Mobile TV! Still no joy

Mobile TV is floundering in one of its bastion (Korea ... other one being Japan). The following is from a report in telecoms.com:

With mobile TV services in the flagship market of South Korea floundering and with few signs that operators anywhere else have found a successful formula for launching such services, most operator and vendor delegates at the recent CommunicAsia Summit in Singapore struggled to find enthusiasm for the fledgling industry.

Some operators and vendors say that mobile TV should be subscription-based, to offer a reliable revenue stream; others say an ad-supported model is the most viable option; and still others argue that a combined pay/advertising approach is the way forward.

Figures from South Korea seem to suggest that both pay-based and ad-supported models have critical weaknesses, which would also apply in other markets in the region. A lot more experimentation and creativity from operators might be required to find the right model.
Those promoting the idea of a pay-based service say that only by charging for content can a business model work. They say operators must team up with content firms to acquire premium content - most particularly sports - that people will be willing to pay a monthly fee to view or even pay for on a per-view basis.

But this line of thinking seems flawed, given that there is a limited amount of blue-chip content for which people will be prepared to pay, most notably live sports events - such as English Premier League soccer games - or highlights of them.

The problem is, of course, that content-rights holders have become adept at exacting a premium price for key sports rights, meaning that mobile TV operators would have to recoup their heavy capital investment by charging high subscription fees.

This is a problem, since the high churn rate experienced by TU Media in South Korea seems to suggest that mobile TV subscribers are extremely price-sensitive.

TU Media subscribers pay just KRW13,000 ($12.60) a month for the service but have been leaving in droves after their initial one-year contracts finish, forcing the firm to offer significantly reduced subscription rates to keep subscribers from deserting the service.

TU Media's experience suggests that mobile TV subscribers will be willing to pay only so much for services and that although blue-chip sports content has a crucial role to play, operators must find a way to acquire the content without paying excessive prices.

On the advertising side of the debate, many delegates at CommunicAsia argued that an ad-based strategy would work best for mobile TV platforms but that operators would have to be extremely creative in their approach.

There is no magic bullet that will provide a successful business model, but there seems to be a reasonable possibility that an attractive model can be built if operators can match the largely young and technology-friendly subscribers viewing mobile TV on their handsets with advertisers desperate to reach such a market.

Intriguingly, conference delegates also discussed the possibility that broadcast-type mobile TV services might never fully take off in the region and that Multimedia Broadcast Multicast Service (MBMS) video streaming over high-speed HSPA and future LTE networks would dominate the market.

The debate has strong proponents on both sides. Many vendors back an MBMS approach, saying that experience shows that broadcast-style services are not what users are demanding and that the more-narrowly targeted VOD-style content being offered on HSPA networks is already proving hugely popular.

The pro-MBMS argument also runs that with HSPA/LTE networks already in place and offering voice, data and video services, why go to the expense of deploying a terrestrial or satellite-based mobile TV network, especially with the expense involved in creating high-quality in-building reception?

Although this is a persuasive argument, it has shortfalls, most notably the fact that even LTE networks will still be point-to-point networks and will be unequipped to operate as point-to-multipoint services, which a full broadcast mobile TV service would require.
The broadcast-mobile-TV lobby argues strongly that the core strengths of broadcast-based networks cannot be replicated by even high-speed mobile networks, which would not be able to support the huge demand that's sure to arise for broadcasts of live sports and important news events.

In reality, the MBMS-vs.-broadcast-mobile-TV debate is spurious, given that both technologies are going to be on the market, and it will be users who determine which is the more successful.
At this early stage, it looks likely that subscribers and operators will use high-speed, quality video streaming for VOD-based "snacking" on content and that full broadcast mobile TV will be used for some live events, for which only a broadcast-style service can supply the quality of service required.

Korean Insight has an interesting section on Mobile TV (but no blogs on this topic for some time). A blog on this topic last year says a lot:

As TU Media started operations in mid 2005 it tried to acquire simultaneous re-transmission rights from broadcasters. This means that S-DMB viewers would be able to watch popular dramas and shows simultaneously with fixed TV. These contents are considered the most popular on both fixed and mobile TV. However, previously have broadcasters been reluctant to share these contents because they wanted to use it for their own T-DMB service. This is why S-DMB had to focus on other contents like sports and news. But the lack of “killer” contents from fixed TV hindered S-DMB development (as shown in the graphic above). Until today it had been able to acquire approximately 1.26 million subscribers. But according to TU Media they need approximately 2.5 million subscribers to be profitable.

But also T-DMB is struggling to build a profitable business. Despite more than seven million T-DMB devices in Korea the advertising revenues are marginal. Which partially is the result of very restrictive legislation on advertising but also broadcasters have failed to develop an attractive mobile advertising value proposition to make this channel more attractive for advertisers.

