Sunday 22 May 2011

LTE World Summit 2011 - Pics and Notes from Day 1

Here are few pics and discussions from the day 1 of LTE world summit 2011. They are quite brief and I will try and add some info from the tweets as well.

Adrian Scrase from 3GPP said that as there are already over 200 operators committed to LTE, its the fastest growing mobile technology ever.

Bart Weijermars from TMobile Netherlands said future growth will be data centric fuelled by appealing terminals, new usages, broadband and ubiquity.

Future challenges include Network sharing, required to keep the cost under control; everyone allowed to roam freely everywhere; all the content is stored in the cloud and voip is the only option for voice.

New services will be possible with the advent of '4G' but care has to be taken because background apps are already using up a lot of capacity.

There is still work that needs to be done on Spectrum, Smartphone challenges and Network of networks.

Huawei has been one of the main sponsors of the event and the award and Ying Weimin spoke on how LTE is the way to more competitive Mobile Broadband.

According to him, wireless solution is a personalized solution and will go everywhere you go.

Spectrum is the main concern though as a combination of low and high frequency will be needed. Hetnets are coming and they will be the future of the networks.

The way forward is to start the LTE with data only and build on top of that. The network should be simple evolution and will contain of cloud baseband, wideband RRU and AAS.

Pocket Wifi is definitely going to be a killer device and Innovative LTE business and Apps will be needed in the long term like Instant LTE broadcast, Wireless Video surveillance, etc.

LTE is faster than expected and this is the reason there are so many operator commitments. Huawei has 40+ LTE contracts and 10 have already been launched. This is just the beginning.

Seizo Onoe from NTT Docomo spoke about Crossy. In fact during Christmas the employees were wishing each other 'Merry Crossy'. Docomo believes that the users dont care about HSPA or LTE so the Crossy is a service they are selling to the users.

Docomo are getting 75Mbps max DL speeds (using 10MHz band). The phones are capable of 100Mbps though (category 3).

Docomo has recently announced 24 new devices. 2 are Wifi routers. Unlimited data plans on the LTE network cost 5000 yens.

On the spectrum side they are expecting the LTE network to co-exist with UMTS and will be using the 2.1GHz band. In fact Docomo thinks that 2.1GHz should be the universal band that all devices should support so In future when the networks are deployed all these existing devices start working without problems.

The RRE equipment that Docomo has been deploying works with both HSPA and LTE.

Japan has already shutdown its 2G PDC networks but other cannot do the same for GSM. Onoe-san believes that we should stop the evolution of 2G as EDGE has still been evolving and we should focus all the energy into LTE evolution.

Onoe-san did emphasise that LTE is 3.9G and not 4G.

I did check with Onoe-san later that as FOMA was not compliant to the '3G' standards completely, is Crossy compliant to the LTE Specs completely and he said it is.

Onoe-san also said that ETWS was very helpful in the recent tsunami in Japan and services like these should be standardised quickly as they will be useful for someone or the other.

Thomas Wehelier from Informa presented the survey results of LTE ecosystem.

In 2011, the LTE deployments will quadruple but 2012 will have most deployments. Spectrum fragmentation is still prevalent but the core bands for LTE are 800MHz, 1800MHz and 2.6GHz. Capacity still cited as a big driver for the deployments.

TD-LTE represents a new market and new opportunity. In fact Ericsson bills this a year of TD-LTE.

Chris Kimm from Verizon spoke on their 2020 vision.

In Dec. 10 Verizon were covering 110 million people in the US by 2013 the plan is that 290 million will have LTE coverage. In fact LTE was launched in new cities on the day. At the moment though only 250K users are using LTE.

The rate of change is breathtaking and as a result CIO has changed from Chief Information officer to chief innovation officer.

In the Q&A, Chris said that they will deploy voice in 2012 using VoLTE. OTT voice will also be ok.
Tommy Ljunggren from Teliasonera spoke of their LTE deployments.

Last year they had 2 'kids' (as he called their networks) but now they have 4 more. Norway and Sweden got their LTE network in 2009. Network in Finland was launched 30th Nov. 2010. Then on 9th Dec 2010 network in Denmark was launched followed by Estonia on 17th Dec. 2010 and finally Lithuania on 28th April 2011. In fact in Estonia the network was launched 6 minutes after the auction.

