Print

Comment

Email

Digg

Del.icio.us

Reddit

There is presently a great deal of industry hype surrounding 'mobile television'. Said by some to be the next 'killer app' of the mobile sector, and dismissed by others as having no sustainable business model, mobile TV is a conjure of possibilities. It lies at the eye of a maelstrom of technologies, network models and frequency bands, waiting for many trials to end and the manifestation of a clue as to the most practical and commercially viable direction.

The notion of delivering television signals to a moving receiver is not an entirely new concept. A number of countries have for years enjoyed live digital television in buses and trains, courtesy of digital video broadcast terrestrial (DVB-T) technology. Utilising coded orthogonal frequency division multiplexing (COFDM) modulation, DVB-T was originally designed with mobile applications in mind. The signal can accommodate variations in signal strength, field strength and multiple reflections that typically reach a receiver in motion.

However, despite this foresight in its development, DVB-T as it stands is not appropriate for broadcasting to handheld devices. Nor is its US counterpart, the advanced television systems committee (ATSC) standard, utilising 8 Vestigial Side Band (8VSB) modulation, which was never designed for mobile applications.

Experiencing TV on handheld devices raises a whole new set of issues that have spawned several new broadcast technology platforms. Attracting the most attention globally is the digital video broadcast to handhelds (DVB-H) standard, which is derived from DVB-T. The important difference is that DVB-H transmits the signal in bursts in order to conserve handset battery life. It also incorporates greater forward error correction, essential for boosting handheld reception.

Another significant difference is the data encapsulation technique. The DVB-H stream is an IP datacast at 200 to 500kbps/program, yielding up to 50 programs in an 8MHz channel. This resolution is sufficient for the tiny handset screen. In contrast, standard-definition DVB-T uses MPEG-2 (or MPEG-4) encoding at 4 to 5Mbps/program, yielding up to five 'standard resolution' programs per channel.

DVB-H is not the only mobile TV platform finding favour. Korea and China are the first to embrace terrestrial digital multimedia broadcast (T-DMB), derived from the Eureka 147 digital audio broadcast (DAB) standard. Moreover, Qualcomm has developed the forward link only (FLO) technology for the delivery of multimedia content. T-DMB, FLO and DVB-H have each addressed the same handset-related issues--battery life, reception and screen resolution--albeit in different ways.

It Starts with Delivery
The choice of technology platform is just one element of delivery-and delivery just one consideration-in the riddle that is mobile TV. Commercial imperatives drive all, and are also dependent on such aspects as consumer viewing habits, handset development, content licensing and government regulatory environment. Yet it is with delivery that the whole mobile TV enterprise gets moving, and delivery infrastructure that represents a significant proportion of capital outlay. Consequently, the question of which delivery model proves best, and most cost-effective, is one of high interest.

Speculation is compounded by the existence of several different industry players. On the one hand, there are the mobile communications carriers. These have an existing subscriber base and perceive mobile TV as a means of extending and differentiating their service. Many have introduced third-generation (3G) mobile TV services based on universal mobile telecommunications service (UMTS) in recent months, while at the same time partnering broadcast-based mobile TV trials.

It is generally well accepted that UMTS-based mobile TV has limitations. The service is here and available now, but the unicast (one-to-one) nature of UMTS means that as the viewer base grows, mobile TV will not be sustainable on this platform--even as UMTS heads towards '3G long term evolution' (3G LTE) or in-band cellular broadcast techniques such as multimedia broadcast/multicast service (MBMS). Recent reports have suggested that it makes more sense to use the spectrum for wireless data services that can be charged at a higher rate than can television.

Mobile carriers are therefore turning to broadcast models for mobile TV. Their quest to utilise existing base station sites has led to the 'cellular overlay' model for mobile TV, where broadcast infrastructure is deployed at mobile base stations to provide mobile TV coverage in a similar way to a cell-based mobile network.

Coverage Adjustments
The broadcast industry approaches mobile TV coverage from the other direction. Conventional free-to-air TV is typically broadcast from centralised high-power transmission sites, supported by supplementary repeater or 'gap filling' stations. It is relatively straightforward to deploy a mobile TV service in the same manner; however, there do need to be adjustments to coverage planning.

Research indicates that the 'high-power terrestrial broadcast' model for mobile TV will require more repeater sites than for conventional television. One reason is because, owing to an increase in reflections at ground level, the forward error correction applied to the signal is increased, resulting in a trade-off in signal-strength that needs to be addressed. It has been reported that a receiver at ground level incurs a signal-strength penalty of approximately -12 to -16dB (depending on frequency band) compared with the average rooftop antenna.

Additionally, consumers have also come to expect their handsets to work indoors and in moving vehicles, each reported to incur another -8 to -12dB (or more) signal impact. The provision of indoor coverage is considered one of the main challenges of mobile TV networks.

A third infrastructure model, incorporating satellite blanket coverage supported by low-power terrestrial repeaters, has been proposed. The repeaters would be co-located at mobile base stations to supplement urban and provide indoor coverage.

A unifying element in all three network models is the convergence of industries that have been hitherto quite separate. Mobile carriers will need to embrace broadcast technology and content; broadcasters (or infrastructure/service providers) will need to team up with carriers, who already have the subscriber base. In fact, it seems logical for mobile TV systems to be intrinsically linked with mobile phone services, which can provide a one-to-one back-channel for interactivity. This could even prove to be a driver for consumer take-up.

The Band Debate
From a technical and practical standpoint, the other major delivery option pertains to frequency band, of which several are being considered: VHF (170 to 240MHz), UHF (470 to 860MHz), L Band (variable depending on region, but generally falls somewhere between UHF and S Band) and S Band (2170 to 2200MHz).