Consumers have embraced this new medium and it is very likely that broadcasters will take mobile TV more serious and endeavor to make mobile TV advertising more attractive for broadcasters. Until 2012 more than 20 million T-DMB devices are expected, so mobile TV has a future in Korea.

E-MBMS out of Rel-8, CBS in

E-MBMS is out of Release 8 and CBS (Cell Broadcast Service) is back in LTE. CBS as far as I am aware is not used much anywhere except Japan. CBS has been added specifically as now there is requirement for EWTS (Earthquake and Tsunami Warning System) .

MBMS - R.I.P.

Even though slow progress on MBMS (Multimedia Broadcast Multicast Service) has been going on for some time, just found out through sources that the two biggest promoters of this technology have put it on backburner. The reason they cite is the lack of interest from operators. They do not have a burning need for Mobile TV technologies as they are still able to cope with the demand by streaming point to point connections.

Somebody told me on condition of anonymity that the big operators in UK are at a breaking point but the things are still surviving because in the peak hours (9am to 5pm) there is not much demand for Mobile TV and the voice occupies the complete bandwidth. Whereas after 7pm and before 7am there is an even distribution of data and voice services. In the buffer zone (7am to 9am and 5pm to 7pm) data is being given low priority and many data calls dont work. This would cause decent revenue loss except that most of the people on data plans have a flat rate package so it does not bother the operators.

This is despite the announcement last month about Huawei and Qualcomm successfully completing their IOT with Telecom Italia. Orange and T-Mobile has been trialling MBMS based on TDtv technology but lets accept the fact that it is TDD-MBMS rather than the FDD-MBMS which other manufacturers like Qualcomm, Nokia and Ericsson are (were) actively working on.

I read this blog some days back and it emphasised what I have been saying for years now that there will always be multiple technologies floating around. MBMS could be a starting point for Mobile TV but as the demand grows it will have to be supplimented by other specialised technologies like DVB-H, DMB-T, MediaFLO, etc, etc.

For the time being, rest in peace MBMS.

Mobile TV Technologies comparison

Saw this new book on Mobile TV "Handbook of Mobile Broadcasting".

Mobile TV has been discussed for long time now but its surprising to see that none of the actual broadcast technologies is being actively used. There are small pockets here and there but no proper deployment. Here is UK, Mobile TV is actually TV on demand which is streamed onto our mobiles. Is it much different in other places? I did write a blog earlier titled '2008 may finally be the year of Mobile TV'.

The book mentioned above gave an interesting comparison of the 4 main technologies which is shown above. I would have liked it to expand it slightly by including DVB-SH and S-DMB.

Finally, heard that ALU trying to do some work on DVB-SH. See this.

And the winner is . . . DVB-H

Brussels has now officially endorsed DVB-H as the mobile TV technology of choice in Europe. This means that member states are now required to "encourage" use of the technology, though the commission has no advice as to how to encourage punters to tune in.

In the UK both T-Mobile and Orange are about to launch trials using the competing MBMS (Multimedia Broadcast Multicast Service) technology, which utilises existing 3G networks and spectrum. The technology for that trial is being provided by NextWave Wireless, and CMO Jon Hambidge is dismissive of EU attempts to mandate a mobile TV technology "when [the network operators] spent billions of dollars on their licences MBMS [was] part of that business case".

Note that in an earlier blog I had mentioned that Mobile TV and MBMS will co-exist. See here.

Viviane Reding, EU telecoms commissioner, has made it clear that if companies don't migrate to DVB-H she'll use regulatory measures to create an EU-wide standard.

Background Material:

The DVB-H standard is a recent extension of the DVB-T standard. It is intended to allow reception of television programs with portable and mobile terminals of relatively small size (the H of DVB-H means “handheld,” which indicates the primary type of devices targeted).

In most cases, the terminal will be a mobile phone. In fact, one of the main goals of DVB-H is to avoid the limitation inherent to UMTS of the number of terminals which can receive the same broadcast television program at one time. The main extensions of DVB-H compared to DVB-T are as follows (their use is signaled by specific TPS bits):

• addition of a 4 k COFDM mode, better suited to the implementation of SFN networks of medium cell size and allowing a reduction of the power consumption of the terminal compared to the 8 k mode;
• addition of a longer time interleaving (double for the 4 k mode and quadruple for the 2 k mode), which improves the behavior in case of signal fading and resistance to impulsive noise;

• transmission of a given service in periodic bursts by a process known as “time slicing” which permits a subscriber to activate the receiver only during a fraction of the time (5 to 10%) in order to reduce the power consumption, thus increasing the battery operating time;
• the ability to increase robustness by means of an optional additional link layer error correction (MPE-FEC) to improve the reception with an integrated antenna of necessarily very limited performances.