Their deployments are in 800/1800/2600 MHz band. This will give them capacity and coverage.

In Sweden the downlink speed is over 20Mbps. In Nordics and Baltics the end users can roam without borders.

Once consumers are using 4G they dont want to go back to 3G. During the royal wedding on Stockholm last year, LTE was used by a TV station to transmit from 6 movable cameras without the need of satellite or any other connection. The transmit was without jitters and a revolution. Nippon TV used LTE to transmit the Noble Peace prize live from Sweden to Tokyo. In fact a bank uses 4G connection as a backup.

TeliaSonera plans to make money by having data caps in place, monthly fees, etc. VoIP would be charged. Right now the charges are €60 for 30GB allowance on LTE.

Cameron Rejali from BT Wholesale spoke on whether the future of mobile was fixed.

According to him WiFi offload is just the start as whenever the speed of the network increases the data usage increases as well.

The network has to do a balancing act. Greater user experience versus network complexity and seamless mobility.

Finally with LTE we will have true convergence at last. The future of mobile is fixed and that of fixed is mobile.

Panel Discussion:

Adrian Scrase brought up the topic of Global Roaming. As there are already 30 bands specified for LTE, do we need a roaming band that should be standardised. Should this band be an Industry initiative or will it be market driven?

The consensus was that this will be market driven.

Question was asked if LTE will be more profitable than previous technology.

NTT Docomo believes that LTE as a technology cannot generate new revenues but the services around it can.

Adrian Scrase asked the question that a lot of Services are defined by the standards but most of them do not get deployed. Does NTT DoComo think ETWS has been defined correctly as per the standards.

Onoe-san from NTT Docomo said that this service has been of utmost importance in the recent tsunami disaster. Even though the service was implemented and available on the phones, it was not used so people were not aware of it. So when the disaster struck everyone was surprised to receive this message. Now everyone knows about this service. Docomo has been using meteorological data since 2009 for this service.

In response to another question Tommy from TeliaSonera said that they will have CSFB next year for voice and VoLTE later. I asked similar question to Onoe-san about the voice support in crossy devices and he said that they will support CSFB. Someone did mention in the panel discussion that VoLTE is not needed and CSFB is enough.

That was my summary of the first day of #LTEWS. You can read the twitter conversations that have much more information.

Thursday 19 May 2011

Twitter Conversations from the 7th LTE World Summit

The following are the twitter conversations from LTE World Summit 2011. If you use twitter, it would be easy for you to understand them otherwise may be a bit of problem.

Tweets By:
Ajit Jaokar (@AjitJaokar)
Dan Warren (@TMGB)
Dean Bubley (@disruptivedean)
Dimitris Mavrakis (@dmavrakis)
Gabriel Brown (@Gabeuk)
Hamad Al-Marzouk (@dukeq8)
Kris Szaniawski (@kszaniawski)
Mico Naddeo (@kingcharles1666)
Olga Holin (@oholin)
Thomas Wehmeier (@twehmeier)
Zahid Ghadialy (@zahidtg)

17th MAY

@dmavrakis: 3GPP: 200 operators have committed to LTE, fastest adoption cycle in mobile

