Showing posts with label South Korea. Show all posts
Showing posts with label South Korea. Show all posts

Sunday, 17 July 2016

Two VoLTE Deployment Case Studies

Back in 2011, I was right in predicting that we will not see VoLTE as early as everyone had predicted. Looking through my twitter archive, I would say I was about right.

The big issue with VoLTE has always been the complexity. In a post last year I provided a quote from China Mobile group vice-president Mr.Liu Aili, "VoLTE network deployment is the one of the most difficult project ever, the implementation complexity and workload is unparalleled in history".

From a recent information published by IHS, there will only be 310 million subscribers by end of 2016 and 2020 is when 1 billion subscribers can make use of VoLTE. I think the number will probably be much higher as we will have VoLTE by stealth.

Below are couple of case studies, one from SK Telecom, presented by Chloe(Go-Eun) Lee and other from Henry Wong, CTO Mobile Engineering, Hong Kong Telecom (HKT). Hope you find them informative and useful.

Friday, 12 December 2014

5G Spectrum and challenges

I was looking at the proposed spectrum for 5G last week. Anyone who follows me on Twitter would have seen the tweets from last weekend already. I think there is more to discuss then just tweet them so here it is.

Metis has the most comprehensive list of all the bands identified from 6GHz, all the way to 86GHz. I am not exactly sure but the slide also identifies who/what is currently occupying these bands in different parts of the world.

The FCC in the USA has opened a Notice of Inquiry (NoI) for using the bands above 24GHz for mobile broadband. The frequency bands above have a potential as there is a big contiguous chunk of spectrum available in each band.

Finally, the slides from ETRI, South Korea show that they want to have 500MHz bandwidth in frequencies above 6GHz.

As I am sure we all know, the higher the frequency, the lower the cell size and penetration indoors. The advantage on the other hand is smaller cell sizes, leading to higher data rates. The antennas also become smaller at higher frequencies thereby making it easier to have higher order MIMO (and massive MIMO). The only way to reliably be able to do mobile broadband is to use beamforming. The tricky part with that is the beam has to track the mobile user which may be an issue at higher speeds.

The ITU working party 5D, recently released a draft report on 'The technical feasibility of IMT in the bands above 6 GHz'. The document is embedded below.

xoxoxo Added Later (13/12/2014) xoxoxo
Here are some links on the related topic:

xoxoxo Added Later (18/12/2014) xoxoxo
Moray Rumney from Keysight (Agilent) gave a presentation on this topic in the Cambridge Wireless Mobile Broadband SIG event yesterday, his presentation is embedded below.

Tuesday, 22 October 2013

Korea Telecom ‘Route Decision System’ for midnight buses

Interesting presentation from Korea Telecom in LTE Asia 2013 about how they use Big Data to decide the night bus routes. Here are two pics which are self explanatory

We will soon start seeing operators using the data being collected from users and this can also be a nice little earner for them.

Thursday, 3 October 2013

Case study of SKT deployment using the C-RAN architecture

Recently I came across this whitepaper by iGR, where they have done a case study on the SKT deployment using C-RAN. The main point can be summarised from the whitepaper as follows:

This approach created several advantages for SK Telecom – or for any operator that might implement a similar solution – including the:

  • Maximum re-use of existing fiber infrastructure to reduce the need for new fiber runs which ultimately reduced the time to market and capital costs.
  • Ability to quickly add more ONTs to the fiber rings so as to support additional RAN capacity when needed.
  • Support of multiple small cells on a single fiber strand. This is critical to reducing costs and having the flexibility to scale.
  • Reduction of operating expenses.
  • Increased reliability due to the use of fiber rings with redundancy.
  • Support for both licensed and unlicensed RAN solutions, including WiFi. Thus, the fronthaul architecture could support LTE and WiFi RANs on the same system.
As a result of its implementation, SK Telecom rolled out a new LTE network in 12 months rather than 24 and reduced operating expenses in the first year by approximately five percent. By 2014, SK Telecom expects an additional 50 percent OpEx savings due to the new architecture.

Anyway, the paper is embedded below for your perusal and is available to download from the iGR website here.

Sunday, 25 August 2013

Centralized SON

I was going through the presentation by SKT that I blogged about here and came across this slide above. SKT is clearly promoting the benefits of their C-SON (centralized SON) here.

The old 4G Americas whitepaper (here) explained the differences between the three approaches; Centralized (C-SON), Distributed (D-SON) and Hybrid (H-SON). An extract from that paper here:

In a centralized architecture, SON algorithms for one or more use cases reside on the Element Management System (EMS) or a separate SON server that manages the eNB's. The output of the SON algorithms namely, the values of specific parameters, are then passed to the eNB's either on a periodic basis or when needed. A centralized approach allows for more manageable implementation of the SON algorithms. It allows for use case interactions between SON algorithms to be considered before modifying SON parameters. However, active updates to the use case parameters are delayed since KPIs and UE measurement information must be forwarded to a centralized location for processing. Filtered and condensed information are passed from the eNB to the centralized SON server to preserve the scalability of the solution in terms of the volume of information transported. Less information is available at the SON server compared to that which would be available at the eNB. Higher latency due to the time taken to collect UE information restricts the applicability of a purely centralized SON architecture to those algorithms that require slower response time. Furthermore, since the centralized SON server presents a single point of failure, an outage in the centralized server or backhaul could result in stale and outdated parameters being used at the eNB due to likely less frequent updates of SON parameters at the eNB compared to that is possible in a distributed solution.

In a distributed approach, SON algorithms reside within the eNB’s, thus allowing autonomous decision making at the eNB's based on UE measurements received on the eNB's and additional information from other eNB's being received via the X2 interface. A distributed architecture allows for ease of deployment in multi-vendor networks and optimization on faster time scales. Optimization could be done for different times of the day. However, due to the inability to ensure standard and identical implementation of algorithms in a multi-vendor network, careful monitoring of KPIs is needed to minimize potential network instabilities and ensure overall optimal operation.

