Background push content download services

Motivation

Nowadays, network operators dedicate a considerable amount of network resources to live streaming, through both broadcast and unicast connections. This live streaming service offering responds solely to commercial requirements: Content must be available anytime and anywhere. However, from a strictly academic point of view, live streaming is only a requirement for live content and not for pre-produced content. Moreover, broadcasting is only efficient when the content is sufficiently popular.

Background push download services use residual capacity in broadcast networks to push popular, pre-produced content to storage capacity in customer premises equipment. The proposal responds only to efficiency requirements. On one hand, it creates value from network resources otherwise unused. On the other hand, it delivers popular pre-produced content in the most efficient way: through broadcast download services.

Overview

Popular Content Delivery Services (CDSs) use either unicast (i.e. HTTP content download and HTTP video on demand) or application layer multicast (i.e. P2P content download) protocols to deliver the content. This means that these services do not benefit from the efficiency inherent to broadcast or network multicast, when the same content is delivered to several clients concurrently. This is because, these services have been designed to work across the Internet, where there is no end-to-end support for broadcasting or network multicasting.

On the contrary, CDSs that use broadcast or multicast connections are referred to as unidirectional CDSs and they allow all clients interested in the same content item to download it from a single connection, thus benefiting from the efficiency inherent to broadcasting. Broadcast CDS can be delivered over television broadcasting networks, like DVB networks. Additionally, any network delivering IPTV services provides end-to-end support for IP broadcast and content download services can benefit from this feature.

There are two different kinds of CDSs: push content download, where the delivery is initiated by the server; and pull content download, where the delivery is initiated by the viewer. Thus, pull CDSs generate traffic in the network after a client request, whereas the traffic generated by push content download services can be controlled on the server side.

Another characteristic common to the different CDSs is that they are pull CDSs and therefore, the bandwidth they use depends to a great extent on the user demand. Contrarily, unidirectional push CDSs are managed by the service provider, who is in control on the generated traffic.

In video broadcasting networks, video streaming services are delivered over reserved network resources of fixed capacity. However, video traffic is by nature variable with time and it follows that the reserved network resources are not used to their full extent at all times. This excess of reserved capacity can be used to provision other services with no instant capacity requirements, as long as they do not compromise the QoS of the streaming services. Clearly, unidirectional background push CDSs can benefit from this excess of reserved capacity, since they do not have any instant capacity requirements and the traffic generated can be controlled at the server side.

Hence, unidirectional background push CDSs send multimedia content to customer premises devices through existing broadcast connections, without interfering with the primary live streaming services. At the client side, the service stores the files delivered by the unidirectional background push CDS in local storage. The download process is transparent to the user: there is no explicit indication from the user on which files to keep in storage. Moreover, the service works as a prefetching cache, downloading files in the background before they are offered to the user. Next figure shows the architecture of the unidirectional background push CDS proposed.

Figure 1. Unidirectional background push CDS architecture

Unidirectional background push CDSs reuse the network infrastructure of television streaming services, while at the same time consume little (and unused) network resources. Therefore, these services turn out to be inexpensive to network operators. Hence, background broadcast push CDS can improve the efficiency of television content delivery with very little overhead in terms of operational expenditures and no need for additional infrastructure.

This kind of background services can be provisioned over any network with broadcast support and can play different roles in combination with other television services, either Linear TV services or VoD services. For the latter, the service could be used to push content to local storage. This way, the background service works as a prefetching cache, thus reducing both the traffic of the on demand service and the access time. Another use case, useful for both linear and on demand services, could be to detach the transmission of advertisements from the main transmission service or to send alternative content for playback under special circumstances, like no proper reception of the main service.

These use cases show that unidirectional background push CDS could provide value for different actors in the value chain. Content providers make more value of the resources dedicated to content delivery; network operators make a better usage of network resources; and television viewers benefit from improved Quality of Experience (QoE).

A list of publications of the research group in this field can be found in [6]-[11].

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] G. Zhiqi, Y. Songyu and Z. Wenjun, “Using Object Multiplex Technique in Data Broadcast on Digital CATV Channel,” IEEE Transactions on Broadcasting, vol. 50, no. 2, pp. 113-119, 2004.
[3] G. Adomavicius and A. Tuzhilin, “Toward the Next Generation of Recommender Systems: A Survey of the State-Of-The-Art and Possible Extensions,” IEEE Transactions on Knowledge and Data Engineering, vol. 17, no 6, pp. 734-759, 2005.
[4] S. Acharya, R. Alonso, M. Franklin and S. Zdonik, “Broadcast Disks: Data Management for Asymmetric Communication Environments,” Mobile computing, The Kluwer International Series in Engineering and Computer Science, vol. 353, pp. 331-361, 1996.
[5] H. Kellerer, U. Pferschy and D. Pisinger, “Knapsack problems,” Springer, 2004.

Related publications

[6] 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.
[7] 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.
[8] F. Fraile, I. de Fez, R. Belda and J. C. Guerri, “Evaluation of a background push download service for personal multimedia devices,” in Proc. International Conference in Consumer Electronics,  Las Vegas (USA), Jan. 2011, pp. 231-232.
[9] 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.
[10] F. Fraile and J. C. Guerri, “Simple models of the content duration and the popularity of television content,” Journal of Network and Computer Applications, vol. 40, pp. 12-20, 2014.
[11] F. Fraile, I. de Fez and J. C. Guerri, “Evaluation of background push content download services to mobile devices over DVB networks,” IEEE Transactions on Broadcasting, vol. 60, no. 1, pp. 1-15, 2014.