameter in the evaluation of both video streaming and file transmission services. In the first, the delay, the losses and the video quality are key aspects for the QoE. Regarding file transmissions, a good QoE is obtained when the file is received correctly with the minimum download time.

In file transmissions users need to receive correctly all the packets that compose a file in order to download it successfully, so channel losses should not occur. However, multicast transport does not guarantee, generally, error free communication and so it is needed to provide protection against errors. Additionally, some multicast file transmissions lack of a feedback channel that can be used by clients to report the servers about the status of their downloads. Therefore, clients are not able to ask for packets lost during the transmission.

In the absence of retransmissions in the transport layer, it is necessary that multicast protocols provide reliability at the application layer to overcome possible packet loss in the communication channel. Among the existing protocols, one of the most used by the different standardization organisms is FLUTE [1].

Overview

FLUTE (File Delivery over Unidirectional Transport) is highly used for delivering multimedia content in unidirectional environments in a reliable manner. One of the key elements of FLUTE is the use of a File Delivery Table (FDT), an in-band mechanism used to inform clients about the files (and their characteristics) transmitted within a FLUTE session. Clients need to receive the FDT in order to start downloading files. In this sense, the delivery of FDT packets and their proper configuration parameters have a great impact on the QoE of FLUTE services.

Figure 1 shows a general overview of a file transmission using the FLUTE protocol. A FLUTE server has a repository of multimedia contents. For example, it could send video, audio, images, documents… The FLUTE server broadcasts the contents in a certain session (identified by the IP address and the TSI), which contains, at least, one delivery channel (identified by the port number). On the other hand, clients follow these steps:

Step 1. Clients obtain through an out-band mechanism the Session Description that contains the transport parameters associated to the session. The way clients obtain the Session Description is independent of FLUTE.
Step 2. Once the clients have connected to a certain session they have to wait until they receive the FDT that describes the files (and their corresponding metadata) that the server is sending.
Step 3. Then, clients are able to identify the data packets they are receiving and they are able to download the files they are interested in.

Figure 1. File delivery using the FLUTE protocol

FLUTE provides reliability using different protection mechanisms. It should be noted that, generally, there are two main error correction techniques, ARQ (Automatic Repeat Request) and FEC (Forward Error Correction). The former consists of retransmitting data that are missed in the communication, whereas FEC allows to reconstruct the original data without retransmissions, through error correction encoding. FEC is mainly used in unidirectional environments, where a return channel does not exist. In this sense, FLUTE works over a FEC block, which is used to protect the file delivery service. Although error correction is generally applied in the lower layers of a communication system, it can be used at higher layers. Specifically, AL-FEC (Application Layer FEC) provides additional robustness to certain services without any modification in the lower layers of a system, through applying FEC coding at transport packet level. Thus, the use of AL-FEC is particularly interesting for provisioning new services over communication networks already in place, since AL-FEC can increase the native reliability of the network to meet the requirements of a specific service, without additional infrastructure [2]. Moreover, AL-FEC may improve the performance of content transfer through wireless communication networks, as it can decrease download times as well as network traffic, since it avoids the request of lost packets. The performance of AL-FEC is somewhat dependent on the complexity of the algorithm used to protect the information. There are different categories of FEC codes: convolutional codes, block codes, fountain codes and hybrid systems. In this sense, the most advanced algorithms fall in the category of rateless codes and perform very close to ideal FEC codes: no matter what is the erasure rate of the channel, receivers need only to acquire an amount of data equivalent to the size of the original file to be able to restore it. Nevertheless, rateless codes require more processing to generate the parity data for a specific file than other AL-FEC codes, such as LDPC (Low Density Parity Check). In environments where the multicast content selection is dynamic, it may be impossible to generate the parity data and insert it in the network on time [3].

LDPC AL-FEC codes provide a good trade-off between performance (download time) and complexity (time required to generate parity and time required to do the decoding process) [4]. LDPC AL-FEC parity can be generated nearly in real time but, unlike rateless codes, the optimum code rate depends on the erasure rate of the channel. When the code rate of LDPC AL-FEC is adjusted to the actual loss rate of the channel, LDPC AL-FEC [5] codes achieve a performance comparable to rateless codes in dynamic loss environments, especially when adaptive LDPC AL-FEC codes are used. Adaptive codes allow servers to send data at an optimum code rate for the channel losses experienced by the clients (watch video).

Moreover, in environments with limited bandwidth the performance of adaptive codes for file transmission can be improved by choosing the best protection that benefits the major part of users.

In this way, the use of file transmission services through multicast networks by employing an appropriate configuration can result very efficient, reducing considerably a resource so valuable as the bandwidth is but without degrading the Quality of Experience of users, as the related publications of our research group in this field [6]-[16] prove.