In practical deployments, these architecture alternatives are not mutually exclusive and could coexist for different purposes, as is realized in a hybrid SON approach. In a hybrid approach, part of a given SON optimization algorithm are executed in the NMS while another part of the same SON algorithm could be executed in the eNB. For example, the values of the initial parameters could be done in a centralized server and updates and refinement to those parameters in response to the actual UE measurements could be done on the eNB's. Each implementation has its own advantages and disadvantages. The choice of centralized, distributed or hybrid architecture needs to be decided on a use-case by use case basis depending on the information availability, processing and speed of response requirements of that use case. In the case of a hybrid or centralized solution, a practical deployment would require specific partnership between the infrastructure vendor, the operator and possibly a third party tool company. Operators can choose the most suitable approach depending upon the current infrastructure deployment.

Finally, Celcite CMO recently recently gave an interview on this topic on Thinksmallcell here. An extract below:

SON software tunes and optimises mobile network performance by setting configuration parameters in cellsites (both large and small), such as the maximum RF power levels, neighbour lists and frequency allocation. In some cases, even the antenna tilt angles are updated to adjust the coverage of individual cells.

Centralised SON (C-SON) software co-ordinates all the small and macrocells, across multiple radio technologies and multiple vendors in a geographic region - autonomously updating parameters via closed loop algorithms. Changes can be as frequent as every 15 minutes– this is partly limited by the bottlenecks of how rapidly measurement data is reported by RAN equipment and also the capacity to handle large numbers of parameter changes. Different RAN vendor equipment is driven from the same SON software. A variety of data feeds from the live network are continuously monitored and used to update system performance, allowing it to adapt automatically to changes throughout the day including outages, population movement and changes in services being used.

Distributed SON (D-SON) software is autonomous within each small cell (or macrocell) determining for itself the RF power level, neighbour lists etc. based on signals it can detect itself (RF sniffing) or by communicating directly with other small cells.

LTE has many SON features already designed in from the outset, with the X.2 interface specifically used to co-ordinate between small and macrocell layers whereas 3G lacks SON standards and requires proprietary solutions.
C-SON software is available from a relatively small number of mostly independent software vendors, while D-SON is built-in to each small cell or macro node provided by the vendor. Both C-SON and D-SON will be needed if network operators are to roll out substantial numbers of small cells quickly and efficiently, especially when more tightly integrated into the network with residential femtocells.

Celcite is one of the handful of C-SON software solution vendors. Founded some 10 years ago, it has grown organically by 35% annually to 450 employees. With major customers in both North and South America, the company is expanding from 3G UMTS SON technology and is actively running trials with LTE C-SON.

Quite a few companies are claiming to be in the SON space, but Celcite would argue that there are perhaps only half a dozen with the capabilities for credible C-SON solutions today. Few companies can point to live deployments. As with most software systems, 90% of the issues arise when something goes wrong and it's those "corner cases" which take time to learn about and deal with from real-world deployment experience.

A major concern is termed "Runaway SON" where the system goes out of control and causes tremendous negative impact on the network. It's important to understand when to trigger SON command and when not to. This ability to orchestrate and issue configuration commands is critical for a safe, secure and effective solution.

Let me know your opinions via comments below.

Friday, 1 July 2011

Summary of 'The Future of Wireless International Conference' #fwic

Here is a summary of the Future of Wireless International conference held in Cambridge on the 27th and 28th of June 2011. The summary is a compilation of my notes with the tweets sent using the #FWIC tag.


Roberto di Pietro, VP Product Marketing and Business Development, Qualcomm CDMA Technologies
• 26 million 3G connections being added every month
• 226% growth is seen in smartphones from 2010 - 2014
• Mobile as a single platform for developers.
• Devices smart enough to know which network to connect to
• Qualcomm arrived on the scene 6 months after everyone but they are the only ones with 4G, 3G and 2G multi-mode chips
• In 2012 they would be releasing the new System Architecture with Single / Dual / Quad cores upto 2.5GHz (Snapdragon next gen)
• Question: Will smartphones die in the future when people move to tablets for everything except for voice/sms and they get simpler phones for that
• Answer: Smartphones will co-exist as companion devices with the tablets and will continue growing for a while.
• In other discussions: QoS will be a big differentiator and offloading would certainly be needed. Femtocells are going to form part of any strategy.
• Network signaling load and need for developers to improve apps design noted in qualcomm keynote here in cambridge

Mr. BongGoon Kwak, Senior Vice President, The head of Mobile Business Fast Incubation Business Department Mobile Business Group in Korea Telecom.
• KT adding 0.5 million users every month.
• Mobile data predicted to grow 26 fold by 2015 (6.2 exabytes/month)
• E = MC^2. Where E = evolution, M = mobile and C = connectivity
• mobile banking users in Korea increased 100% to 18 million due to smartphones
• smartphone ARPU up 32% on feature phone
• KakaoTalk ( users have increased which has in turn reduced the SMS ARPU
• NaaS (Network as a Service) is a new trend

Mr. Edward Zhou, CMO of Western European Region, Huawei Technologies Co. Ltd.
• states they have 5300 people in Europe but only 65% are from local market
• No. 2 telecom solution provider with revenues of $28 billion
• has 110,000+ employees with 150 nationalities worldwide, more than half work in R&D
• By 2020 there will be 5.5B MBB (Mobile Broadband) users as opposed to 1.5B FBB (Fixed Broadband)
• 70% of companies (especially SMEs) will be using cloud based services.

