Mobile TV

The buzz around mobile TV has been growing steadily. Following the launch of mobile TV via streaming technology over (3G) point-to-point connections, operators are now forming their plans for broadcast/multicast networks – necessary for cost effective delivery of video content on large-market scale. Dozens of trials have told the mobile communications sector that consumers enjoy mobile TV and, more importantly, that they are willing to pay for it — on average between €5 and €10 a month. Nokia added more momentum recently with a report from DVB-H pilot implementations in Finland, UK, Spain and France (with a total of approximately 1800 users).

Large scale success of mobile digital broadcast technology is not imminent however. Operators are still trying to grasp the business models and content format for mobile TV, and there is no technology standard for multicast/broadcast delivery. Since a single video user may consume up to 100 times the amount of data as a voice call (peak rate 13 kbps, average 4 kbps), while he or she may generate only approximately one quarter as much revenue, the cost per byte delivered needs to be an order of magnitude less than a voice cellular network. Several different technologies are in competition: DVB-H, FLO, MBMS/MCBCS, IPWireless’ TDtv, OMA BCAST, Korea’s DMB, and Alcatel’s S-band DVB-H announced in 3GSM 2006.

• DVB-H is largely based on the DVB-T specification for digital terrestrial television, adding to it a number of features designed to enable the receivers to be less power hungry, and the particular environments in which such receivers must operate. This power reduction has been achieved by time slicing so that the receiver is only switched on in those time intervals when viewing the channel of interest. These intervals could be anything between a few milliseconds and a few seconds. It therefore reduces power consumption by being switched off for the rest of the time when non-required data is being transmitted. There is therefore a trade off between the data rate required for the service and how much this can be packed into short bursts to save the battery power of the receiver. DVB-H uses Coded Orthogonal Frequency Division Multiplex (COFDM) with a bandwidth of either 6, 7, or 8 MHz. Additionally it uses a range of different types of modulation from QPSK up to 64QAM and this enables it to have a very high data rate. DVB-H also employs additional forward error correction to further improve the mobile performance of DVB-T. The specifications for IP Datacast are essential to the convergence of broadcast networks and mobile telecommunications networks that will almost certainly be central to the majority of commercial launches of DVB-H services. The level of vehicular mobility supported by DVB-H is unclear.
• Multimedia Broadcast and Multicast Service (MBMS) and Broadcast and Multicast Service (BCMCS) are the results of the work of 3GPP and 3GPP2 on broadcast/multicast services in GSM/WCDMA and CDMA2000 respectively. The specifications of mobile broadcast services were functionally frozen in 2004. MBMS and BCMCS introduce a set of functions that control the broadcast/multicast service (Broadcast/Multicast Service Center, BM-SC in MBMS), broadcast/multicast routing of data flows in the Core Network and efficient radio bearers for point-to-multipoint radio transmission in a cell. MBMS in 2G uses GPRS/EDGE PDCHs in with RLC/MAC with or without ARQ (PDAN or blind repetition respectively). In WCDMA a new physical/local channel structure is introduced, with soft combining for FACH. MBMS/BCMCS support full mobility.
• IPWireless’ TDtv system is MBMS implemented over UMTS TD-CDMA in the unpaired TDD band which is part of most operator’s spectrum license. TDtv is attractive because it avoids impacting other voice and data 3G services in the FDD band. On the other hand, terminal support of TD-CDMA and its frequency band is a burden. Vehicular mobility supported up to 200 kmph (and 400 kmph later this year).
• OMA BCAST is working on the specification of broadcast/multicast-related service-layer functionalities that can be applied to mobile and non-mobile digital broadcast networks. For instance, OMA BCAST addresses content protection, service and program guides, and transmission scheduling. It is agnostic of the underlying broadcast/multicast distribution scheme, which could be MBMS, BCMCS or a non-mobile digital broadcasting system such as DVB-H.
• Qualcomm‘s Forward Link Only (FLO, or MediaFLO) is an OFDM based air interface designed specifically for multicasting. FLO is said to support mobility up to 200 kmph, optimize terminal power consumption, frequency diversity, time diversity and low access times. It uses QVGA at 30 frames/s plus stereo audio, completed with compression technologies such as H.264 and AAC+ to support high quality video channels at 360 kbps average.
• Terrestrial – Digital Multimedia Broadcasting (T-DMB) is based on Eureka 147/DAB (Digital Audio Broadcasting), which has already had some success in Europe, as the broadcast channel to deliver TV, audio, video and data to mobile devices. South Korea, with key players Samsung and LG, is leading the development of T-DMB with commercial networks launched in 2005. Like DVB-H, T-DMB uses COFDM but on carriers with only 1.712 MHz bandwidth. T-DMB uses a large number of these carriers, each carrying low rate data. The carriers are arranged so that they are orthogonal and do not interfere with each other, which allows very efficient use of the bandwidth used. For DAB and DMB each transmission occupies approximately 1.5 MHz bandwidth and for the VHF broadcasts the transmission contains 1536 carriers.
• Last month, Alcatel announced a combination of DVB-H-based terrestrial and satellite broadcast delivery in the S-Band, a frequency band reserved for satellite usage immediately adjacent to the WCDMA spectrum and available across continents. For the broadcast of a significant number of mainstream TV channels, Alcatel proposes to reuse 3G radio sites and antennas and to extend them with S-Band repeater capability, providing optimized quality as well as indoor and urban coverage.

