TD-SCDMA, TDD and FDD

After my posting on TD-SCDMA so many people asked me about what TD-SCDMA is. I am surprised that so many people are not aware of TD-SCDMA. So here is a quick posting on that.

TDD and FDD Mode of Operation

Basically most of the UMTS networks in operation are Frequency Division Duplex (FDD) based. There is also another variant called the Time Division Duplex or TDD. In reality there is more than one variant of TDD, so the normal 5MHz bandwidth TDD is called Wideband TDD of WTDD. There is also another name for WTDD to confuse people, called the High Chip Rate TDD (HCR-TDD). There is another variant of TDD as would have guessed known as the Narrowband TDD (NTDD). NTDD is also known as Low Chip Rate TDD (LCR-TDD) and most popularly its known as TD-SCDMA or Time Division Synchronous CDMA.

"Synchronous" implies that uplink signals are synchronized at the base station receiver, achieved by continuous timing adjustments. This reduces the interference between users of the same timeslot using different codes by improving the orthogonality between the codes, therefore increasing system capacity, at the cost of some hardware complexity in achieving uplink synchronization.

The normal bandwidth of FDD or TDD mode of operation is 5 MHz. This gives a chip rate of 3.84 Mcps (Mega chips per second). The corresponding figure for TD-SCDMA is 1.66 Mhz and 1.28 Mcps.

Assymetric operation in TDD mode

The advantage of TDD over FDD are:

• Does not require paired spectrum because FDD uses different frequencies for UL and DL whereas TDD uses the same frequency hence its more easy to deploy
• Channel charachteristics is the same in both directions due to same band
• You can dynamically change the UL and the DL bandwidth allocation depending on the traffic.
  

The dis-advantage of TDD over FDD are:

• Switching between transmission directions requires time, and the switching transients must be controlled. To avoid corrupted transmission, the uplink and downlink transmissions require a common means of agreeing on transmission direction and allowed time to transmit. Corruption of transmission is avoided by allocating a guard period which allows uncorrupted propagation to counter the propagation delay. Discontinuous transmission may also cause audible interference to audio equipment that does not comply with electromagnetic susceptibility requirements.
• Base stations need to be synchronised with respect to the uplink and downlink transmission times. If neighbouring base stations use different uplink and downlink assignments and share the same channel, then interference may occur between cells. This can increase the complexity of the system and the cost.
• Also it does not support soft/softer handovers

Timing Synchronisation between different terminals

By the way, in Release 7 a new TDD mode of operation with 10 MHz bandwidth (7.86 Mcps) has been added. Unfortunately I dont know much about it.

You can read more about TD-SCDMA in whitepaper 'TD-SCDMA: the Solution for TDD bands'

You can find more information on TD-SCDMA at: http://www.td-forum.org/en/

Multi-Standards Radio Base Station (MSR-BS) in 3GPP Release 9

I wrote about Future Mobile Terminals earlier which will probably be Multiservice, Multinetwork and Multimode. A similar approach would be needed for the network side. 3GPP is working on Release-9 feature of Multi-Standard Radio (MSR-BS). The 3GPP Spec 37.900 is not yet available but a draft should be available soon.

Research and Markets have already released a report arguing about the benefits of MSR-BS. Last year Ericsson released the RBS 6000 series products that has MSR support. Huawei and Nokia Siemens Networks are also working on similar products under different guises. Martin has blogged about this topic as well earlier in case you want to refer to.

According to Research and Markets report the terms used for this technology is Multi-Standard Radio Base Station (MSR-BTS/MSR-BS), Multi-Mode Radio Base Station (MMR-BTS/MMR-BS) and Multi-Radio Access Technology (Multi-RAT). The name in standards usually is MSR-BS.

So what is MSR-BS? The 3GPP definition is: Base Station characterized by the ability of its receiver and transmitter to process two or more carriers in common active RF components simultaneously in a declared RF bandwidth, where at least one carrier is of a different RAT than the other carrier(s).

In very simple terms, a single Base Station will be able to simultaneously transmit different radio access technologies from a single unit. So a unit may be for example transmitting GSM, WCDMA 2100 and LTE 2600 simultaneously.

The number of technologies supported by a BTS will be an implementation choice. With technology maturing it wont be surprising to have upto 4-5 different technologies in a MSR-BS in the next five years.

The advantage the mobile operator will have will not only be monetary but there will be possibility of space saving. But as the old english proverb says, they will be "putting their eggs in a single basket". If one unit stops working then the coverage in the area goes down. There may not be an option to fallback on different technology.

The way this MSR-BS are implemented will be definitely based on Software Defined Radios (SDR). The advantage with SDR will be that in different parts there is a slight frequency variation for different technologies like GSM-850 is specific to USA whereas the rest of the world uses GSM-900. These small variations will easily be customisable with these MSR-BS and optimisations wont be too far off.

Different Band Categories have been defined for different scenarios. For example Band Category 1 involves deplyment where GSM wont be present. Only LTE and WCDMA is present there. Band Category 2 involves frequency bands where GSM, EDGE, WCDMA and LTE may be present. Band Category 3 is designed with TDD and TD-SCDMA in mind.