Mr. R. Swaminathan, Senior Executive Vice President, Reliance Communications Ltd.
• Low cost mobile networks and devices helped drive innovation in low cost business models in Rural India
• Customisation is a mecessity for the rural market.
• One offering includes a fixed phone that uses Mobile as a backhual using the Yagi antenna
• 15 operators in rural India. Voice tariff went from 20cents to 1 cent. Entry cost reduced by 95%
• ARPU in rural India is $2.
• Telecom operators have done innovations to keep costs to minimum
• Phone to tablet is best evolution for Indian rural market, using visual images and txt to speech technologies not smartphones
• Good to have some text to voice and vice-versa apps
• Ends with saying that there are 870 million people in rural India and possible market size is $25 billion that can be exploited

Kanwar Chadha, Chief Marketing Officer and Board Member of Cambridge Silicon Radio
• Innovations in location-aware wire-free connected world
• Spoke on view of local business vs global, very entertaining perspective , assume nothing and be careful of interpretation
• Example is the initial GPS cost $3700 but was still successful in Japan because guys wanted to show it off to their dates.
• Maslow's hierarchy of needs dont work for India as its more important to have entertainment (TV) than roof.
• FM very succesful in India but nowhere else.
• Sat Navs will not succeed in India because addresses and maps not very well mapped. Things like coupons, sms will be very successful

Innovation Hothouse: Mr. Christian Leicher, Member of the Executive Board at Rohde & Schwarz GmbH & Co. KG.

Session on start-ups very interesting
• Augmentra talked of GPS based smartphone apps. Users can share and get paid when someone else download what they share. Their guidebooks, etc are trusted by half the search and rescue mountain teams
• Oxems have a solution for the new plastic pipes that are being deployed. The normal metal detectors cant detect these pipes so they have a RFID based solution.
• Pneumacare has a non-contact medicare solution that can be used to track people with respiratory problems
• has a unique app discovery solution that can reach upto 6 millions users in 90 different countries.
• Cambridge temperature concepts has a solution that can increase the chances of fertility without IVF to the same levels after 6 months use.

Interesting points from the breakout sessions:
• Mike Bowerman of Alcatel-Lucent: Soon we will see pricing based on time of day, location, etc. Infrastructure sharing lower costs but it means that coverage from some location can completely vanish.
• John frieslaar of Huawei talks about how many will be connected to networks and the cause of demand
• Stephen temple says industry must spur innovation not gov.agree but will gov let us?
• 75% of UK mobile data consumption is driven by BBC iplayer, YouTube and adult videos says Sam Leinhardt of Penthera
• Ed Candy, 3: Apps evolving from Handset Apps to Widgets to Intelligent Browsers based
• Content is king but context is queen
• O2 in UK started putting data caps and lost 7K customers in London. They were using 7% of network capacity so O2 happy to get rid of them

Stephen Baily, General Manager, BBC R&D
• BBC R&D iPlayer usage on tablets is 3million/mth 2% of total
• Dual screens being explored by BBC with a universal controller API. The proposal has been submitted to W3C.
• Working on Dual-Screen concept where iPad becomes a complimentary device to TV (See
• BBC R&D iPlayer usage on tablets is 3million/mth 2% of total
• BBC is looking again at mobile broadcasting based on DVB-t2m standard
• 90% of broadcast os normal schedule than the time shifted one.

Dr. Tapani Ryhänen, Laboratory Director, heading Eurolab (Cambridge and Lausanne) of Nokia Research Centre
• Imagining tomorrow devices, creating technology today
• Morph concept video

• Nokia Research Center in Cambridge working in lots of futuristic technologies like Data driven Apps, Stretchable electronics, Bend to zoom, flexible phone and display
• Another video that I wasnt able to locate on Youtube

Few points from The "Can big wireless deliver on the promise of a big society?" Panel Debate
• Motorola's David Chater-Lea: "Due to spectrum needs we're going to see breakdown of barriers between commercial & private networks"
• Neul/Ofcom's William Webb: "To get a truely wireless society we need more small cells and increased backhaul. Then we need FTTH"
• Otherwise we're going to have situation where wireless will be held back by the wired network
• Public safety: should governments use private networks or commercial networks & give priority to emergency services over customers?

Graham Fisher, Former CEO of Orange Labs R&D, BathCube:
• Net neutrality doesn't work in a world of finite resources
• High end phones expectaions include screens that can work in sunlight, AR, 3D, etc.
• When it comes to retail price plans mobile operators are all in a bargain basement, they need to reintroduce value

Dan Reed, Corporate Vice President, Technology Policy and Strategy and eXtreme Computing Group at Microsoft
• The uber change happening is collision of computing/comms/content. We need to work out how to work together

Ken Blakeslee, Chairman of WebMobility Ventures:
• Digital natives vs digital immigrants
• Is mobile too inward looking?
• We're moving from hardware to software driven marketplace where communities are the new currency
• Users can be bought and bribed, communities can not

Interesting Obervation:
• Cambridge Wireless - run largely by women as an organisation but 95% of attendees at Future of Wireless conf #fwic are male

Poll of #fwic audience returns 50:50 re: whether mobile infrastructure should be common wholesale solution vs competitive between operators

Hopefully you have enjoyed this summary!

Sunday, 26 June 2011

Second hand report from the Femtocell World Summit 2011 (#FWS11)

Here is a summary from the Femtocells World Summit 2011 that I have compiled from different blogs and twitter. I was unable to be there, due to the expense, location and timing of this event it simply wasn’t feasible for me to attend. I am also disappointed that the organisers are not more welcoming of bloggers and do not understand how valuable our participation can be for the summit. Peronally, I would have taken a few pics of the exhibition, as I have done in the past, as it would have provided a better idea about the event to people in different parts of the world. Anyway, summay as follows:

DAY 1 began with Simon Saunders from the Femto Forum. Some of his points:

60% of consumers are interested in femtocells now Another interesting statistic was that there are now more 3G femtocells in the world than there are 3G macrocells, which again agrees with data stating that 60% of operators think small cells are more important than macrocells in the success of LTE.

According to the Ubiquisys blog: Simon’s thoughts are best summed up with a sort of rallying cry he came up with: “Our cells are small but our goals are not”!

This was followed by, Thilo Kirchinger, Principal Product Manager forVodafone Group. He discussed Vodafone’s operational stance on femtocells and small cells, and during which confirmed that Vodafone would indeed be launching LTE femtocells.