References

[1] T. Paila, R. Walsh, M. Luby, V. Roca, and R. Lehtonen, “FLUTE – File delivery over unidirectional transport,” IETF RFC, vol. 6726, Nov. 2012.
[2] H. T. Chiao, S. Y. Chang, K. M. Li, Y. T. Kuo, and M. C. Tseng, “WiFi multicast streaming using AL-FEC inside the trains of high-speed rails,” presented at the IEEE Int. Symp. on Broadband Multimedia System and Broadcasting (BMSB), Seoul, Korea, Jun. 2012.
[3] J. Peltotalo, J. Harju, and M. Hannuksela, “Reliable, server-friendly and bandwidth-efficient file delivery system using FLUTE server file format,” presented at the IEEE Int. Symp. on Broadband Multimedia System and Broadcasting (BMSB), Bilbao, Spain, May 2009.
[4] E. Paolini, M. Varrella, M. Chiani, B. Matuz, and G. Liva, “Low-complexity LDPC codes with near-optimum performance over the BEC,” in Proc. Advanced Satellite Multimedia Systems (ASMS), Bologna, Italy, Aug. 2008, pp. 274-282.
[5] V. Roca, C. Neumann, and D. Furodet, “Low Density Parity Check (LDPC) Staircase and Triangle Forward Error Correction (FEC) Schemes,” IETF RFC, vol. 5170, Jun. 2008.

Related publications

[6] R. Belda, I. de Fez, F. Fraile, V. Murcia, P. Arce, and J. C. Guerri, “Multimedia System for Emergency Services over TETRA-DVBT Networks,” in Proc. of the 34th EUROMICRO Conference on Software Engineering and Advanced Applications (SEAA), Parma (Italy), Sep. 2008, pp. 142-149.
[7] F. Fraile, I. de Fez, and J. C. Guerri, “Modela-TV: Service Personalization and Business Model Management for Mobile TV,” in Proc. of the 7th European Interactive TV Conference (EuroITV), Leuven (Belgium), Jun. 2009, pp. 27-30.
[8] A. Gil, F. Fraile, M. Ramos, I. de Fez, and J. C. Guerri, “Personalized Multimedia Touristic Services for Hybrid broadcast/broadband Mobile receivers,” IEEE Transactions on Consumer Electronics, vol. 56, no. 1, pp. 211-219, 2010.
[9] I. de Fez, F. Fraile, R. Belda, and J. C. Guerri, “Implementación y evaluación de la codificación LDPC para la transmisión de ficheros en entornos unidireccionales,” in Proc. of Jornadas de Ingeniería Telemática (JITEL 2010), Valladolid (Spain), Sep. 2010, pp. 229-236.
[10] I. de Fez, F. Fraile, R. Belda, and J. C. Guerri, “Evaluation of adaptive LDPC AL-FEC codes for content download services,” in Proc. of the IEEE International Conference on Multimedia and Expo (ICME), Barcelona (Spain), Jul. 2011, pp. 1-6.
[11] I. de Fez, F. Fraile, R. Belda, and J. C. Guerri, “Performance evaluation of AL-FEC LDPC codes for push content applications in wireless unidirectional environments,” Multimedia Tools and Applications, vol. 60, no. 3, pp. 669-688, 2012.
[12] I. de Fez, F. Fraile, R. Belda, and J. C. Guerri, “Analysis and evaluation of adaptative LDPC AL-FEC codes for content download services,” IEEE Transactions on Multimedia, vol. 14, no. 3, pp. 641-650, 2012.
[13] I. de Fez, F. Fraile, and J. C. Guerri, “Effect of the FDT transmission frequency for an optimum content delivery using the FLUTE protocol,” Computer Communications, vol. 36, no. 2, pp. 1298-1309, 2013.
[14] R. Belda, I. de Fez, F. Fraile, P. Arce, and J. C. Guerri, “Hybrid FLUTE/DASH video delivery over mobile wireless networks,” Transactions on Emerging Telecommunications Technologies, vol. 25, no. 11, pp. 1070-1082, 2014.
[15] I. de Fez and J. C. Guerri, “An adaptive mechanism for optimal content download in wireless networks,” IEEE Transactions on Multimedia, vol. 16, no. 4, pp. 1140-1155, 2014.
[16] I. de Fez, M. Gil, J. Fons, J. C. Guerri, and V. Pelechano, “A personalized system for scalable distribution of multimedia content in multicast wireless networks,” Multimedia Tools and Applications, vol. 74, no. 21, pp. 9595-9621, 2015.