In order to allow the best use of these extensions, TV programs or other broadcast services are transmitted to mobile terminals as elementary streams (ES) formatted as IP (Internet Protocol) datagrams. The use of the IP protocol is, however, different from the one in TV by ADSL using DVB-IP: in DVB-H, the IP datagrams are encapsulated according to the so-called multiprotocol encapsulation (MPE) and then inserted in an MPEG-2 transport stream for transmission (in DVB-IP, it’s the transport stream which is IP encapsulated). This operation consists of encapsultaing the IP datagrams in DSM-CC sections by adding a header and a CRC termination. These sections are then segmented into MPEG-2 transport packets.

In order to realize the desired time-slicing, sections are not transmitted immediately, but are accumulated in order to form records of a maximum size of 191 kb, which will correspond to the duration of the time slice allocated to a service. These records can be represented as a table of 191 colums by a maximum of 1024 rows on which an optional additional error correction called “MPE-FEC” can be applied. This MPE-FEC consists of a Reed–Solomon coding RS (255,191) applied to words of 191 bytes made of the lines of this table. This will produce a second table made of an RS word of 64 bytes for each line of the original table. The result will be a new table of 255 colums by a maximum of 1024 lines which will be read column by column for transmission.

The DVB-H standard can be used in the UHF TV band with usual DVB-T channel widths (6, 7, or 8 MHz, depending on the region) or in other frequency bands (e.g., L-band in the United States around 1.67GHz with other channel widths, 5MHz in this case).

One of the problems with the use of the UHF band for TV reception in a GSM phone is the proximity of the high part of the UHF band (up to 862 MHz) to the GSM 900 transmit band of the terminal (880 to 915 MHz). Taking into account the current filtering capabilities, this prevents in practice the possibility of using the high UHF channels (>750 MHz) in a TV receiver integrated into an operating GSM phone.

The DVB-H standard can in principle use all the combinations of modulation parameters allowed by the standard (QPSK to 64-QAM, etc.) but, given the required robustness of this application, in practice only QPSK and 16-QAM with FEC of 1/2 or 2/3 are realistically usable, which permits bit-rates of 5 to 11 Mb/s in an 8MHz channel (without MPE-FEC correction). The video encoding used will be mainly H.264 with a CIF or QCIF resolution and bit-rates in the order of 256 to 384 kb/s.

Various experiments took place in Europe from 2004 onward to test the technical performances of the system in order to define the characteristics of the network, and to find out the user acceptance and expectations in order to establish a viable business model. The reactions of the test sample have been positive or enthusiastic everywhere. The first commercial DVB-H services started in Finland and Italy in mid-2006.

For more information see:

Do I hear TDtv again?

From Fox Business:

BARCELONA, Spain, Feb 12, 2008 (BUSINESS WIRE) -- NextWave Wireless Inc. (formerly IP Wireless) a global provider of mobile multimedia and broadband technologies, today paved the way for handset manufacturers to easily participate in the global market for TDtv by announcing a TDtv Device Integration Pack. Designed to meet the strict cost, size, and power consumption requirements established by some of Europe's largest mobile operators, NextWave's TDtv Device Integration Pack includes a low-power TDtv System in Package (SiP), a complete MBMS software stack, and MediaFusion multimedia client software from PacketVideo Corporation (PV). NextWave's turnkey solution will enable device vendors to deliver TDtv handsets to market in 2008 in support of the TDtv initiative announced today by NextWave, Orange and T-Mobile UK.

"Our TDtv device integration module provides a simple and inexpensive way for device manufacturers to integrate the power of TDtv into their next-generation WCMDA handsets," said Dr. Bill Jones, CEO of NextWave Network Products. "We're confident that a growing number of device manufacturers will support TDtv as mobile operators begin to accelerate wide-area deployments of TDtv systems."

TDtv is an innovative 3GPP MBMS solution that provides mobile operators an opportunity to profitably deliver up to 28 high-resolution TV channels, digital audio channels, and other IP data-cast services to an unlimited number of concurrent customers. By operating on existing 3G spectrum and with the unique ability to support multi-carrier spectrum pooling and network sharing, TDtv represents a breakthrough in reducing the cost of implementing mobile television and provides UMTS operators around the world with a powerful multimedia and advertising platform. Currently, more than 150 mobile network operators in over 50 countries have access to the spectrum needed to deliver TDtv to more than half a billion subscribers.