Thilo also spoke about how he sees femtocells integrating and being used by people in home environments. For example, instructions for home femtocells should be as simple, with as little technical information as possible, limiting potential confusion for the end user, while voice communications is still the biggest draw for this kind of residential femtocell (despite the fact that people tend to use a lot of data for things like app browsing when at home).
There are now 9 Vodafone subsidiaries with commercial femtocell service – almost a third of the total – and more are to follow shortly.
Research showed that some 34% of the UK either have insufficient or unsatisfactory indoor mobile coverage and Wi-Fi only partly solves the issue.
In summary, he'd like to see accelerated standardization of the Iu-h interface, for the femtocell supplier ecosystem to start engaging with the Connected Home industry and for femtocell costs to reduce further.
David Chambers of asked how operators, such as Vodafone, with strong brands of being the best mobile network and coverage could reconcile asking customers to pay for a box to fix poor coverage problems. Thilo felt that femtocells were complementary (especially for growing indoor use) and by offering both (ie great outdoor macrocell coverage plus great indoor femtocell service) it gave them competitive advantage. Another question related to 3rd party broadband internet – he reported that this hadn't been a problem, especially where customers conduct a speed test as part of the pre-sales process.

Telecom Italia’s Ferruccio Antonelli took the third slot of the day with a presentation focusing on the company’s commercial trial and proposed launch of femtocells in Italy.

Telecom Italia Mobile (TIM to the locals), has always been a bit of a trendsetter in the mobile industry and is one to watch. They have the highest penetration rate and smartphone takeup of any European country. They will launch femtocell services next month (the precise date is commercially withheld), with Alcatel-Lucent providing two sizes of femtocell (seems very similar to Vodafone products).
It won't be mandatory to use Telecom Italia broadband – any third party wireline/cable broadband can be used. While the pricing also can't be revealed, their billing system will be flexible enough to offer different prices when customers are using their "femtozone" at home.
It sounds like it’s been a time of experimentation in Italy for femtocells thus far, but signs are looking good, with Ferruccio stating that femtocells will see launch in the second half of 2011. There was also some discussion on Twitter stemming from Telecom Italia’s idea of a ‘femtozone’ tariff or simply keeping pricing the same.
A major issue for their implementation was the regulatory requirement to know if the femtocell has been moved (so that emergency services go to the right address) – this is checked by ensuring that at least one external macrocell ID is the same as when the unit was first installed and/or that the Telecom Italia broadband IP address matches.
Unusually, TIM want to have SIM cards to authenticate their femtocells – so for example faulty femtocells can be replaced and by swapping the SIM card would automatically reconfigure for that customer.

Some insights from South Korea was provided by Samson Tae-Yong Kim from SK Telecom, whose presentation focused on using femtocells for data offload.

Of particular note was the disparity raised by Kim in terms of data usage between different types of phone. For example, some smartphone users are consuming as much as 1 gig of data on an ‘all-you-can-eat’ plan in the same amount of time that it would take a feature phone user to consume 10 megs. It’s also worth mentioning at this point that 20% of mobile phone users in South Korea have smartphones, and this number is sure to grow.
South Korea Telecom (SKT) plan to deploy some 10,000 public hotspots before the end of the year, many equipped with both Wi-Fi and cellular. They've previously used a lot of repeaters to ensure excellent (voice) coverage, but now need to bring in heavy additional capacity and higher speeds.

Alcatel Lucent: Steve Kemp looked at how data usage is now ballooning – indeed, that we are now “nearing the practical limits of information theory” –
This is a generation that is watching 2 billion Facebook videos a month and 2 billion YouTube videos a day.
Alcatel-Lucent expects mobile traffic growth to be in the order of 30x in the next five years.
Just look at the iPad – users consumer twice as much data (and signalling) as the average iPhone user.
What’s the problem? Signal to noise. As Claude Shannon at Bell Labs in 1948 theorized, a network is limited intrinsically by the noise generated by the media and the users. As you get more users, it degrades the overall capacity of the network.
LTE, despite being more spectrally efficient than 3G, has a theoretical capacity limit, under Shannon’s Law, of 3.5 mbps per hertz.
The answer to this inescapable fact is to make the cell size smaller so that spectrum is more efficiently used. And use beam forming to focus spectrum where you need it, away from interference
Kemp then moved onto the business case for femtocells. You need initially to improve customer retention because keeping customers is a whole lot easier than gaining new ones.
Femtocells result in a 60% downlink improvement, and a 26% uplink improvement. With lower latency, customers are happier with their voice calls at home and churn less. You can build a business case for home femtocells on this alone.
Metro femtocells have even more compelling business case. The more traffic is offloaded onto small cells, the users on macro cell also see a service improvement.
Steve also raised a point that kept reappearing through the morning: iPads (and tablets in general) are far more data hungry than iPhones/smartphones, which is certainly food for thought when considering the sudden surge in popularity of these devices.
Alcatel-Lucent also announced their femtocell application developer kit, which is based on the recently published Femto Forum femtocell API specification. Already 23 developers have signed up to use it, with the first application to be made available by Telecom Italia when they launch.

As the morning progressed, it was the turn of Nigel Toon, CEO at Picochip, to present his thoughts and findings on the impact of femtocells on network performance and capacity.

Nigel noted that voice communication is still one of the most important reasons why people select a carrier. Nigel also raised the point that no one really knows by how much mobile data traffic usage is expanding (or due to expand), with various different proposals raised during day one of FWS 2011 alone.
Mobile data traffic exploding – you guess by how much. Is it 30x, 50x or even 1000x ?
Problem is carriers capex can’t grow at 1000x
Currently carriers spend, on average, 20% of their revenues on capex. And the cumulative amount of capex is increasing 8% year on year).
Need to serve customers more efficiently and at a lower cost.
Today a user in the middle of macrocell might only experience 40kbps. Tomorrow, with femtocells, the user can enjoy 8mbps while increasing the performance (less crowing on the macro) to 170 kbps.
The key to low cost deployment is self-organizing, self-configuring, interference management and remote management.
Picochip reaffirmed the issue of replacing repeaters with additional capacity, suggesting that rationing wasn't the right answer for customers who have grown to love data access. The web will only increase reliance on data connectivity and network operators will need to respond by building out a new network layer to meet demand.