More information as and when available

LTE Band 13 mystery

Interesting observation from Hooman Razani in LTE university blog:

It was in a classroom, after I had opened Technical Specification 36.101 for handset radio transmission and reception that I suddenly noticed for the first time something odd about Band 13.

I said to the class without hesitation, Band 13 [UL = 777 MHz-787 MHz & DL = 746 MHz-756 MHz] has the uplink and downlink reversed! In contrast to all other FDD bands in the table (and many other bandplans elsewhere), in this band, the higher frequency is allocated to the mobile device and the lower frequency to the base station. The normal practice of FDD paired band allocation has to do with attenuation properties of the carrier frequency and the availability of the power. The mobile side has usually been allocated the “easier” lower-frequency portion of the band which experiences smaller amount of attenuation and therefore requires less power for closing the uplink. A reversal of this allocation strategy in band 13 is a sure sign that some other problem is lurking in the background, namely interference in the form of spurious emissions.

The 700MHz band is divided into four paired blocks, A, B, C and D (an unpaired E Block is also available). The A and B blocks are guard bands. The C blocks (LTE Band 13) in this upper 700MHz band were the main object of last year’s auction. 12MHz in the middle of this spectrum belongs to the Public Safety.

A – C – D – B – [Public Safety] – A – C – D - B

A look at the structure of the auctioned UHF band shows that upper portion of Band 13 is closer to the public safety band. In order to reduce the effects of interference to the public safety band and other bands due to intermodulation effects of this carrier, the higher more “difficult” frequency has been allocated to the lower powered handset. On the other hand we can expect a slight enhancement in indoor coverage for band 13, due to further lowering of the downlink frequency .The question to ask is if this maneuver is enough to suppress the spurious emissions caused by the handsets to nearby bands. Further details after initial deployment will shed light on the mysterious case of band 13.

Digital Dividend Spectrum

3G Americas have released a whitepaper titled, "3GPP Technology Approaches for Maximizing Fragmented Spectrum Allocations". There are quite a few interesting things that caught my attention. One of them is this section on Digital Dividend Spectrum about which I have mentioned before. Here is what the whitepaper says:

The Digital Dividend refers to the reallocation of significant amounts of spectrum as a result of the switchover from analog to digital TV, a phenomenon occurring across the globe. Historically, analog TV operates in the UHF band between 470-862 MHz.

The analog to digital switchover will free a substantial amount of spectrum for new services, including digital television and mobile broadband.

Mobile services will need at least 100 MHz of this spectrum for mobile broadband. The results of WRC 07 incorporate this vision, as well as the need to promote harmonization of these bands, as reflected in the following above.

As is planned, there is no one globally harmonized Digital Dividend spectrum band. Further, the APAC countries have the flexibility to adopt the Region 1 or Region 2 plans. The important task of promoting harmonization and aligning band plans as far as possible with WRC-07 agreement, in order to realize the benefits of harmonization for their citizens, now falls to policymakers to accomplish.

In fact, the EC recently launched a consultation on Digital Dividend spectrum. Noting that importance of taking prompt action “to prevent the emergence of fragmented national legacy situations” that would stymie the development of future equipment and services in the 800 MHz band, the consultation proposes that the EC undertake two urgent actions by autumn of 2009: (1) Member States that have not completed the digital switchover would be requested to confirm switch off of analogue TV under national law by 1 January 2012; and (2) the EC would draft a Commission decision, for regulatory opinion in the autumn of 2009 and formal adoption at the beginning of 2010, on technical harmonization measures for transitioning the 790-862 MHz band to non-broadcast uses.

Credit to 3G Americas for publishing these interesting white papers.

UK: Ofcom releases 3G coverage maps

Ofcom has just released (or as The Register puts it; found under the sofa) 3G coverage maps for UK. Its useful for people who dont live in big towns but planning to take out contracts on dongles/data services. They can now quickly check which operator to go for.

These 3G coverage maps by mobile operator were prepared in January 2009. They represent the area where we have assessed the mobile operators met a minimum coverage threshold set by Ofcom (see technical notes below). The shaded areas on the maps indicate areas where customers have the possibility of making and receiving a call outside over a 3G network (but with no guarantee of being able to do so). They do not indicate areas where customers are able to access higher data rate services.

All operators produce their own coverage indicators on their websites which are likely to provide more reliable guidance to network availability in any given area. The accuracy and detail of the maps are not to the same level as the mobile operators publish. These maps show UK-wide general coverage and are not suitable for zooming in to see specific locations i.e. a particular house or street. Also they are not suitable for assessing the quality or depth of coverage within the indicated areas (e.g. different operators may be able to offer better or worse data rate services or support a smaller or greater number of users).

You can see the PDF of the coverage maps here.

LTE activity gathering pace

Wireless internet access is going to be a better, richer experience than fixed link access Professor Michael Walker, group R&D director at Vodafone told Wireless 2.0 conference in Bristol, organised by Silicon South-West.

“People think wireless can’t compete with fixed link, but it can”, said Walker, pointing out that the 100Mbit/s of FTTH is the same as the theoretical maximum throughput of LTE.