"When we looked at the available mobile TV technologies, TDtv was one of the technologies that impressed us the most, both from a performance as well as from a market opportunity perspective," said Michael Thelander, CEO of Signals Research Group, LLC. "With spectrum available across Europe and many parts of Asia and with two major operators now moving forward with an initiative, this is a market that handset vendors should take the time to explore."

NextWave's TDtv Device Integration Pack includes everything device manufacturers need to create a TDtv- enabled handset. The Device Integration Pack includes an integrated TDtv System in Package (SiP) measuring approximately 10x10 millimeters, a complete software suite that includes the required MBMS software stack, and TDtv radio access network controller software. The TDtv SiP includes a TDtv baseband chip, RF chip, receive filters and passives, and interfaces directly with the handset's application processor. This low power (active power consumption under 60 mW) solution enables device manufacturers to better meet network operator requirements for sleek and highly-attractive mobile TV handsets with internal antennas, and includes filters that allow for seamless coexistence with existing 2G and 3G services.

Meanwhile in Guardian:

T-Mobile and Orange will today announce a partnership to run a commercial trial in west London of a new mobile TV technology which could allow handset users to tune in to up to 100 channels.

The technology, TDTV, has been developed by US-based NextWave Wireless at its British unit in Chippenham, Wiltshire, and could provide a cheaper and more efficient way to get broadcast TV on to mobile phones. The trial, due to start in late summer, will see several thousand Londoners given either a new handset - made by a far eastern manufacturer rumoured to be LG - or a wireless receiver, no bigger than a matchbox, which will transfer the channels to their mobile phones.

The six-month test will see Orange and T-Mobile share their masts in London and install equipment that will allow them to broadcast 24 high-quality TV channels including several from the BBC and BSkyB, and 10 digital radio stations. It follows technical trials of the service carried out by Orange in Bristol last year. Orange and T-Mobile are also inviting O2, Vodafone and 3 to take part in the London test.

TDTV uses a slice of the 3G spectrum which Britain's five networks spent £22.5bn buying eight years ago and which has so far lain dormant. As a result, TDTV works with the phone companies' systems, making it easy to bill customers.

TDTV is more efficient and has more capacity for channels than other mobile TV solutions. Orange, T-Mobile, Vodafone and 3 are all offering mobile TV through their 3G networks but they suffer from congestion if more than a handful of customers use the service in the same place. TDTV uses a different part of the 3G spectrum and many more users can watch TV simultaneously.

The European Union has proposed using a Nokia-backed standard called DVB-H for mobile TV in member countries, but there will be no spectrum available for it in Britain until the analogue TV signal is switched off in 2012, and the operators will have to pay if they want it.

DVB-H, which O2 tested in Oxford two years ago, can carry only about two dozen channels while TDTV could have up to 100.

Ericsson Networks to launch MBMS in March 2008

Ericsson Networks last month successfully completed IOT (Inter-operability Testing) with EMP (Ericsson Mobile Platforms ... Handset people) and LG.

The plan is to have early launch systems in March 2008 followed by full deployment in September 2008. The only limiting factor might be the handsets which are not ready yet and have many problems that need to be sorted out.

According to CIO, Australians might be the first to get their hands on MBMS networks.

From what I have seen, Nokia is much ahead of its rivals in the MBMS game but they are keeping mum on this topic ... atleast for the moment. Sure it will be an interesting 2008.

SFN for MBMS in Release 7

MBMS in Release 6, can combine signals from different cells and this can be used to boost the reliability of the received signal. Though this concept is good, the other neighbouring cells are still causing interference since the adjacent cells are not orthogonal due to different scrambling codes. If the same scrambling code were to be used in all cells and the cells were to be synchronised, the other cells’ interference would turn into multipath
propagation. The received signal looks as if there is only a single base station transmitting, but with more multipath components. Therefore, this approach is called an SFN (Single Frequency Network).

Similar approach is also used in other Mobile-TV technologies like DVB-H.

The SFN approach is available in Release 7 and can be implemented with relatively minor modifications to the radio network while providing a major gain in the broadcast data rates. On the other hand, SFN transmission requires that the whole 5 MHz carrier is allocated for MBMS usage only, which requires that the total amount of spectrum for the operator is large enough.

Source: WCDMA for UMTS – HSPA Evolution and LTE, fourth edition. Edited by Harri Holma and Antti Toskala, John Wiley & Sons, Ltd.

TD-MBMS ready to roll out

Spreadtrum Communications Zhongxing Telecommunication Equipment (ZTE) have announced that the ZTE R&D Centre (Shanghai) has demonstrated TD-MBMS network services using Spreadtrum's TD-SCDMA/GSM/GPRS dual-mode chipset solution.