Nitin Bhas from Juniper Research discussed the principals of mobile data offload and onload, where ‘offload’ means data migration from mobile to fixed, and ‘onload’ vice-versa. The spectre of tablets such as the iPad and smartphones being data hogs was once again raised during Nitin’s presentation, as was the important of the ‘offload’ of data due to this very reason.Mobile data traffic from smart phones, tablets and other devices to grow to 14,211 petabytes by 2015. This will be equivalent to 18 billion video downloads. By 2016, 63% of this will be offloaded to Wifi or femto.

Bill Chang, chief planning and strategy officer, UMobile explained that UMobile is a new challenger in Malaysia, challenging three well entrenched incumbents Digi (leader in price), Maxis (leader in products) and Celcom (leader in coverage.)
Malaysia has 120% penetration, expected to rise to 150% within 5 years. 28 m population.
70% of market revenues come from 8% coverage area. Highly urbanised. So when UMobile launched in 2007, made sense to target where 70% of the revenue was coming from.
Currently has 1200 node Bs and roaming onto 2G partner network.
Price is in decline in Malaysia, ARPUS are falling for voice. The market has reached revenue saturation.
Operators need data centric growth and they need it to be low costs business case. Makes sense to use femtocells. (In Malaysia, smartphones make up 65% of new phone sales)
Umobile has limited capex, so trialling femtos with Alcatel-Lucent. Using home and hotspot femtos.
Plan to launch femtos commercially. Will improve indoor coverage, data offload, reducing roaming costs (because they have to pay their 2G partner) and bundled services.
Malaysian govt has target of 75% BB penetration by 2016.
“Its a no brainer for us to give away femtos for free”
However their strategy is somewhat hampered by a local regulation (tax) of around US$600 per cellsite – not really significant for macrocells, but a serious problem for thousands of femtocells.

Continuous Computing launched their "Femtotality" software product. No longer limited to just the protocol stacks, they've invested an additional 150 man years in their application layer (I believe this figure includes an acquisition, otherwise their 200 staff would have been working a lot of overtime) and now offer SON (Self Optimisation), remote management and configuration features too.

NTT DoCoMo was able to restore cellular service after the earthquake/tsunami in just 6 weeks after 4,900 cellsites were put out of service in the Tohoku region alone - femtocells were part of the solution. They plan to switch off their 2G service next year and have already launched LTE. They intend to deploy LTE femtocells as soon as possible.

Finally, Broadcom’s Shlomo Gadot gave a provocative presentation where he outlined a compelling vision for femtocell technology. He sees no reason why Wi-Fi hardware should be cheaper than femto in future, and named integration as a key trend. Following this trend, Shlomo gave more details of the forthcoming integrated WiFi/Femto/ADSL residential gateway, the first of its kind, announced by Ubiquisys earlier that same day.


Dr Alan Law of Vodafone Group talking about femtocells beyond the home.

Vodafone’s vision started with consumer cells, and great things are happening both at home and abroad with this arm of their femtocell operation. But where do you take femtocells when looking beyond residential?
Vodafone has been trialling its enterprise and rural cells, and some interesting information emerged when Dr Law recounted some statistics from their rural and enterprise test deployments. The amount of dropped calls noticeably decreased when voice and data was offloaded onto the femtocell – which means better quality of service for Vodafone’s customers. There are still some challenging aspects to rural deployment such as IP transport and power locations, but on the whole results were positive.
Vodafone’s enterprise femto trials have also been successful, with data services noticeably enhanced in enterprise environments when femtocells were brought into the mix. The company’s ‘Metrozone’ concept would provide extra network capacity for data offload in denser urban areas.

Next there was a fascinating presentation from Rick Vergin, CEO of Mosaic Telecoms. He represents a rural telco, and outlined the problems of serving customer who live predominately in farmland or forest. It is desperate to deploy femtocells to not just plug gaps, but create coverage for the first time. Cellular coverage is the chief concern: macrocells can provide coverage to population centers (towns over 200 people) and microcells can support where people gather regularly (schools, for instance). But thousands live outside this coverage area.

First problem is geography: most of Mosaic’s customers live towns with 200 people up to a small city with 9000. But the 9000 square mile coverage area within its 3G license, comprises mostly farmland or forest – and potentially 100,000 people.
Mosaic runs 3G in band IV, a relatively underused part of the spectrum from a global perspective. This has caused unprecedented problems with femtocell vendors, with Airvana, Technicolor and Arcadyan all contracted only to subsequently drop out one at a time. Finally, with the guidance of Nokia Siemens, Ubiquisys was selected.
Farmland is not so bad, but forest is very challenging for the Mosaic’s 35 macro cell sites. CEO Rick Vergin lives 200m from a main road, and 2 miles from the nearest macro cell. On the road, he has line of site and 4 bar coverage, at home he barely has 1 bar coverage. Many of the potential customers in their licensed area have no coverage.
The femtos will bring coverage to people with currently little or no coverage. Moscaic has no intention to use femtos to create ubiquitous coverage – that would be way too expensive. But what they can do is give subscribers coverage most of the time: at home, at school, at the cafe. It will only be on the journeys between that they may have no bars.
The rural customers of Mosaic will also benefit from LTE because it will be used to backhaul the femto traffic and also provide broadband access for the first time (remember many of these properties will be far away from an exchange and may not use satellite or microwave. Mosaic will use the 750MHz LTE for residential broadband access, and bundle VoIP and femto/cellular with it. (750MHz is much more spectrally efficient than its 1700/2100 MHz 3G spectrum).
This is a great case study for not just the 1000 rural US telco but for any operator that either operates in the rural segment or has universal access obligations.