“LTE capacity on 20MHz is an order of magnitude higher than HSPA,” said Walker. He said that, “in the first real field trials,” average downlink speeds of 15Mbit/s, with 4.5 spectral efficiency, were achieved. “Wimax takes three times more spectrum”, he said.

“We decided with LTE that we would make sure the technology works before we buy spectrum,” said Walker.

To that end, Vodafone has been working with China Mobile and Verizon to make sure LTE has compatible standards.

Walker regards talk of a killer app as silly for LTE as it was for 3G. “LTE is just about access,” he said.

He predicted the gradual relative demise of the person-to-person phone calls, referencing data that showed 11 times more wireless traffic is being generated by community chatting than by person-to-person calls.

Ericsson, one of the world’s leading suppliers of mobile phones and related network solutions, has warned that it could be 2012 before the first true next generation Mobile Broadband networks gain a good foothold in the UK. The deployment of Long Term Evolution (LTE) technology (aka - 4G), which could deliver download speeds of 150 to 1000Mbps, is being hampered by problems with releasing the needed 900Mhz spectrum.

Presently both O2 and Vodafone own some of the older 2G (900MHz) spectrum, which Ofcom is seeking to have redistributed to rival operators ( Orange , T-Mobile and Three (3) ). This could then be converted for use by 3G/4G voice and Mobile Broadband technologies, such as HSPA and LTE .

Nokia Siemens Networks (NSN), although providing WiMAX solutions for Taiwan operators, plans to launch commercialized LTE (Long Term Evolution) solutions in 2010, Mike Wang, NSN's general manager for Taiwan, Hong Kong and Macau.

NSN has already received LTE solution orders from Japan-based NTT DoCoMo, which is expected to start offering LTE-based services in 2010, Wang stated.

In addition, NSN has also landed orders for the installment of LTE trial networks for T-Mobile and Singapore-based service provider Mobile One, he added.

LTE-enabled chipsets are expected to hit the market starting in the second half of 2009, followed by LTE USB modems and network cards in 2010 and other LTE CPE such as handsets, PDAs, tablet PCs in 2012, Wang predicted.

Japan's DoCoMo is reported to be planning an investment of ¥ 343 billion (US$3.4 billion) for the investment, while KDDI and Softbank Mobile have budgeted ¥ 515 billion and ¥207.3 billion, respectively. According to the Nikkei newspaper, among the four, LTE services are forecast to attract around 36 million subscribers, with DoCoMo projecting 17.74 million LTE customers.

The total investment will top ¥1 trillion (US$10 billion)

DoCoMo, KDDI and Softbank Mobile expect roughly 30% of their existing customers to switch to 3.9G services. DoCoMo is also reported to be expecting to upgrade or deploy some 20,000 LTE enabled base stations by 2014. The network should cover 50% of the population and commercial services will start in 2010.

LTE provides downlink peak rates of at least 100Mbit/s, 50 Mbit/s in the uplink and RAN round-trip times of less than 10ms. Fujitsu recently announced that, in collaboration with NTT DoCoMo, they had successfully completed field testing for LTE, using 4x4 MIMO technology, which resulted in data transmission speeds in the range of 120 Mbps (using 10 MHz bandwidth) in Sapporo's urban environment.

NTT DOCOMO has selected Alcatel-Lucent' Ethernet transmission solution to provide the backhaul network of its Long Term Evolution (LTE) service. Financial terms were not disclosed.

Alcatel-Lucent said its packet optical gear will provide NTT DOCOMO with cost-effective and flexible Ethernet-based aggregation and transport from the base stations to the core network. Specifically, the Alcatel-Lucent solution, based on the 1850 Transport Service Switch (TSS) and its universal switching technology, will provide multipoint Ethernet connections between cell sites supported by strong operations, administration and maintenance capabilities, as well as carrier-class protection and network management.

China Mobile Ltd., recognizing that future growth of its 3G services is not yet certain, is looking to secure its future with a nearly parallel development of Long Term Evolution (LTE) capabilities, according to its annual report filed with the Securities and Exchange Commission (SEC) recently.

High on the list of limitations TD-SCDMA imposes are the availability, functionality, and price of terminals, and the support for international roaming. The latter particularly impacts China Mobile's high-end and business customers.

China Mobile remains committed to TD-SCDMA, but at the same time it's going all out to develop converged time division duplex (TDD) and frequency division duplex (FDD) products for LTE, the proto-4G wireless technology set to be deployed by mobile operators worldwide.
Developing a simultaneous LTE strategy will enable China Mobile to limit the time it is dependent on TD-SCDMA, and also help it counter the constraints it's currently experiencing as a result of the TD-SCDMA sector's limited ecosystem.

The Chinese vendors have labored too long under the market perception that they deliver low cost equipment, but can't do cutting edge. To fight this image and enhance their position among tier one carriers, Huawei and ZTE are throwing everything - including their huge credit lines for vendor financing - at gaining early LTE trials, which puts them in the spotlight even if commercial roll-outs are often two years or more away. The latest points go to ZTE, which has won field trial contracts with Telefónica in Spain and CSL in Hong Kong.