@dmavrakis: Bart Weijemanrs, MD of T-Mobile NL on stage
@dmavrakis: T-Mobile, TeliaSonera and Orange teaming up to drive device evolution for LTE at 1800MHz in Europe.
@disruptivedean: T-Mo NL speaker at #LTEWS sees network sharing as important. But also flexible to serve diff customers Sharing & flexibility compatible?
@disruptivedean: According to TMo NL speaker WhatsApp apparently got to 70% adoption on NL smartphones in 3 months! No wonder SMS revs fell at KPN.
@disruptivedean: Killing the MB traffic myth. TMo NL at #LTEWS says biggest problem is mass signalling load from simultaneous smartphone users
@dmavrakis: T-Mobile NL: Mobile needs developer guidelines to promote and to understand interdependency between internet and telco worlds
@dmavrakis: T-Mobile NL: spectrum, smartphones, voice, backhaul and seamless user experience between diff networks still need work for LTE
@twehmeier: I wonder if progress on LTE1800 in Europe will be slowed by legal challenges in the same way UMTS900 was hampered in past five years
@disruptivedean: So, who has been fired at operators, vendors or standards bodies for completely missing impact of app signalling when LTE was designed?
@dmavrakis: Common denominator in discussions and presentations so far in #LTEWS: Spectrum is the biggest challenge.
@twehmeier: Funny how the operators talk up the signalling challenge, whilst the vendors seem to think there are *quick* fixes. Quick=pricey?
@twehmeier: If planning to be “most operator-friendly platform” wouldn't you be spending BIIIIG on educating apps devs on being signalling aware?
@TMGB: @twehmeier I've been banging on about signaling aware apps and cross-app update correlation for months. Who do you think should spend big?
@disruptivedean: @TMGB @twehmeier Nobody bothers telling / helping developers to create signalling-lite apps. Or orients user pricing around signalling load.
@TMGB: @disruptivedean handset vendors hold the relationship to the app community and could implement the smarts for 'connect once, update many'
@disruptivedean: @TMGB yes & no. Apple was supposed to be trying but no sign yet. Best approach I've seen is SEVEN's signalling proxy & client in IP stack
@disruptivedean: Seems like consensus LTE1800 is critical for success. 2600 not viable for wide usage, DigDiv not enough capacity. regulator headache
@Gabeuk: If I heard right, T-Mo Netherlands CEO implied 70 percent of domestic smartphone users are using WhatsApp. Wow, that's viral.

@oholin: Nice to see that the focus is on enabling services above single connection technologies

@oholin: Huawei talking about LTE enabled services
@dmavrakis: Huawei believes Pocket WiFi will be a killer application for LTE
@dmavrakis: Huawei also looking at applications focusing on LTE uplink, including video broadcasting for professionals.
@kingcharles1666: Pocket Wifi. Like mifi but faster.
@oholin: Need to check Huawei's numbers on commercially launched networks. They say 20. Don't think that's right
@twehmeier: @oholin Depends on how strict you interpret “commercial launch". A few are technically pre-commercial trials for sure.
@oholin: @twehmeier I guess they should have better used our data though for their purposes 20 seems a better number

@dmavrakis: Seizo Onoe of NTT DoCoMo now on stage to talk about their LTE service called Xi (Crossy)
@dmavrakis: NTT DoCoMo: LTE service launched in Tokyo, Nagoya, Osaka (Dec 2010) with a phased coverage increase during 2011/2012.
@dmavrakis: NTT DoCoMo decided to share its 2.1GHz spectrum between UMTS and LTE and will deploy LTE at 1.5GHz during 2012.
@oholin: Docomo says spectrum needs to be used more efficiently and shared between technologies. I am a fan of efficiency
@disruptivedean: DoCoMo pitching LTE2100 shared with HSPA2100. Recommends all phones support LTE2100. Use my band & give me scale economies please!
@dmavrakis: Do most operators have enough spectrum at 2.1GHz to implement NTT DoCoMo's suggestion? I.e deploy LTE on existing 3G spectrum
@twehmeier: @dmavrakis Most in Europe have 2x10MHz. Very tough to refarm anytime soon to LTE. 2.1GHz is where all smartphone traffic goes...
@dmavrakis: @twehmeier Very true. Risk of cannibalizing UMTS capacity would scare most operators in Europe.
@dmavrakis: Deploying LTE at 2.1GHz makes sense in Japan, where indoor systems (DAS, tuned for 2.1GHz) are heavily deployed in urban areas
@kingcharles1666: Top tip from NTT: stop evolution of old technology. They will shut off 2G soon!
@TMGB: @kingcharles1666 NTT DOCOMO 2G is PDC, only deployed in Japan. Choice between turning off GSM vs. WCDMA is not so straight forward #LTEWS
@twehmeier: @kingcharles1666 If only operators in Europe had invested to ensure 3G coverage was >or= 2G like DoCoMo.