Peter Agnew of Colt Telecom took to the stage to present his views on what it takes to overcome the barriers to launching a femtocell service through fixed and mobile collaboration. If that sounds like a bit of a mouthful, all will become clearer in a minute!

Colt Telecom is a large pan-european fixed line operator, working in 21 countries with organisations such as major banks. Peter proposed that in working together with a fixed line operator such as Colt, mobile operators will have an ally in femtocell deployment, aiding connectivity, quality of service and increasing the mobile operator’s access to enterprises.
In essence, what Peter and Colt are proposing is ‘femto-as-a-service’ (‘FaaS’), which was met with some figurative nods of approval on Twitter. Peter finished his presentation by noting that for something like FaaS to work, self-organising network technology would almost certainly need to play a role in such a deployment.
It’s an important development for operators wanting to take their first steps in femto, which often starts with the low-risk bit low-volume enterprise route. This solution is the first to remove the barrier of high up-front gateway and integration costs, and the subsequent reliance on volume in the business case.
Another approach, and its not one that COLT said it would necessarily be offering, is to provide in-building installs (as long as there is not radio planning). It makes sense for a business telco with experience of firewalls, LANs and so on to assist both enterprise and mobile operator in this area.
In dense metropolitan areas, most subscribers are sitting within an office. It makes sense to bring coverage closer to these users, and not charge the enterprise for this (either for the access points or in-building cabling). It improves the coverage of the enterprise subscribers and for everyone else in the macro – both are sufficient incentives for the mobile operator to foot the bill.
However, more bandwidth available means more consumed. COLT asks, do mobile operators have the fixed-line infrastructure and core-network to cope with the increase in backhaul requirement?

Cisco’s Mark Grayson, spoke about mobile offload architectures. One of Mark’s main points that resonated with the Twitter audience following the #FWS11 hashtag was that the cost for networks is dealing with the non-uniform peaks in mobile internet demand.
In their previous experience with large sporting events like the Superbowl, Cisco noted that the volume of traffic leaving the stadium was greater than the volume entering – all thanks to social media services such as Facebook, YouTube, etc. with people sharing content, something that Intel’s Steve Price raised later on.
Mark suggested that the move to small cells will require a change in mindset, and put forward a suggestion for using converged Wi-Fi/femto architectures for macro offload of indoor traffic – and he also noted that cellular small cells would need to prove themselves at the high densities already deployed with WiFi.

Ubiquisys’ CTO and Founder Will Franks, with a presentation on the next generation of small cells.

Will started things off with a brief discussion on the evolution and naming of small cells, describing how things have progressed from early residential femtos, all the way to some of the especially advanced outdoor and rural models.
The building blocks for the next generation of intelligent small cells, Will stated, are 3G, LTE and Wi-Fi. This, combined with the continuous adaptive behaviour offered by our self-organising network technology, helps Ubiquisys small cells to form part of the recently discussed ‘Edge Cloud’ – something also raised in Intel’s presentation.
Will went on to describe how small cell hotpots will be deployed in the real world, and broke down small cell technology into layers. Starting with the hardware platform (featuring Texas Instruments’ simultaneous dual-mode 3G/LTE), through continuous self organization and self organizing networks, and on to edge cloud computing platforms (Intel) and cloud control systems.
Ubiquisys reported that Softbank Japan have been able to deploy rural femtocells in just 3 days using satellite backhaul. Their "self optimising femto grid" even works for clusters of rural femtocells at 2km range.

Competitive operator Network Norway, thinks it has the answer for small businesses in Norway.
Combine mobile centrex with femtocells. Norway is a country that was at the leading edge of fixed-mobile substitution.
According to Network Norway, 64% of all calls originate on a mobile and 79% of call minutes terminate on a mobile. This is a very mobile friendly country and, believed Network Norway, businesses would be very receptive to mobile centrex.
The problem is buildings: all that concrete, glass and basements make ditching the desk phone an impossibility unless you can bring the mobile network indoors. DAS (distributed antennae systems) are too expensive for most small businesses. Femtocells are not.
Network Norway launched a small business femtocells to make their Mobile Centrex service more compelling. The mobile PaBX service offers hunt groups, stats on attendant function, private number plans, conferencing etc.
What is interesting to me is that they have built smartphone apps (for Ovi, iPhone and Android) which allows users to set up conferences and see presence/availability in contacts (which comes from femtocells).
In other nomenclature, this is called “collaboration”. Or even unified communications, if you use the IM, email and SMS functions on your smart phone.
So benefits for small businesses: flexible communications, collaboration, guaranteed coverage in the office, seamless experience, no capex.

The last presentation day 2 featured Steve Price of Intel, with a look at how to ‘differentiate the small cell user experience with an intelligent, application enabling architecture’.

The internet and mobile internet are both growing rapidly, with the “Gigabit Generation” particular fixated on social networking, which now has a considerable impact on network traffic at large. Service providers are now presented with a great opportunity, Steve said, as they can now take advantage of the fact that they are directly involved in the process.
The next step is to make sure that intelligence is present throughout the network – and just as important is its location. These intelligent services ensure that the user will be getting a better experience in the end.
The two key trends identified by Intel were cloud RAN, with China Mobile named as an example, and edge cloud, where the Intel-Ubiquisys collaboration was given as a prime example.

Individual Contribution: Tom Lismer
Residential Femtocell Access Point Design and Technology Innovation: Picochip
Non-residential femtocell access point design and technology innovation: Alcatel-Lucent
Femtocell Network element design and technology innovation: ip.access
Femtocell Application: New service or technology: Alcatel-Lucent
Progress in commercial deployment: Huawei
Commercial deployment – Marketing Campaign: Vodafone
Commercial Deployment – technical implementation: Vodafone
Contribution to Femtocell Standards: Nokia Siemens Networks
Enabling Technology: Texas Instrument
Social Vision: NEC
Judges Choice: Rakon

Complete Details on Femto Forum Website here.