As a new TD-SCDMA multimedia service, TD-MBMS targets the mid to high-end segments of the 3G mobile market. TD-MBMS is now technically feasible on the TD-SCDMA network built using ZTE's equipment. With the adoption of Spreadtrum's SC8800D TD-SCDMA/GSM/GPRS dual-mode chip and platform, TD-MBMS could be applied to mobile phones, providing smooth images and clear sound. This successful demonstration of MBMS based on the TD-SCDMA standard indicates that the TD-MBMS technology is ready for commercialization.

This new TD-MBMS development is another step in the joint commercialization of TD-SCDMA by Spreadtrum and ZTE, following the strategic partnership agreement between Spreadtrum and ZTE announced on August 29, 2007. Dr. Datong Chen, CTO of Spreadtrum, said, ''We are very pleased to be a part of ZTE's successful demonstration of the industry's first TD-MBMS services. We believe this not only enriches the growing level of 3G TD-SCDMA multimedia services, but also enables TD-SCDMA mobile phones to satisfy the diverse requirements of its targeted users. In addition, Spreadtrum's TD- SCDMA/GSM/GPRS dual-mode chip and platform can equip the handset manufacturers with competitive technical advantages of TD-SCDMA mobile phones." Mr. Yuhong Duan, ZTE's General Manager of TD products, said, ''ZTE has been focused on the research and technical evolvement of the TD-SCDMA standard for several years, and it is the first to support TD-MBMS on the system side in the industry.

MBMS but no Multicast

From the discussions going on in the industry, it seems that MBMS is being implemented by everyone but no one is concentrating on Multicast. This means we should now change MBMS to MBS but ofcourse it doesnt sound as good ;).

In the discussions going in RAN5 on the testing side, Nokia, Qualcomm and Ericsson has indicated that they are not interested in Multicast at the meoment (and probably for long time) and Mobile TV using MBMS will only be using Broadcast services.

The main reason for Multicast not being implemented is that it is quite complex and the purpose it was designed to serve can be served by using authentication at the Application Layer which i think is far enough.

Frankly if operators are really interested in utilising the potential of MBMS (excluding Mobile TV) then they will have to use Multicast sooner or later.

Can anyone think of why Multicast would be better than application layer security?

MBMS battle heats up

The MBMS battle has started heating up with new developments and announcements nearly every week.

First of all there was this announcement from ZTE, China

China's ZTE Corp. announced that it has, in collaboration with Qualcomm, successfully completed a Multimedia Broadcast Multicast Service (MBMS) testing based on the 3GPP R6 standard conducted at its Shanghai R&D center last month.

The first testing MBMS inter-operability test (IOT) completed, test results show that 128- and 256Kbps high-speed transmission of MBMS TV programs broadcast and multicast services can be smoothly delivered over cellular networks. The success of MBMS IOT marks the readiness of commercial delivery to the mass market.

"Qualcomm is happy to cooperate with ZTE in conducting the MBMS testing, an area which we believe is part of the core development of ZTE's W-CDMA products," said Frank Meng, president of Qualcomm Greater China. "This is a major contribution in the development of MBMS and we are confident that the results of the testing will help our customers provide more competitive W-CDMA products and solutions moving forward."

Then yesterday, Anite submitted first MBMS Conformance Test which they were able to pass on Nokia NoRM-6 and Qualcomm 7200 UE. Other System Simulator manufacturers are not far behind with Anritsu and R&S also focussing heavily on MBMS test cases development.



Anritsu Protocol Test System (PTS) is shown in the figure above.

At present it seems Nokia and Qualcomm are leading the way on the UE side but in past i have heard about Ericsson and Motorola being MBMS ready soon.

'3' starts MBMS trials

The mobile operator '3' has again taken a major step forward by starting Mobile-TV trials using Multimedia Broadcast Multicast Service (MBMS). The trials are in conjunction with Ericsson. The following from the press release:

3 Italy uses DVB-H for its Mobile TV offering. However, MBMS is being developed under the auspices of the global 3GPP standards body.

MBMS uses existing 3G networks and spectrum for content delivery, building on existing infrastructure. To deliver MBMS, upgrades are made to the existing network as well as content and broadcast servers. 3G mobile phones with support for MBMS are expected to be available in 2008, said the company.

MBMS is a different approach to Mobile TV combining both broadcast and unicast shows (though multicast is more accurate). It also gives consumers opportunity to interact by voting, sending messages, accessing downloads of related content and special promotions from advertisers.

"MBMS as part of the 3G evolution is an attractive technology not only because of its flexibility and efficiency, but because it's quick and easy to deploy and leverages existing infrastructure," said Kursten Leins, Strategic Marketing Manager - Multimedia, Ericsson. "MBMS allows an unlimited number of users to watch the same mobile TV program at the same time in the same area, as well as enabling valuable user interaction with advertisements, campaigns and programs."