@TMGB: Tracking all the 'signaling kills the network' type tweets from #LTEWS - someone should make the point that LTE connections are always-on.
@disruptivedean: @TMGB multiple layers of signalling. Eg extra reporting on network quality and QoE from device midcall on volte...
@dmavrakis: @TMGB What about early deployments where devices will roam between LTE and 3G? Apparently HTC Thunderbolt is v aggressive #LTEWS
@TMGB: @dmavrakis if you're on LTE, you have to be connected else there is no signaling path (it's over IP).
@TMGB: @disruptivedean and agree on multiple layers, but is/was the connect/disconnect signaling that caused original problems.

@disruptivedean: No obvious mention anywhere on #LTEWS exhibition floor of SMS over SGs. Does it actually exist?

@dmavrakis: We are back! @informatm's @twehmeier on stage now presenting global survey results of the LTE ecosystem.
@dmavrakis: @twehmeier: 2011 will see LTE deployments to quadruple, but we will have most deployments during 2012
@kszaniawski: @twehmeier on stage presenting global survey results of the LTE ecosystem. 65% respondents set to launch in next 18 months
@dmavrakis: @twehmeier: Core bands emerging for LTE. 800, 1800, 2600MHz each have 15-20% of operator interest
@kszaniawski: @twehmeier on survey results - capacity cited as big driver in LTE survey >interesting as not what we're seeing with early launches
@oholin: @twehmeier talking about the results of first LTE survey
@dmavrakis: @twehmeier: TD-LTE represents a potential new entry into mobile market for greenfield operators and vendors #LTEWS
@kszaniawski: @twehmeier on survey results - 62% think primary solution for voice over LTE will be IMS-based VoIP (VoLTE) not surprisingly
@kszaniawski: @twehmeier on survey results - 58% think LTE will be complementary to fixed broadband. Only 17% think LTE will heavily substitute
@dmavrakis: @twehmeier: Most mobile operators believe LTE and WiFi will coexist and most of them will also need to have a carrier WiFi strategy
@dmavrakis: @twehmeier: 60% of mobile operator respondents believe that small cells will be necessary for LTE. #LTEWS
@disruptivedean: 62% of surveyed MNOs reckon IMS VoLTE will be primary voice tech for LTE. 23% reckon OTT-style, 13% CSFB. Wonder if 2012 will be same
@disruptivedean: LTE smartphones (on 3GPP networks) will need to have Grade-A SMS usability. 2 yrs ago we were told SMS was fixed for LTE. But where is it?
@oholin: @twehmeier: the factor which will have the greatest impact on the growth of LTE maturity: ( customer education) based on survey

@dmavrakis: Verizon's Chris Kimm on stage now to discuss about the role of 4G LTE in Verizon's 2020 network vision.
@dmavrakis: Verizon launched in 38 markets and 60 major airports covering 110 million subscribers during December 2010.
@kszaniawski: Chris Kimm of Verizon - how can I buy everything as a service & not have to commit. This the business opportunity for 4G
@dmavrakis: Verizon expects to enable VoLTE during 2012. Question via @disruptivedean
@disruptivedean: VZ speaker #LTEWS parachuted in from non-US corporate division, not VZW. Little new info. Claims VoLTE still OK for 2012, but also OTT VoIP
@dmavrakis: TeliaSonera's Tommy Ljunggren on stage now to discuss about their experience with LTE.
@oholin: TeliaSonera: trust this technology
@dmavrakis: TeliaSonera reiterates same message as last year on LTE: Mobile operators should stop trialing and start deploying.
@dmavrakis: Valid concern of a multinational operator: TeliaSonera prays that regulators will grant same frequencies to its opcos (800,1800,2600)
@kszaniawski: Tommy Ljunggren of TeliaSonera says need to get momentum going behind initiative for 1800 spectrum band as is most promising for #LTE
@dmavrakis: TeliaSonera aims to harmonize spectrum across its footprint: 900, 1800 (2G); 900, 2100 (3G); 800, 1800 and 2600 (4G) - all MHz
@kszaniawski: Tommy Ljunggren of TeliaSonera says won't offer VoIP for free, you will have to pay for it if you’re not an exclusive subscription
@dmavrakis: TeliaSonera is charging according to volume for MBB: LTE tariff is €60 with 30GB of monthly allowance. HSPA is around €35
@disruptivedean: TeliaSonera intending to charge extra for VoIP on LTE. Good luck with that. Start by defining 'voice' maybe? VPNs, workarounds etc.
@Gabeuk: @disruptivedean he wasn't specific thou. re: carrier VOIP vs OTT VOIP and what would be blocked/charged. No detail at all
@disruptivedean: @Gabeuk I know. His vendors & lawyers must be having fun defining 'voice' vs. 'telephony' vs. 'streamed audio' vs. 'sound' & metadata #LTEWS
@Gabeuk: TeliaSonera exhorting other operators to lauch LTE, but no actual new info or data to make the case.