This comes shortly after Huawei highlighted its LTE deal with Netcom of Norway, and the Chinese duo are, early market estimates indicate, coming close to Ericsson in terms of their penetration of stage one LTE trials - and ahead of Alcatel-Lucent and Nokia Siemens.

Telefónica confirmed reports by Light Reading that ZTE's trial will take place in the third quarter. While success at the Spanish firm may not translate into meaningful revenue any time soon (and Ericsson has already conducted LTE trials with Telefónica too), it will still be a major boost for the vendor. It has not been as successful in getting tier one western carrier deals as its compatriot - although its overall wireless equipment market share grew at a similar rate to that of Huawei in the past 12 months, its base is far more concentrated on the lower margin developing markets.

By contrast with Telefónica, CSL is already a major customer for ZTE and one of the first commercial users of its software defined base station platform, which it is currently rolling out in an IP-based HSPA+ network across Hong Kong. In a recent interview with Telecoms.com, Tarek Robbiati, CEO of the Telstra subsidiary, said: "Further consolidation will come in the next three to five years. In the end there will be only three [infrastructure vendors] left, and two of them will be Chinese. The European vendors are just too slow."

With LTE firmly at the top of the hype curve in 2009, WiMAX is somewhat overshadowed in the headlines, but did make a strong impression at last week's CommunicAsia show in Singapore, highlighting how the technology has gained a higher profile in Asia than in Europe - because of the importance of the Taiwanese ODMs and vendors like Samsung, as well as a large number of deployments in countries like Japan, Vietnam, Korea and Taiwan.

800 or 2600MHz: Frequency Impact on Fixed Deployments

From a presentation by Richard Keith, Director of Global Strategy, Broadband Access Solutions Home & Networks Mobility, Motorola at LTE World Summit

Drivers for TD-LTE

From a presentation by Matthias Reiss –Head of LTE Radio, NSN

Cell Radius on Different LTE Frequencies

I mentioned earlier about the cell radius being 300 metres at 2.6GHz. Here is a diagram showing that.

LTE World Summit - Day 1 roundup

Before we start, I should mention that there have been no discussions showing LTE is better than WiMAX, etc. It is now assumed that WiMAX is no longer a threat and a competitor. There is also missing a discussion on Femtocells. I suppose that some Femto related discussions are planned for day 3.

I am going to cover the discussions in detail in the coming days (months) but here are some interesting tidbits.

It started with a presentation from Marc Fossier, France Telecom. An interesting fact about France he mentioned is that 40% of voice traffic is VoIP. They are present in 30 countries (generally by the name Orange) and they have 128 million customers. LTE rollout is planned for post 2011. They will be deploying WiMAX in some markets like Romania and some African countries. The main reason being that there is no 3G network there. Another key point of his presentation was that Mobile TV should be independent of the technology being considered and TDD works fine for that. Also SON (Self Organising Networks) will be very important feature that is needed when LTE is rolled out.

Klaus-Jurgen Krath from T-Mobile Germany gave T-Mobile statistics of 148 million customers and revenue of 9.2 Billion Euros. One of the things he mentioned is that at 2.6GHz, the cell radius is 0.3Km but if we move to 800MHz then the cell radius is around 3Km. Personally I am not sure if the frequency is something we should worry too much about. WiBro in Korea is I think using 2.3GHz and is working fine. Another thing mentioned is that the Backhaul is generally quite bandwidth limited so that needs to be improved. The operators should use Microwave or Fiber to overcome this backhaul problem. LTE Migration will also take time and initially it would be the third layer on top of GSM and 3G/HSPA but by 2020 it should be dominant technology. Right now Voice and SMS problem has to be solved asap. Personally I think SMS is not much of a problem and SMS can work on CS as well as PS. Operators can use SMS over PS for the moment. Femtocell according to Klaus is very important for LTE but Femtos will be successful only if they are Plug and Play for both operator and the consumer. His final point was that we should not dissapoint the customer by showing the peak rates, rather we should focus on average rate.

Ulf Ewaldsson from Ericsson mentioned that 80% of broadband subscribers by 2014 will be mobile. Another thing he said is that for LTE, it is possible to deliver 1GB for < 1 euro. (Capex only). Ericsson vision is that by 2020 there will be 50 Billion mobile broadband connections. Ericsson will be ready with LTE soon and by the year end peak rate of 42Mbps will be possible.

Alex Sinclair from GSMA talked on expanding the LTE ecosystem. There were lots of interesting facts and figures that I will have to mention seperately. There was also an intersting mention of the GRX (GPRS Roaming Exchange) that will need a completely new discussion.

In the Panel Discussion, Marc mentioned that Orange is not considering Femtocells for Residential market because of many unsolved technical problems like Price, Security, Spectrum Management, etc. On the business side they may have Femto, Pico or Micro depending on the need. Ed Candy from '3' mentioned that operators should focus on Macro rather than Femto because lots of spare capacity available. He mentioned that there is a business problem because somebody has to subsidise Femto, also too many Femtos can cause RF blackspots and operational dynamics are yet to be worked out.

Elio Florina from Telecom Italia (TiM) talked about LTE and HSPA deployment in Brasil. The main point being that initially they want to have complete 3G/HSPA coverage and then in future think about LTE.