"3 pioneered 3G in Australia, so it was a great opportunity to see the country's first MBMS technology trial run on Australia's first 3G network," Leins added. Since the MBMS signal can be pin pointed to specific geographies, it's also possible to broadcast different mobile TV programs to different areas, giving a locally-specific customer experience and also providing highly targeted mobile advertising opportunities.

"We are very happy with the trial - the technology worked well and apart from delivering a good customer experience, it's extremely efficient in terms of network traffic and capacity, and provides new levels of customer involvement with their programs," said Michael Young, Director Technology & Services, 3 Australia.

The technical trial was held in 3's Sydney Head office over 6 weeks and run by Ericsson who developed the trial system. Using four specially designed prototype handsets, Ericsson also installed equipment to simulate the content and broadcast servers on a section of 3's network so the customer and network experience could be seen.

"The trial has been very interesting, and we'll continue to work with Ericsson to keep a close eye on the technology and the handsets to support MBMS as they develop," Young added.

More on MBMS can be found here.

C-Mobile: 3GPP MBMS for systems beyond 3G

Came across the C-Mobile website, searching for some information and the site caught my eye.

The strategic objective of C-MOBILE is to foster the evolution of the mobile broadcast business by providing enhancements to the 3GPP MBMS for systems beyond 3G.

Having worked with MBMS for some time and having completed atleast 4 trainings, the topic definitely holds my interest.

C-MOBILE will help to understand how best to organise and schedule MBMS content from the BM-SC through the core network, radio access, to the end users. Since this is multimedia content, interactions with the 3GPP IP Multimedia Subsystem are expected and C-MOBILE will explicitly investigate how best MBMS can make use of capabilities provided by the IMS.

The current concept of group communication is narrow within Release 6 MBMS specification.
C-MOBILE will research, investigate and define ways to use multicast technology to support personalized services and in particular the concept of multicast content community where users also contribute to the multicast service.

Key market and business requirements for multicast-broadcast services will be identified to aid defining research directions, leading also to new business models involving the various players.

To that end it is critical to understand the needs of multicast-broadcast users, network operators, and content providers.

The project intends to make important contributions to the standardisation bodies and to prove experimentally or via system level simulations innovative concepts.

There are no high profile names with C-mobile yet but there is Qualcomm and 3 UK in the participants list.

Some documents of interest are available here.

Push-to-share over MBMS

During one of my MBMS trainings last month in Anritsu, i set everyone a task of defining a service based on MBMS. One of the services mentioned was Push to Talk (PTT) over MBMS.

Theoretically it should be possible to use MBMS for PTT. The voice in the UL is sent via a normal CS RAB. On the Downlink the data is Broadcast using MBMS. Since we would like the data to be sent to a particular set of users, it would be Multicasted rather than Broadcasted. Also this would mean that only the users in a particular Service Area will be able to receive this.

From operators point of view, Service Area should be big enough so that the user is seamlessly able to move a wide geographical area. At the same time it should not be too big because localised services (and adverts) can generate more revenue.

A bit of Googling and i came across some patents that are trying to do the same. Following is an extract from a patent Fresh Patents:

[0009] Use of a PoC application server together with a multimedia
broadcast/multicast service (MBMS) server for providing multicast transfer of
data in the downlink direction has been suggested. In the uplink direction PoC
typically uses Real Time Protocol (RTP) traffic unicast. In the uplink PoC
clients send speech data to the PoC application server, which then directs the
speech data packets either to the MBMS for the downlink leg to those
participants who receive the speech via multicast service or directly to those
recipients who prefer to receive via unicast directly from the PoC server. Use
of multicast in downlink direction improves the spectral efficiency in the case
of group communications with great number of participants. In addition, without
multicast it may not be possible to support large group sizes, if the
participants are located geographically in the close proximity.

Another was a patent on free patents:

0032] Herein, the MBMS service refers to a service for transmitting the same
multimedia data to a plurality of recipients through a wireless network. In this
case, the recipients share one radio channel, thereby saving radio transmission
resources. For example, the MBMS service includes a stock information service,
sport broadcast service, Push-to-Talk (PTT) service, and the like.

In fact C-Mobile is working on something similar. One of their documents highlight the limitations of the current MBMS architecture and suggests how we can improve the architecture in future for B3G Architecture.

My personal feeling is that till the Architecture is eveolved enough, PTT may not be very practical over MBMS but we should be able to use Push-to-share over MBMS. Some interesting short video clip or Breaking News Clip or maybe personal Valentine messages, etc can be shared using MBMS. It now needs to be seen if some operator picks on this idea and how soon.