@dmavrakis: Cameron Rejali of BT Wholesale on stage now to discuss whether the future of mobile is fixed.

@dmavrakis: Welcome everyone to #LTEchat. Live from #LTEWS, during a live panel including mobile operators and @twehmeier.
@dmavrakis: Global roaming argument brought up: 30 spectrum bands standardized by 3GPP so far from a global standard.
@dmavrakis: Consensus is that some harmonization should happen around spectrum. Natural forces rather than industry initiative likely
@dmavrakis: NTT DoCoMo expects that roaming will be an incremental process starting with some key bands
@dmavrakis: TeliaSonera highlights CSFB and VoLTE problem. Operators have some breathing room since most LTE devices are now data only
@dmavrakis: Question to panel: Is the profit potential of LTE credible?
@dmavrakis: TeliaSonera: mobile industry should sort out the roaming business model before switching on LTE.
@dmavrakis: NTT DoCoMo: Not sure whether LTE can allow a new premium revenue stream compared with existing HSPA/HSPA+ networks
@kszaniawski: NTT DoCoMo statement about RoI interesting. How do you assess/measure the business case for LTE network investment?
@dmavrakis: NTT DoCoMo on tsunami warning system (ATW): it's now a necessity and standardization needs to happen quickly
@dmavrakis: TeliaSonera highlights the lack of vision from regulators, so operators bet on all bands since they don't have visibility#
@dmavrakis: Spectrum issues dominate operator panel discussion. Regulators cited as primary culprit.
@kszaniawski: DT more philosophical than TeliaSonera saying that you will see the main bands determining developments altho a lot later
@kszaniawski: Doesn't seem to be much sense of urgency over using VoLTE either. Leisureliness seems to be a bit of a theme at conference

@disruptivedean: Opinion from AM of #LTEWS is negative. Spectrum fragmentation, slow rollout & uptake. And when VoLTE is ready, old telephony mostly obsolete

@twehmeier: Loved Question on iPlayer. Good reminder that this industry can be incredibly insular when you've got head down doing your day job

@twehmeier: Skype's biggest objective is to monetize more users. Strikes me that operators should be able to help them with that. Partnerships time.

@oholin: Not pointing fingers but I call to minimise the use of "ubiquitous" especially since it's used without understanding what it means

@disruptivedean: Stupidest idea of the day at #LTEWS 'email & social network only' data plan. Cue Facebook putting VPN, video & web proxy in next app update
@disruptivedean: Good to hear Ericsson at #LTEWS explicitly describing VoLTE as a *telephony* standard and not a general *voice* app one
@disruptivedean: Spoken to one silicon company at #LTEWS who agrees with me: may see dual-radio LTE + 2G voice phones as alternative to VoLTE or CS fallback
@zahidtg: @disruptivedean I was talking about exactly the same thing with Adrian Scrase in lunch :) #LTEWS. Good option as radios are becoming cheap
@disruptivedean: Unconvincing groupthink about role & control of WiFi by mobile operators at #LTEWS. Needs much better & granular approach to user needs
@disruptivedean: Seems to be a belief at #LTEWS that all smartphone WiFi usage is 'offload'. Industry needs to understand there's many other WiFi use cases