Surprisingly there wasnt much coverage from Day 3. My observation is that by the third day, the people get really tired and its just the analysts who are still around learning, discussing and participating as much as they can. The only summary I found is from the Think Femtocell blog. Here are few interesting points:

The femto vendor community seems to be frustrated by the slow rollout of Femtos by the network. The technology has been proven and from what I see, if a network is rolling out Femtos, they are getting good reviews and reception from the user community, even though they may have to shell out a few bucks.

Verizon reported tremendous success when using their femtocell (the Verizon Wireless Extender) to reduce churn. They've also successfully offloaded heavy users from their macro network in Chicago, by sending them a free femtocell – both improving speeds for those high users as well as releasing capacity on the macro network for others to benefit from. Their femtocell solution works well and they're very happy with it. You still can't buy a femtocell in a Verizon store because It doesn't fit with their corporate branding of having the best network.

In contrast, Vodafone don't seem to have suffered any loss of brand image by promoting Sure Signal – their network brand remains strong and is arguably strengthened by saying they are the only one who can truly guarantee full service indoors anywhere (assuming there is a DSL line to connect with). Vodafone Ireland jokingly apologised for the lower approval figure than Vodafone Greece during their femtocell trials - only 96% (against 98% in Greece) would recommend them to their friends and family. They explained how they had carefully crafted their marketing message to celebrate the positive aspects of their customer's individual homes (thick woods, stone buildings, basement flats etc.) and how simple it was for them to have 5 bar coverage.

Comcast have built out a lot of Wi-Fi hotspot capacity in addition to their wireline/cable services. They believe in the long term, the usage mix of traffic on wireless will be a similar profile to wireline today – say 50% entertainment (including video), 20% web surfing; a total of 13GB/month. Comcast has deployed some 5000 WiFi hotspots so far, and plan to build out 100K over the coming years.

Wi-Fi has some new features coming – the new HotSpot 2.0, which Comcast will be trialling later this year. Greater use of the 5GHz spectrum will help reduce congestion in high traffic areas. Sports stadiums seem to be the biggest challenge – many users wanting to watch video at the same time, with others trying to use Mi-Fi (cellular to Wi-Fi adaptors) at the same time/in the same spectrum.

Contela explained how they use femtocells in Korea to offload data traffic. Unusually, the system deals with voice and data traffic differently – switching voice calls to the normal macro network while handling as much data traffic as possible through femtocells and Wi-Fi.

TOT, Thailand, a relatively new entrant to mobile explained how they can install femtocells at public payphone booths as a quick way to find sites with backhaul connectivity (using DSL) and power. Getting the height of the unit is important – it needs to be slightly out of reach. They also showed their disaster recovery solution – which uses femtocells + satellite backhaul and can be rapidly deployed. In these situations, providing a fixed/wireline phone service isn't useful – most people now have all their phone numbers held in their mobile phone and not written down. Mesh backhaul, linking clusters of femtocells to each other using wireless and aggregating the backhaul to a few egress points, is also a useful option – a maximum of 5 "hops" using a so-called spine and rib architecture matches urban street layouts.

Stuart Carlaw from ABI Research. Growing number of employees have more than one phone they use in the office (one corporate + one personal). Both phones have mixed voice/data use. After some retrenchment in 2009, voice has continued to grow and is now 779 minutes average for corporate users. Video and picture messaging are being used by enterprise users (on their corporate liable phone) more than ever before. The growing demands of employees are giving their IT departments a major headache, for which enterprise femtocells will be a major part of the solution.

The Femtocell Application developers toolkit from Alcatel-Lucent isn't locked into their solution. Applications developed and tested using their SDK should also work with any other femtocell system that also conforms to the Femtocell Application API.

There were a number of operators present at the conference who are clearly there in an active capacity. Most were pretty tight lipped about their plans, but all seem to acknowledge that femtocells will play some part in the story.

Some Final thoughts from the Ubiquisys Blog.

The latest Informa femtocell market status report, produced for the Femto Forum this week, confirms the strong growth trend with nine new commercial launches in the past quarter alone.

Both operators and vendors alike were talking about femto technology being used in public-space small-cell hotspots to provide a capacity boost in high demand areas. At least half of the presentations touched on this topic in one way or another. Is it because the growth in data demand is beginning to be felt? Or is it that the low opex and backhaul costs of femto are making a strong business case? In any case, many of the questions about public space small cells were mentioned, such as interoperability with the macro layer and how the necessary high density deployment of small cells will be achieved. The questions were mentioned, but solutions were not – a sure sign of innovative work in progress.

Colt Telecom unveiled femtocell infrastructure as a service. Because many operators want to make their first femto launch into a low-risk segment, they often opt for SME (small business) rather than consumer segments. Yet the lower volumes in SME can damage the business case, because the upfront costs of the core gateway and systems integration are shared between fewer customers. By offering an incremental managed service cost, fixed line provider Colt might just have made it easier for mobile operators to start femto services.

Broadcom unveiled a fully integrated femto residential gateway, Texas Instruments won an award for their powerful new 3G/LTE SoC, and Intel presented a future powered by compute platforms in both cloud RAN and edge cloud environments.

There was a degree of consensus that LTE will be seen first in small cell hotspots, the same hotspots that need to deal with a deluge of 3G data demand over the next few years. Several speakers mentioned that this calls for small cells that can run 3G and LTE simultaneously, like those new SoCs from TI.

A few years ago you would have seen quite a few femto vs. Wi-Fi presentations, but no more, which is quite a relief to us, as we have been behind combined femto-Wi-Fi devices since 2008. There was much discussion of harmonisation in home and business environments. In public spaces, the idea of tri-mode small cells replacing Wi-Fi hotspots was raised. These would maintain the Wi-Fi capability, but add 3G and LTE cellular, opening the possibility of using cellular’s invisible “login” to replace Wi-Fi’s usual usernames and passwords.


Pics Source - Ubiquisys Blog

Report compiled from:

Wednesday, 15 December 2010

SON in Heterogeneous Networks

Another presentation from the SON 2010 Conference based on yesterdays theme of HetNets.