Alan Hadden from GSA mentioned about GSA activities. I was surprised to hear that only 55,000 people view their website annually. This is because a simple blog like mine regularly gets between 20,000 and 25,000 views per month. Again there were lots of stats that I will write about later. Interesting point from his discussion was about the Digital Dividend band that can be used by LTE to increase the coverage.

Dave Marutiak, Microsoft spoke about Bearer Aware Applications. Dean has blogged in detail about this.

Ed Candy from '3' in his presentation highlighted an important point about the user experience. He said that for any service, user experience should be memorable so even after 2-3 years if the user has to use the same application than he should remember how it works. I think that there are hardly any services like this except those that are intutive to use. In Dec. 08, '3' saw people using 34million skype minutes and 68 million Facebook page views. It should be remembered that Facebook phone was launched in Dec itself. His message was that '3' will not be moving to LTE anytime soon because the current network is delivering whatever is required.

Roberto Di Pietro from Qualcomm spoke about the challenges faced by the chipset manufacturers. The main challenges is due to the fact that there are millions of combinations of different bands that the UE may need to support which is not practical and possible. They are focussing mainly on 2.6GHz at the moment and when the Digital Dividend band is available then they will implement chipset in that band as well.

Jorgen Lantto from ST-Ericsson and Sami Jokinen from Nokia gave their views on chipsets and handsets implementation respectively. According to Nokia, LTE handsets will be available by 2010. Initial deployments will be in FDD but later in TDD-FDD both.

Liesbet Van der Perre from IMEC discussed about the SDR implementation for LTE devices. Her presentation was more of updates on the previous IMEC info I have covered here.

Finally, I managed to squeeze myself into an Agilent workshop called 'LTE at the movies'. There were two parts to that 'MIMO MIA' and 'Honey who shrunk my mega bits'. The workshop lated 2.5 hours and I got a free Agilent LTE book. Lots of technical details that I will post seperately.

Please feel free to comment or correct me if I made any mistakes. You can also read Dean Bubley's take on the first day here.

I shouldnt finish without thanking the organisers who have done a good job in organising the event and for such wonderful hospitality (and food :).

LTE Status: Jan 09 updates

According to this news, Verizon Wireless says that the Next-Gen 'LTE' Wireless Network Still On Track For 2010 Launch:

Despite a tanking economy and a potential delay getting access to the spectrum Verizon bought in last year's FCC auction, the telco says it's still on track to roll out its next-generation wireless network next year.

On the company's earnings call today, Verizon COO Denny Strigl said he's still hoping LTE will be commercially available in the first half of 2010, with in-house testing to begin later this year.

Meanwhile in Sweden, Swedish telecommunications operator TeliaSonera has signed contracts towards the launch of an 4G high-speed wireless network in 2010, saying it aims to be one of the first with an LTE network up and running.

TeliaSonera said on Thursday it had signed up Ericsson to construct its initial Stockholm LTE (the Long Term Evolution of 3G) network, and Huawei for the initial Oslo network. Both equipment makers said it was their first commercial LTE contract.

The telco said it is still evaluating suppliers for further LTE networks across Sweden and its other markets.

India gets ready for 3G

So here comes 3G in India. It’s been long coming as the data needs were increasing rapidly in almost all the Indian states. With the existing cellular infrastructure not capable of holding huge traffic particular for data, arrival of 3G was imminent.

The Indian Department of Telecoms (DoT) has published its official timetable for the award of its 3G licences across the country as well as a breakdown of how the relevant spectrum will be allocated across the telecoms circles.

As expected, the state-owned operators BSNL and MTNL each have been reserved one block of 2x5MHz in each circle, with the exception of Rajasthan (State in North West India) which will have no 3G spectrum at all. The number of blocks of spectrum in the private auction differs depending on the circle (see the spectrum table, below).

The auction for the 15-year licences is planned for Jan. 15, 2009. In the majority of 3G service areas there is 25 MHz of paired frequency bandwidth available which relates to four blocks of 2x5 MHz spectrum available for auction in addition to the block reserved for the state-owned operators, Bharat Sanchar Nigam (BSNL) and Mahanagar Telephone Nigam (MTNL). Spectrum is rather limited in many other areas, including the major metro circle of Delhi where only two 2x5MHz blocks will be available to private operators.

All of the 3G spectrum will be in the 2.1 GHz band and in the 2.3 GHz and 2.5 GHz frequency bands, a separate auction for Broadband Wireless Access (WiMAX). In both these auctions, which will take place two days after the 3G auction, bidders are restricted to just one block of spectrum per service area.

The table below shows the proposed spectrum layout.

Service Area (Indian Cities or States)	Paired frequency bandwidth to be allotted	Paired frequency bandwidth to be allotted
Delhi	160	15
Mumbai	160	25
Kolkata	80	25
Maharashtra	160	25
Gujrat	160	15
Andhra Pradesh	160	25
Karnataka	160	25
Tamil Nadu	80	25
Kerela	80	25
Punjab	80	25
Haryana	80	25
Uttar Pradesh(e)	80	25
Uttar Pradesh (w)	80	10
Rajasthan	0	20
Madhya Pradesh	80	25
Bengal	80	25
Himachal Prades	30	25
Bihar	30	25
Orrisa	30	25
Assam	30	25
North East	30	5
Jammu And Kashmir	30	25

Satellite based Mobile Internet of the future

Background: The current US military satellite communications network represents decades-old technology. To meet the heightened demands of national security in the coming years, newer and more powerful systems are being developed.