@kszaniawski: Dimitris Mavrakis now speaking at #LTE World Summit about how to reduce the cost of LTE network deployment
@oholin: @dmavrakis presenting now on cost per gigabyte: cost depends on network
@twehmeier: There is no single cost per gigabyte. Varies for every single operator and especially correlated to utilisation, says @dmavrakis
@twehmeier: Come on @dmavrakis, say it, our traffic-led business case found LTE wasn't necessary inUK until 2014. Non-traffic biz case is driver
@oholin: @dmavrakis very practical presentation so far with lots of case studies done by the network cost modelling tool
@kszaniawski: UK LTE vs HSPA case study - @dmavrakis arguing HSPA/HSPA+ will be perfectly capable of handling capacity in the UK in 2015 timeframe
@kszaniawski: Maintaining/installing new sites & backhaul are key determinants of cost in developing markets says @dmavrakis: see Egypt case study
@oholin: Judging by the number of listener's @dmavrakis' cost per gigabyte presentation is right on the money
@oholin: @dmavrakis LTE is just a piece of the puzzle

@kszaniawski: Wi-Fi offload business case is complex - 2015 US market case study shows even passive upgrades need management
@twehmeier: Lots of focus on selling LTE into enterprises at #LTEWS. Greater willingness to pay the premium versus consumers.
@twehmeier: Love the fact that Sprint have done a whole presentation on what they've done with “4G” WiMax so far without mentioning WiMax once!
@twehmeier: Sprint talked abt need to educate developers w/o saying who shd do it. Device OEMs talk it up in their plans, but are they doing it?
@kszaniawski: 4G? “When I use a word” Humpty Dumpty said, “it means just what I choose it to mean, neither more nor less”
@dmavrakis: Sprint saying that network and spectrum needs to be decoupled and networks need to move to software based platforms. #LTEWS
@dmavrakis: Sprint's vision is nice, but cognitive radio is nowhere near market and even today's software upgradable radios have limitations.
@twehmeier: So much spectrum chat. LTE WS 2010 just called-it wants its PPT decks back. Or, spectrum issues got much worse in last 12 months.

@twehmeier: 10 pres down+not a single traffic/revenue scissor slide! What refreshing change! Well done Sabah for laying down law

@oholin: Sad that some vendors still think that LTE is the whole picture and not just a piece in the puzzle

@twehmeier: Who coined hetnet term then? It is everywhere. Spoke to guy at Qualcomm who claimed it from early days of UMB. Any other claims?
@dmavrakis: @twehmeier I think someone from Huawei is claiming it also.

@oholin: Can't believe the term killer app also still in use

@twehmeier: Bengt of Northstream talking sense as always and handing out a lesson in spectrum and licensing history. Audience listening intently
@dmavrakis: Bengt Nordstrom at #LTEWS: Mobile devices take normally 2-3 years to mature to a market ready version. Same with LTE devices.

@zahidtg: Tim Hubbard, BT Wholesale: In January 10 million devices downloaded catchup TV in UK
@gabeuk: Good presentations from Jaime Lluch, Telefonica (small cells) and Tim Hubbard, BT (IP VPNs for mobile operators)
@zahidtg: I heard it many times today in #LTEWS where operators are talking about Coverage Layer (low freq) and capacity layer (high freq)
@zahidtg: #SON will help in Interference management in HetNet's. #LTEWS - I think SON can only help in limited ways for interference management.

@disruptivedean: Can you really call an HSPA subscriber network-limited to 512kbit/s max speed "mobile broadband" subscriber? Don't believe all the stats!
@zahidtg: Trying to get my head around. Panelists said that in Hetnet's, interference problems are less than no Hetnet's. Right?

18th May

@kingcharles1666: Day 2 of the #ltews (@ LTE World Summit)

@disruptivedean: Future of Voice roundtable at #LTEWS went well - thanks to all attendees for good discussion. I'm also doing a Spotlight pres at 11am
@disruptivedean: Interesting debate about dual-radio GSM+LTE phones as LTE voice easy option. Not popular with some network vendors. ZTE is advocate though
@disruptivedean: Lot of debate on Circuit Switched Fallback for LTE Voice. How long are call setup times in real world? User experience for dropped 4G data?