Presented by Seungpyo Hong of Institute of Network Technology, SK Telecom.

Friday, 22 October 2010

IMB and TDtv (and DVB-H)

Its long time since I blogged about TDtv. Its been quite a while since I heard about TDtv. Apparently its been superseded by IMB, aka. Integrated Mobile Broadcast.

IMB is used to stream live video and store popular content on the device for later consumption. This results in a significant offloading of data intensive traffic from existing 3G unicast networks and an improved customer experience. The multimedia client features an intuitive electronic program guide, channel grid and embedded video player for live TV viewing and video recording. All IMB applications can be quickly and cost-effectively adapted to support all major mobile operating systems and different mobile device types, including smartphones, tablets and e-readers.

IMB was defined in the 3GPP release 8 standards, and was recently endorsed by the GSMA as their preferred method for the efficient delivery of broadcast services. In June 2010, O2, Orange and Vodafone – three of the five major UK mobile operators – announced that they have teamed up for a three-month trial that will explore IMB wireless technology within a tranche of 3G TDD spectrum.

This spectrum already forms part of the 3G licenses held by many European mobile operators, but has remained largely unused because of a lack of appropriate technology. Currently, 3G TDD spectrum is available to over 150 operators across 60 countries, covering more than half a billion subscribers. IMB enables spectrally efficient delivery of broadcast services in the TDD spectrum based on techniques that are aligned with existing FDD WCDMA standards. This enables a smooth handover between IMB and existing 3G networks.

Issues that previously limited uptake of IMB, or IPWireless' tdTV system, have now all been addressed. Namely, the standard now allows for smooth handover between IMB and unicast delivery; has the potential to be integrated onto a single W-CDMA chip rather than requiring a separate chip; and has resolved interference issues with FDD W-CDMA, at least for spectrum in the 1900MHz to 1910MHz range.

IP Wireless already had a trial at Orange and T-Mobile in the UK (which have just agreed to merge), but in that pilot each 5MHz segment only gave rise to 14 TV channels per operator. The new standard could support 40 separate TV channels if two operators shared their TDD spectrum.

The GSMA announced its support and is backed up with additional support from both IPWireless and Ericsson as well as operators Orange, Softbank and Telstra.

There have been recently quite a few bad news for DVB-H and on top of that IP Wireless has announced that Samsung is going to be releasing phones with IMB support so it may be that we will see IMB sometime next year.

The GSMA paper that details IMB service scenarios and System requirements is embedded below:

Thursday, 29 October 2009

LTE definitely needed and coming next year...dont mention Voice and SMS please

The unremitting growth in data traffic will bring about a 3G network capacity crisis for some mobile network operators as early as 2010. This dire scenario, according to a new study from Unwired Insight, will only be avoided by the early deployment of LTE, and the acceptance that additional LTE spectrum will be required to satisfy this demand.

With 3G traffic volumes set to increase by a factor of 20 by 2015--driven by many technology factors and also dramatic reductions in mobile data pricing--Alastair Brydon, co-author of the new study, points to the example of mobile broadband pricing that has fallen as low as US$2 per gigabyte, "which is nearly half a million times smaller than the price per gigabyte of an SMS message."

Brydon believes that early LTE will be necessary for the following reasons:

  • As 2G users continue to migrate to 3G services, the available capacity per 3G user will decline rapidly in networks utilising HSPA, to less than 100MB per user per month in some cases. LTE will be essential to counter this decline.
  • While LTE promises peak data rates of over 100Mbps, this is only possible with wide allocations of spectrum, and particularly good radio conditions. Average data rates from practical LTE networks will be nowhere near the peak values.
  • Network operators will have an insatiable appetite for LTE spectrum, to stand any chance of keeping up with forecast traffic demand. For some operators, 10MHz of spectrum will be able to support forecast traffic levels only until 2011. A further 10MHz will be needed by 2012 and another 10MHz in 2013.
Unwired Insight claims LTE's ability to relieve the capacity constraints of HSPA networks will be limited initially, until operators can acquire additional spectrum and seed a sufficient number of LTE devices in the market place. "But, we don't expect to see LTE handsets until 2011," the company warns.

Fourteen operators have committed to LTE rollouts next year, up from 10 in March, the research firm said. It predicts the LTE network gear market will be worth more than $5 billion by 2013, dominated by E-UTRAN macrocell (eNodeB) deployments.

It also expects the LTE customer base to top 72 million by 2013, mostly users with laptops, netbooks or dongles, with the first smartphones expected to hit the market after 2011.

In another forecast, Informa Telecoms and Media said Japan would account for more than half of Asia's 14.4 million LTE subscribers by 2015.

NTT DoCoMo, Japanese rival eMobile and China Mobile will be the first to launch LTE in the region, Informa said, with Hong Kong's CSL likely to follow soon after.

But rollouts in the region may be hindered by delays, as Japan and Hong Kong are so far the only Asian countries to have awarded spectrum for LTE.

Regulators in other nations are scrambling to free up enough spectrum, Informa added. Even in Japan, there is not enough 2100MHz spectrum available to support DoCoMo's full LTE plans, so it will use its newly allocated 1.5GHz for LTE from 2010.

According to news sources in South Korea, LG Telecom (LGT) quietly revealed their intention to migrate to LTE for 4G service in South Korea. LG-Nortel and Samsung will provide the multi-mode base stations which are part of the company's green network upgrade. SKT and KTF (now part of KT), the other two mobile operators in the country, have already announced their LTE migration path for 4G previously. Unlike SKT and KTF who will migrate from HSPA to LTE, LGT will go from EV-DO to LTE, similar to the case of Verizon Wireless.

It was probably a matter of time for LGT to announce the LTE migration plan since it was only running EV-DO network, and this officially puts LGT on the LTE camp. Now, my speculation is that other major EV-DO operators (noticeably, Sprint) who haven't announced such plans will follow the same path down the road since WiMAX does not seem to be a viable migration path for the FDD part of the network.