Advances in information technology are fundamentally changing the way military conflicts are resolved. The ability to transmit detailed information quickly and reliably to and from all parts of the globe will help streamline military command and control and ensure information superiority, enabling faster deployment of highly mobile forces capable of adapting quickly to changing conditions in the field. Satellite communications play a pivotal role in providing the interoperable, robust, "network-centric" communications needed for future operations.

Military satellite communications (or milsatcom) systems are typically categorized as wideband, protected, or narrowband. Wideband systems emphasize high capacity. Protected systems stress antijam features, covertness, and nuclear survivability. Narrowband systems emphasize support to users who need voice or low-data-rate communications and who also may be mobile or otherwise disadvantaged (because of limited terminal capability, antenna size, environment, etc.).

For wideband communication needs, the Wideband Gapfiller Satellite program and the Advanced Wideband System will augment and eventually replace the Defense Satellite Communications System (DSCS). These satellites will transmit several gigabits of data per second—up to ten times the data flow of the satellites being replaced. Protected communications will be addressed by a global extremely high frequency (EHF) system, composed of the Advanced Extremely High Frequency System and Advanced Polar System. These systems are expected to provide about ten times the capacity of current protected satellites (the Milstar satellites). Narrowband needs are supported by the UFO (Ultrahigh-frequency Follow-On) constellation, which will be replaced by a component of the Advanced Narrowband System

Lockheed Martin Space Systems, Hughes Space and Communications and TRW have formed a National Team to build the Department of Defense's (DOD) next generation of highly secure communication satellites known as the Advanced Extremely High Frequency (AEHF) system.

The Advanced EHF programme provides the follow-on capability to the Milstar satellite programme. It provides the basis for the next generation military communications satellite system, for survivable, jam-resistant, worldwide, secure, communications for the strategic and tactical warfighter. The system replenishes the Milstar constellation in the EHF band.

Each of these Advanced EHF satellites employs more than 50 communications channels via multiple, simultaneous downlinks. Launch of the first AEHF satellite is planned for April 2008 with the second AEHF satellite scheduled for launch in April 2009.

The fully operational Advanced EHF constellation will consist of four crosslinked satellites, providing coverage of the Earth from 65° north latitude to 65° south. These satellites will provide more data throughput capability and coverage flexibility to regional and global military operations than ever before. The fifth satellite built could be used as a spare or launched to provide additional capability to the envisioned constellation.

Current Status: After being plagued with project overruns and a scaling back of the final system, the US military's next generation satellite communications network is another step closer to reality, with completion of the payload module for the third and final Advanced Extremely High Frequency (EHF) satellite.

Although the EHF band is a relatively lightly used part of the electromagnetic spectrum (30-300 GHz), it is for good reason. Atmospheric attenuation is the biggest problem faced in this band, especially around 60 GHz, however the frequencies are viable for short distance terrestrial based communication links, such as microwave Internet and telecommunication links (which already operate in this band). Millimetre wave radar, probably best known as the radar that can see through your clothes but not your skin, also operates in this band.

Designed to avoid problematic frequencies that are more susceptible to attenuation, but accepting increased overall atmospheric attenuation, are an increasing number of military and civil satellite systems that are using this band for uplink and downlink, as well as inter-satellite communication. Inter-satellite communication is really where EHF equipment shines (no atmosphere, small antennas, high data rates).

Civilian systems are currently around the Ku band (Intelsat), providing data rates of up to 2-4 Mbps (14 GHz uplink, 12 GHz downlink) however these rates have still to trickle into everyday user's hands for remote and mobile Internet access. It is more common that an aggregator will access this link/rate and use that to then portion out local Internet access. Systems such as this are in use for remote Australian territories like Cocos and Christmas Islands, and formed the backbone of Boeing's stillborn Connexion in-flight Internet access. High ongoing access costs (basically a share of the overall cost of the satellite) and limited access slots help keep the technology away from everyday use at this time. Militaries and governments around the globe also lease access on these circuits when they need the added capability, with Intelsat and Inmarsat systems being used in the first Gulf War.

Advanced EHF is designed to provide 24 hour coverage from 65 North, to 65 South across the K and Ka sub bands, and when combined with the prototyped Extended Data Rate (XDR) terminals and systems, will offer up to 8.2 Mbps data rates for around 4,000 terminals in concurrent use per satellite footprint (whether that scales to 12,000 systems in concurrent use globally isn't clear from source material).

Within the tri-satellite constellation, inter-satellite EHF links will allow terminals on opposite sides of the globe to communicate in near real-time without the use of a terrestrial link. Combined with smaller, directional antennas and the various options for anti-jamming technology, it represents a significant military capability for the US.