@kingcharles1666: According to vodafone speed is a killer app
@twehmeier: Vodafone Germany has 20,000 customers on its LTE network
@disruptivedean: Vodafone Germany has 20k customers for fixed-wireless LTE800 services. ADSL replacement & also offering own NGN fixed-style VoIP.
@oholin: The "number 1" operator has not got the number 1 presentation!
@oholin: Vodafone and Telefonica say: we want multi-mode devices as soon as possible
@Gabeuk: Few good stats on LTE performance in *loaded* cells from Vodafone Germany CTO at #LTEWS

@disruptivedean: Yota speaker at #LTEWS . Prime example of #happypipe operator. More than 1m users, now transitioning to shared wholesaler for 4 Russian MNOs
@disruptivedean: Yota believes future of telecom is for shared infra owners abstracted from retail operators & app/svcs = Under the Floor players UTF+MNO+OTT
@kszaniawski: Yegor Ivanov of Yota, Russia: justifying wholesale model he says it's no secret that many big operators not yet ready to deploy #LTE
@oholin: Yota - the competitive advantage is how you serve your customer! Let's Yota!

@disruptivedean: No recognition at #LTEWS that phone calls will need varying tiers. Call to important client should have higher QoS vs casual call to friend
@tmgb: PRT @disruptivedean: Call to client should have higher QoS vs call to friend (implies you care about the money more than about your mates?)
@AjitJaokar: @disruptivedean #LTEWS too many people in telco ind cannot ans why people make calls!
@AjitJaokar: @disruptivedean #LTEWS voice going from quality of service to quality of experience
@AjitJaokar: @disruptivedean #LTEWS humans don't interact in 'sessions'
@tmgb: @AjitJaokar agree QoS v QoE - QoS = call gets established and I can hear what is being said, everything else is QoE (@disruptivedean)

@dmavrakis: Getting ready for keynotes at #LTEWS, starting with Barclays capital discussing whether there are too many operators in market
@dmavrakis: LightSquared EVP of ecosystem development is on stage now to present about their disruptive business model
@dmavrakis: LightSquared operates 3 satellites today and will be operating 4 next year when they launch, covering 100% of North America
@dmavrakis: LightSquared will focus on network and service layer while partners on the rest: customer/device mgmt, marketing, distribution
@dukeq8: What an excellent speach ! Wholesale market for LTE is a sheft in the industry!
@twehmeier: LightSquared says will never have over 500 employees. Most telcos count employees in 1000s, many in 10s of 1000s, some 100s of 1000s
@dmavrakis: LightSquared shared surprisingly little information, nothing new.

@twehmeier: MetroPCS has 200,000 LTE connections. Quick back of envelope estimate - I'd say first million subs on LTE globally has been passed
@dmavrakis: @twehmeier Do you think biggest driver of LTE connections today is leap-froging from CDMA to LTE? Including Leap and Verizon
@twehmeier: @dmavrakis Yes, because they were happy to go dual antenna 4 voice and therefore haven't dithered launching handsets. Handsets drive scale

@twehmeier: Did i hear that right? Samsung to launch VoLTE handset in 4Q11 with Cellular South in the US?
@dmavrakis: He confirmed this in Q&A RT @twehmeier: Did i hear that right? Samsung to launch VoLTE handset in 4Q11 with Cellular South in the US?
@dmavrakis: Samsung speaker touting device expertise to drive networks business, including inhouse IOT.
@tmgb: @dmavrakis @twehmeier Samsung/Cell South - oh happiness :-)

@dmavrakis: Patent pool Sisvel's founder will present now on LTE IPR. Lets see if things will be less complicated that 3G
@dmavrakis: Theoretically, LTE IPR will be much more complicated when there will be 3 competing patent pools and bilateral agreements
@dmavrakis: Sisvel's founder mentioned that some meeting is taking place during the #LTEWS to create a new patent pool for LTE.
@dmavrakis: Sisvel: 4000 IPR declarations for LTE/SAE. Interdigital, Qualcomm, Nokia and Samsung leaders (ETSI database March 2011)
@dmavrakis: Interesting statistics for GSM by Sisvel: 561 families declared as essential but only 158 granted as essential
@twehmeier: Interesting analysis patents role in pushing up costs of DVD players. Retail cost in China =$20. Sums of patent licences if all paid=$19.
@kingcharles1666: LOL RT @dukeq8: I'm confused now that I choosed the wrong industry! I should be a lawyer.