OMA seeks to ease mobile TV pain

The Open Mobile Alliance's recently-unveiled BCAST Enabler specification is designed to create a 'write once, run anywhere' environment' for broadcasters and other content providers. The spec - if widely adopted - could have significant implications for the concept of mobile TV 'roaming'.

In theory, it means broadcasters will be able to deploy their programming across the whole gamut of broadcast mobile TV platforms - DVB-H, DVB-SH, DMB, DAB-IP, ATSC-M/H etc - with little or no tweaking.

Because it works with any IP-based content delivery technology BCAST Enabler can also be used for the delivery of programming across cellular systems like 3GPP MBMS, 3GPP2 BCMCS and mobile unicast streaming systems, such as 3G streaming.

What benefits will OMA BCAST offer broadcasters and broadcast network operators?

• The specification enables broadcast-only mode for delivering services. It also allows broadcast-only terminals and free-to-air content with service and content protection capability.

• The specification is agnostic to access network meaning that the same service offering can be delivered over broadcast channel, interaction channel or both. Being agnostic to underlying architecture allows integration of the broadcast offering with operators or independent delivery over the interaction channel, which is controlled by broadcaster.

• Service interactivity is well specified and caters for broad range of services including interactive and direct feedback from viewers. Also, the service interactivity is not bound to the cellular channel – WLAN or a similar network can also be used. The use of the interaction channel allows personalization of services and service guides.

• The Service Guide enables the broadcaster to associate broadcast
programming with on-demand content. In addition, it supports both broadcast and on-demand delivery of the Service Guide itself.

What benefits will OMA BCAST offer terminal manufacturers?

• The Mobile TV Enabler specifies features for a common TV & video service layer that are currently not addressed by other specifications but still needed to ensure interoperability for large-scale terminal availability.

• Enables economies of scale by leveraging same technologies for both
broadcast and interactive channels. This means vendors can build an
economically viable terminal base that can be used by operators/carriers or broadcasters or jointly by both.

Mobile TV and MBMS will co exist

Someone brought my attention towards a Digitimes article where some people from Israel-based mobile chip designer Siano Mobile Silicon are talking about Mobile TV and MBMS. Some of the interesting points below.

Q: And what about multicast?

A: (Jashek) Again, multicast will end up placing a strain on the system bandwidth. The current MBMS (Multimedia Broadcast Multicast Service) capacity is limited to 2Mb/sec, while a broadcast system will provide bandwidth of 16-32Mb/sec, which is the bandwidth needed to support about 20 channels. Upgrading any existing cellular network so that it supports MBMS at 15-20Mbps (while not hurting the voice capabilities of the network) requires an investment that is by far larger than building a good mobile TV broadcast system.

We believe video-over-cellular services such as MBMS will continue to exist, but will gradually focus on “on demand” services, while actual mobile TV services will use a broadcast platform.
(Raab) Content will be broadcast to users, but users will be involved in the content, such as in programs that involve voting. And the way to create profits from this is to get more people involved in the service and bundling services to increase the amount of data that is going through the network, but in such a way that it does not strain the system.

Q: But who will build the broadcast infrastructure? Do you expect broadcasters and cellcos to be competitors or partners?

A: (Jashek) Most operators are facing the question of whether they should invest themselves or whether they should partner with a broadcaster to develop the infrastructure. In Italy, Telecom Italia Mobile (TIM) has deployed a mobile TV service where it is the service provider, even though Mediaset (a broadcaster) built the primary broadcast infrastructure. On the other hand, 3-Italia have made their own investment into a DVB-H network, and they enjoy a very good attach rate.

In the US, Qualcomm's subsidiary MediaFLO has solved this dilemma for the operators by building the network itself. The only thing Verizon or Cingular had to do was sign a contract with Qualcomm and offer the service.

Thus, different models exist. The relationship between broadcasters and cellcos will be one of the key issues affecting the success of mobile TV in the future. Most broadcasters already have the spectrum, as well as the content. They are currently using that for analog terrestrial TV, but in the future it will be used for digital mobile TV. However, cellcos already have a network that supports interactive programming. They also have an infrastructure in place for service and billing.

The question is how well can cellcos and broadcasters get along. What TIM has done, is take revenues from its mobile TV service and split it evenly with the broadcaster. In the future, we expect to see a similar type of model where broadcasters focus on broadcast services and operators focus on interacting with the customer.

Q: You mentioned that current analog TV spectrum will be allocated to mobile TV in the future. Can you add more color to that statement and explain how that will affect the development of the mobile TV market?