Already plans are being drawn up for the Transformational Satellite Communications System (T-Sat) which will replace Advanced EHF starting sometime in 2013, however it is already facing funding troubles. This could be problematic, with Advanced EHF still struggling to reach capability and the final launch not scheduled until April 2010. Dropping the fourth satellite of the Advanced EHF constellation has been planned to give the USAF time to implement T-Sat more rapidly.

If GPS and remote imaging (think Google Earth) have proven anything, it is that technology initially developed for military purposes, and extremely expensive for initial civil use, will eventually reach the point where it forms part of our daily lives without us ever being conscious of the massive investment to get to that point.

LTE Roll out updates from the 4th LTE World Summit in London

As I mentioned in an earlier post, I got the opportunity to attend and meet with the industry experts in the 4th LTE world summit. There were some very interesting discussions and debates and some announcements about the rollout of LTE. Here is a quick summary of the announcements and news. I am sure to have missed some and will expand on some of the topics in later posts.

Karri Mikkonen, Director, Corporate Strategy, TeliaSonera in his presentation said that TeliaSonera to be an early LTE adopter with rollout planned Mid 2010. They have already bought licenses in Sweden, Norway, Finland, Denmark, Baltics. According to them LTE enables very convenient mobile data usage scenarios, and is one tool to win battle in mobile data, among others.

Bill Huang, General Manager, China Mobile Research Institute gave a very interesting perspective on mobiles in China which I will expand in a later post. According to his presentation, China Mobile will be deploying TDD based LTE (TD-LTE) option probably around 400Mhz spectrum. Trails will start by Mid 2009 and pre-commercial launch will happen around end of 2009. By Q2, 2010 there will be large scale commercial trials involving around 15,000 base stations.

Remi Thomas, Director of NGMN project and head of Network Strategy, France Telecom, France said FT plans to introduce HSPA+ and opt for 'wait and see' approach for LTE. The earliest they want to even think about LTE is after 2010 and if practical the rollout may occur in 2012. Even with HSPA+ they will opt for all the software changes only and not go for any hardware changes. So we wont see MIMO anytime soon with HSPA+ according to them. They also have plans to rollout LTE Femtocells when available to check the technology and iron out the problems with LTE technology.

Erik Ekudden, Vice President, Group Function Technology, Ericsson, Sweden in his presentation said that E\\\ will be commercially releasing equipment in 2009. Terminal HW including support FDD and TDD modeof LTE. For FDD, initial support will be for IMT core band of 2.1GHz and also IMT extension band of 2.6GHz and US 700MHz spectrum. For TDD the initial support will be for IMT extension center gap in 2.6GHz spectrum and 2.3 TDD band in 2.3GHz band.

In a question answer session, Dr Howard Benn, Director of Cellular Standards, Motorola Mobile Devices mentioned that Motorola already has a working LTE UE but not in Form factor (probably development board). He did not expand on the details.

Nick (Norikazu) Yamasaki, Manager Standards Strategy section, Emerging Technologies and Spectrum Division, KDDI Corp. in his presentation said that KDDI is a CDMA2000 operator but since with UMB nearly dead (my words) they have decided to eveolve to LTE. LTE deployment willbe started around 2012 but will co-exist with the CDMA 2000 network which they will support for another 10 years. Right now they have EV-DO Rev.A but they may also opt for Rev.B

One of the big problems that was discussed many times in the conference was that Release 8 LTE standards have no solution for the normal CS voice call. There are some hacks around it but voice part will only be solved in time for Release 9. This could delay decisions by some operators to roll out LTE networks untill after Release 9. I will write a post on this later.

Femtocells and the stealing of Spectrum

When Femtocells are finally rolled out, it would be possible for anyone to create their own little mobile cell anywhere to enhance their coverage. At least that is what the Femtocells are supposed to help with. This would also mean that the spectrum would be open to abuse by someone who wants to abuse it.

Let's take a scenario in which someone buys a Femtocell from an operator in UK. The Femtocells will be operator specific since they will contain lots of parameters and addresses that would be terminating in the operator network. Then that person can take the Femtocell away to another country (say India) and connect the Femtocell to an Ethernet port in India. The IP packets would be routed via IP to the operator and the user is now connected via Femtocell to the UK operator even though he in in India. He would get the same treatment as in case he was in UK.

Let me point out that this would be illegal because the Spectrum in India would belong to an operator in India or this spectrum may be used for something completely different.

The operators and the device manufacturers are aware of this potential abuse. As a result they are going to use a two step approach. The first is that they would allow Femtocell to register from a registered telephone line via an IP address. They may have access to ISP data or would be aware of the range of IP address being used by the ISP. The Femtocell user will hence have to register their Telephone line and ISP with the network operator and if they change them then this would need to be informed to the operator. The second is that they would check the location of the device via GPS. This can have two problems. The first is the cost of the Femtocell will increase and the second is that unless the Femtocell is near a window or an open area, there would be no GPS signals received and the GPS approach may not work. One of the obvious use of Femtocell in London city for example is in the basements where there is absolutely no coverage due to their location.

Note that from the above you can see that even if the Femtocells are advertised as PnP or Zero Touch, etc., there would still be some overhead that will always be required.