Most popular globally for digital terrestrial television is the UHF band, which has also seen the most mobile TV activity to-date. It has good propagation characteristics and, if deployed using the terrestrial broadcast model, should be capable of providing coverage of a large city using 20 to 50 repeater sites. Qualcomm in the USA is using this model for its commercial MediaFLO service (using the FLO platform), but, as other trials have shown, it is also ideal for deploying DVB-H.

The UHF band is also suitable for networks deployed using the cellular overlay model, since UHF frequencies are just below conventional global standard for mobile communications (GSM) or US 'Cellular' code division multiple access (CDMA) frequencies. This type of network is being trialled in many countries across Europe.

One of the main challenges associated with the UHF band is the limited availability of spectrum in most parts of the world, but especially Europe. Some governments are considering assigning two or three UHF frequencies for DVB-H mobile TV services, which can be deployed as single frequency networks (SFN). Although it makes network configuration more complex, an SFN is a highly efficient use of spectrum, and a network of two or three overlapping SFNs could be a promising option.

The VHF Band III has even better RF propagation characteristics than UHF. It is not suitable for the cellular overlay model, since the antennas would be too large for existing base stations; but it is an ideal candidate for the terrestrial broadcast model, where city coverage could be achieved with just a handful of repeaters. From a network deployment perspective, VHF would appear to offer the lowest roll-out costs coupled with the best indoor coverage.

Factoring in Availability
Korea and China are both deploying commercial T-DMB mobile TV services in VHF Band III, as per DAB services. To-date, there has been no move to deploy DVB-H in VHF Band III; however, since DVB-T services operate in VHF Band III, there seems little reason why DVB-H would not as well. The main obstacle is again one of spectrum availability-of the four considered bands it has the most limited availability in most countries-coupled perhaps with convention.

The other two bands-L Band and the satellite S Band-are emerging as contenders. Both provide reduced terrestrial propagation and in-building coverage compared with the lower frequency bands, but have the advantage of being more readily available. L Band looks set to support a commercial deployment of DVB-H mobile TV services in the US; S Band is that proposed to support a DVB-H based hybrid satellite/terrestrial repeater model.

Irrespective of which frequency band is selected, the signal polarisation is also under examination. The FLO systems being deployed use circular polarisation (CP), which is a combination of vertical (VP) and horizontal (HP) components. It has been speculated that a CP signal may facilitate reception at the mobile handset regardless of orientation. This may, however, be a moot point, since the multiple reflections experienced by HP and VP signals can alter the polarisation, effectively producing a mixture of polarisation components by the time the signal reaches the handset.

Vertical polarisation is favoured at present by both DVB-H trials and T-DMB deployments. In the latter case, this probably harks back to the DAB convention, since radio signals are often VP to enhance reception by car antennas. Use of VP also enhances isolation from HP television signals at similar frequencies. Most DVB-H trials are using VP, although at least one utilises a HP signal. Ultimately, the selection of polarisation will depend upon the receiver performance when faced with multiple signals from reflections, plus the indoor penetration of the signal.

Which Way Forward?
The future of mobile TV depends on many factors; but if it is proved that consumers want mobile TV-and are prepared to pay for it-then half the battle is won. The network model will then be determined by how cost-effectively networks can be deployed and the availability of frequencies and licenses. This is likely to differ on a case-by-case basis.

Utilising existing infrastructure will be a key element. It is not difficult to incorporate mobile TV services into existing broadband terrestrial broadcast systems, particularly if the systems were initially designed to accommodate additional services or channels. The most significant capital outlay would come with the deployment of additional repeater stations.

If, on the other hand, a mobile TV network is deployed as a cellular overlay, this will involve a significant shift in broadcast infrastructure philosophy. The quest to deploy television antennas at existing mobile base stations (hundreds, perhaps thousands, of sites) will encounter the same challenges as experienced by mobile phone carriers-the demand for low-profile, environmentally friendly antennas; the mandate for low emissions; site-by-site negotiations; and the trade-off between capex and opex. It could also promote utilisation of the higher-frequency L Band and its inherently more compact infrastructure.

Co-location interference issues also need to be considered when overlaying mobile TV and wireless communications services. With UHF frequencies so close to the GSM 900MHz receive band (usually 890 to 915MHz) and the CDMA 800MHz receive band (usually 824 to 849MHz), careful frequency planning and coordination will be required. Moreover, if the broadcast signal is too high in power, it could cause 'blocking' in the sensitive GSM or CDMA receivers, unless RF filtering is deployed. Similar situations arise with both the L Band and S Band frequencies, which are all in the vicinity of high-band GSM, CDMA and UMTS services.

In addition, it is likely that all mobile TV network topologies will ultimately need to incorporate dedicated wireless indoor solutions (WINS) to provide coverage inside multi-level buildings, large campuses (such as airports and shopping malls) and underground road tunnels and metro systems. These could be integrated with existing broadband WINS systems for mobile wireless communications.

True Convergence
Clearly, for mobile TV to succeed as a commercial venture, it will involve many players in the wireless sector: mobile phone carriers, broadcasters, handset manufacturers, content providers, infrastructure groups, base station OEMs, government and licensing bodies.

The quest to maximise the bottom line will ultimately reveal which network model, technology platform and frequency band combine to form the most viable option for a specific country or market. And it will be dependent on which provides the most attractive and accessible model for consumer uptake. Whatever the outcome, it will represent a true convergence of multiple technologies. From this will materialise the true meaning of mobile TV.

Mike Dallimore
VP, Broadcast, Towers and Defence Systems, Radio Frequency Systems
vadmail@cybermedia.co.in

Page(s)   1