Guidelines and directions in the EU are unclear. Many European vendors and operators appear to favour the homegrown DVB-H standard. DVB-H requires new spectrum but the preferred radio channels are not available throughout Europe, which means certain countries like Finland, France, Spain, Italy and the Netherlands may adopt it commercially in coming months and years, but others may take much longer. On the other hand, MBMS allows European mobile operators to simply upgrade their existing network to offer broadcast/multicast services in their existing spectrum. Besides the reuse of their legacy network investments, mobile operators could also retain existing unicast TV business models. Opponents say MBMS could consume substantial mobile network capacity, even on WCDMA/HSDPA, and is an inefficient use of paired spectrum. Ericsson simulations for a GSM/EDGE bearer show that, if there are two users in the cell, the MBMS multicast bearer is as efficient as a regular unicast connection and becomes much more efficient with an increasing number of users.

The simulations indicate that video broadcasting at 40 kbps requires two to four timeslots with EDGE channel coding in PDAN mode depending on the number of users whereas four timeslots are needed in blind repetition mode. Note that a regular point-to-point EDGE channel can provide streaming at the same bitrate using two timeslots, but only to one user. For three users, six timeslots are required; for four users, eight, and so on.

On WCDMA the MBMS radio transmission cost is independent of the number of subscribers in the cell. Referring again to the same Ericsson study, with MBMS on WCDMA 3GPP R6 one 5 MHz cell carrier can support 16 multicast MBMS channels at a user bitrate of 64 kbps per channel (for vehicular A3 channel model, 80ms TTI and single receive antenna). Dual antenna receive diversity can double the MBMS capacity, allowing for 32 channels per cell carrier.

In Belgium the IBBT MADUF project (Maximize DVB Usage in Flanders) will soon kick off. The project will explore the possibilities for mobile television through DVB-H, gathering operators Telenet, Belgacom and Proximus, VRT as manager of the transmitter park and suppliers of communications solutions Siemens, Option and Scientific Atlanta. Research units from KUL, VUB, UG, and IMEC will participate as well. The objective is to study the economical, societal, legal and technological potential of mobile television. The basic technological research is focused on interactivity through mobile phone networks and specific encoding mechanisms. The results will contribute to the final standardisation of DVB-H. The project has no prospect for an imminent commercial launching of mobile TV. Indeed, with commercial launches of wireline IDTV lagging a few years behind other EU countries, it remains to be seen how quick digital and interactive television services will be adopted by the traditionally conservative Belgian consumer.