Even if we assume that both the above approaches are being used, it may stop mass market fraud but may not be able to deter individuals who are smart enough to work around them. For example the user in India (example in the start) may use VPN to tunnel the IP packets to their home or registered address in UK and from there the packets will go to the operator network. Similarly it is not too difficult to fool the GPS receiver into believing its location.

The operators are aware and working on something better then the above strategy. I have not come across any papers yet suggesting work around these problems.

This also highlights an important problem regarding emergency calls. Should the emergency calls go via Femtocell or should they be re-directed to Macro cell. Again a clever algorithm would be needed for this. There could be a configurable parameter in the Femtocell which can check during the startup if Macrocell is present or not. If Macrocell is present then emergency calls should be re-directed and if not present then the user should be able to initiate it via Femtocell.

There are probably many more problems that would be highlighted once Femtocells are rolled out.

Momentum Building for UMTS 900MHz

According to a recent paper published by GSA, momentum is building for introducing UMTS 900 i.e. WCDMA-HSPA systems in 900 MHz band, used today by GSM/EDGE networks, to help operators to extend voice, data and mobile broadband services coverage by leveraging the advantages of lower frequencies. UMTS 900 is on the roadmap of several manufacturers. Three
commercial UMTS 900 systems have launched, and 20 user devices have been announced by 6 manufacturers.

I have blogged on 900Mhz band in past. Technical specifications for WCDMA-HSDPA in the 900 MHz band (UMTS 900) were completed by 3GPP in December 2005. The 900 MHz band, denoted as Band Class VIII, is defined as paired bands in the range 880 to 915 MHz (uplink), and 925 to 960 MHz (downlink).

A Manx Telecom trial confirmed 30% improved inbuilding penetration compared to 2100 MHz, and 40% in deep indoor penetration. With HSDPA, throughput could increase by 10%, raising overall network capacity 5%. A key finding was the ability to hand over calls between base stations operating at different frequencies. The trial confirmed a GSM 900 operator could re-use sites for UMTS without having to redesign and re-deploy the network, thus significantly reducing operational costs.

Dell to power laptops with HSPA

Ericsson has a press release saying that Dell will use its high-speed HSPA mobile broadband technology in next-generation laptops due in the second quarter of 2008. The modules will be built in to the Dell laptops, according to a press release by both companies. By June, Intel is expected to roll out its next-generation "Montevina" mobile chipset, which will be used as the foundation for the next generation of Centrino notebooks.

Although Montevina was expected to usher in the next-generation WiMAX technology, the apparent delays underlying Clearwire's WiMAX rollout may have pushed Dell to seek an alternative broadband choice.

According to a Dell spokeswoman, the choice to include Ericsson's HSPA technology was as much about compatibility as throughput. If a customer takes a 3G-enabled laptop with him or her to Europe, it might work, "but it's not a seamless transition," Dell's Anne Camden said. The HSPA technology is more uniform throughout the globe, she said.

But it's also true that Dell wanted a broadband solution now. "Mobile broadband delievers a broadband experience today, and that is what we need," Camden said. "We want to deliver a great broadband experience. We're certainly looking at WiMAX support in future products."
Dell is the second major PC vendor to sign on to use Ericsson's HSPA technology, after Lenovo.

According to Ericsson, both Dell's business customers and consumers will use the new modules. Interestingly, Ericsson built in a GPS component into the HSPA modules, meaning that location services will be also be built in.

Market projections indicate that in 2011, approximately 200 million notebooks will ship annually and Ericsson anticipates that 50 percent of those notebooks will feature a built-in HSPA mobile broadband module. Users will increasingly have the option to take their broadband connections with them, delivering on the promise of full service broadband, which is anytime, anywhere access from the screen or device of choice.

Meanwhile:

Winners of Sweden's 2.6GHz spectrum auction can now look to rapid deployment of advanced mobile networks, with Ericsson poised to deliver end-to-end HSPA and LTE technology. The auction is the first held in the world to license according to the harmonized band arrangement decision by the European Conference of Postal and Telecommunications Administrations (CEPT).

As a front runner in allocating the 2.6GHz frequency band, the regulator Swedish Post and Telecom Agency has adopted a harmonized spectrum allocation as defined by CEPT. The allocation will facilitate economies of scale for operators and secure the availability of standardized terminals, allowing roaming between countries for users. Auctions of the 2.6GHz band in Austria, Netherlands, Italy and the UK are scheduled for 2008.
LTE and HSPA, the preferred technologies for the 2.6GHz band, enable a superior, mass-market user experience, enhancing demanding applications such as mobile video, blogging, advanced games, rich multimedia telephony and professional services.

Ericsson's solutions help operators leverage their network investments by providing optimal voice communication and mobile broadband services. Ericsson employs scalable architecture and allows seamless network expansion, providing an efficient migration path to broadband, regardless of the legacy technology in place.

Ericsson's offerings for the 2.6GHz band are based on its multi-standard RBS 3000 and RBS 6000 series. These energy efficient base stations support WCDMA/HSPA/LTE and GSM/EDGE/WCDMA/HSPA/LTE respectively. Ericsson's RBS suite offers the smallest base stations on the market and facilitates low-cost migration and easy network integration. HSPA is already commercially deployed in more than 185 networks in 80 countries, with more than 600 devices launched.