A: (Jashek) I should note that the development of mobile TV will go hand in hand with the migration of terrestrial analog TV to digital TV. For example, if you look at the DVB standard (DVB-T for terrestrial TV and DVB-H for mobile TV), which will be the DTV standard deployed in the most markets worldwide, currently about 30 countries have DVB-T networks, while another 30 will join in one to three years. Once the DVB-T networks are in place, you will see huge growth in DVB-H support because it does not take much investment to add DVB-H support to a DVB-T network.

Getting back to your specific question, a lot of countries have allocated spectrum to mobile TV on a temporary basis. Once governments start turning off their analog services in 2010, that spectrum will be allocated to mobile TV on a more permanent basis, and you will see a big jump in the size of the market.

We expect to see 120-130 million mobile TV users worldwide by 2010, with DVB-H being the number one platform. By 2012-2013 when more markets switch off their analog services, we expect to see 300-400 million people enjoying broadcast mobile TV.

Q: As you mentioned, DVB-H will be deployed in the most markets, however the global mobile TV market remains fragmented. Can you comment on the implications of how such a fragmented global market might affect the development of mobile TV?

A: (Raab) Obviously, with the huge expected size of the mobile TV market, a lot of different organizations would like to have a piece of the pie. Hence, a number of broadcast mobile TV technologies have been developed. Eventually, economy of scales will not allow more than about four technologies to survive in large volumes. It looks like the partitioning will be geographical.
(Jashek) DVB-H has its stronghold in Europe, where it was originally pushed by local players such as Nokia and Philips, and where DVB-T, the "mother" of DVB-H, has strong momentum. We have no doubt that DVB-H will dominate mobile TV in Europe, and DVB-T will also be supported on some hand-held devices. DVB-H is also expected to be the dominant standard in Southeast Asia – Taiwan, Singapore, Vietnam, Malaysia – and the Pacific Rim. In countries with vast rural areas, such as Russia or Canada, we expect that, around 2010-2011, DVB-H will be unified with DVB-SH (the satellite version of DVB-H). This will optimize the coverage with respect to the infrastructure investment required.

MediaFLO seems to be the winner in North America, although we would not be surprised if DVB-H will also be deployed there.

In Japan, as well as Brazil and a couple other South American countries, ISDB-T will dominate. And South Korea will continue with its T-DMB for some time, although being the only nation to have large-scale deployment of this standard will make it difficult for Korea to maintain it for many years. In China, the homegrown standard known as CMMB (S-TIMI) will be the main platform for mobile TV.

(Raab) Another thing to remember is that not only are the standards fragmented but so is spectrum support.

With the big picture being so unclear, device makers are looking for help to develop a solution that fits as many markets as possible. That’s why three years ago Siano came up with the concept of a multi-standard and multi-band mobile TV chip solution.

Our chips currently support the DVB-H/T, DAB and T-DMB standards, as well as covering the VHF, UHF, L1 (1450-1490MHz) and L2 (1660-1680MHz) spectrums. In addition, we will very soon have ISDB-T supported, while CMMB and MediaFLO are also on our roadmap. We are members of the CMMB working group, and the FLO Forum.

Q: Several mobile TV trials have been hampered by a lack of handset support, why is that?

A: (Raab) Handset makers need to digest and endorse a new technology – new types of antennas, receiver chips, software, etc. This is not easy. Some of the first few DVB-H phones were bulky, use antennae that were too long (making them unacceptable for most users), and have a reception sensitivity that was not that great.

The above diagram is from a Vodafone presentation ( Mobile TV from pure Broadcast to Interactivity, 19th Oct 2006 ). It shows how Mobile TV technologies will coexist with MBMS and the traditional unicast services

MBMS Enhancements in Release 7

MBMS will be undergoing enhancements in Release 7 and this work item is seperate from E-MBMS or Enhanced MBMS which is part of 3GPP Long Term Evolution (LTE).

MBMS is being enhanced in Release 7 and IMS will be able to use MBMS transport. The advantage of this approach is:

• MBMS reception is possible over IP accesses (e.g. I-WLAN)


• Higher MBMS bit rate services possible (e.g. HDTV)


• Support for adaptation of MBMS to the QoS resources provided by the access network(s)


• MBMS services will be available regardless of access technologies and other services will be able to usse MBMS transport

This MBMS Enhancement is still under development and the following is being investigated:

• Radio Interface Physical Layer: Introduction of new transmission schemes and advanced multi-antenna technologies


• Layer 2 and 3: Signalling optimisations


• UTRAN Architecture: Identify the most optimum architecture and functional splits between RAN network nodes
  

3GPP website lists some of the aims and objectives of these MBMS enhancements but theey do not look correct. They are copied from the LTE requirements documents. I will be revisiting this topic when more information is available