@dmavrakis: CTO of CSL Christian Daigneault on stage now to discuss their LTE/DC-HSPA network
@AjitJaokar: #ltews mobile penetration in HK is 160 perc. @tomiahonen is in the right place :)
@twehmeier: Wonder how many infrastructure contracts sweetened by clauses that ops must do conf circuit singing & dancing about vendor's network?
@dmavrakis: CSL claims Hong Kong is highest market for tablets with 17% market penetration. Sweetspot for LTE?
@twehmeier: CSL HK are basically deploying in way many ops wd in ideal situation. 3G@ 2100+900, LTE @ 2.6+1800. Missing piece is LTE sub-1GHz.
@dmavrakis: CSL has a USB modem that supports everything from GSM 900 to DC-HSPA to LTE at 2.6GHz

@kingcharles1666: Very conservative view from FT orange on VoLTE and migration.

@oholin: 5 ITM analysts discussing LTE drivers and trends at the analyst spotlight
@oholin: @dmavrakis: "technology evolves with operator business models"

Please feel free to add comments and suggestions.

Tuesday 10 May 2011

Advanced IP Interconnection of Services (IPXS) in 3GPP Rel-11

The following is edited from the 3GPP documents:

IP is being introduced in both fixed and mobile networks as a more cost-effective alternative to circuit switched technology in the legacy PSTN/PLMN, as well as the underpinning transport for delivering IMS based multi-media services.

In order to ensure carrier grade end to end performance, appropriate interconnect solutions are required to support communications between users connected to different networks. There are currently a number of initiatives underway outside 3GPP addressing IP Interconnection of services scenarios and commercial models to achieve this; for example, the GSM Association has developed the IPX (IP Packet Exchange). Also, ETSI has recently defined requirements and use case scenarios for IP Interconnection of services. These initiatives require the use of appropriate technical solutions and corresponding technical standards, some of which are already available and others which will require development in 3GPP.

Moreover, new models of interconnection may emerge in the market where Network Operators expose network capabilities to 3rd party Application Providers including user plane connectivity for the media related to the service.

The main objective of IPXS is thus:
To specify the technical requirements for carrier grade inter-operator IP Interconnection of Services for the support of Multimedia services provided by IMS and for legacy voice PTSN/PLMN services transported over IP infrastructure (e.g. VoIP).

These technical requirements should cover the new interconnect models developed by GSMA (i.e. the IPX interconnect model) and take into account interconnect models between national operators (including transit functionality) and peering based business trunking.

Any new requirements identified should not overlap with requirements already defined by other bodies (e.g. GSMA, ETSI TISPAN). Specifically the work will cover:
Service level aspects for direct IP inter-connection between Operators, service level aspects for national transit IP interconnect and service level aspects for next generation corporate network IP interconnect (peer-to-peer business trunking).
Service layer aspects for interconnection of voice services (e.g. toll-free, premium rate and emergency calls).
Service level aspects for IP Interconnection (service control and user plane aspects) between Operators and 3rd party Application Providers.

To ensure that requirements are identified for the Stage 2 & 3 work to identify relevant existing specifications, initiate enhancements and the development of the new specifications as necessary.

The following is a related presentation on Release-8 II-NNI with an introduction to Rel-9 and Rel-10 features.

The 3GPP references can be seen from the presentation above.

European Commission conducted a study on this topic back in 2008 and produced a lengthy report on this. Since the report is 187 pages long, you can also read the executive summary to learn about the direction in technical, economic and public policy.

Sunday 8 May 2011

What is '4G' ?

One of the most popular posts is Dilbert's definition of '4G'. So I decided to go back and see how and if the definition of '4G' has changed over the time.

Back in 2008, '4G' was more of WiMAX. Here is a video from that time:

2009 video of Cisco where they are pushing for 4G = IP :

2010 video that says 4G = fast :

Now from 2011, where consumers in US are being asked What '4G' means: News

And finally my '4G' FAQ from 2006 that is seriously outdated and needs updating :)

Wednesday 4 May 2011

New Security Algorithms in Release-11

I did mention in my earlier blog post about the new algorithm for 3GPP LTE-A Security. The good news is that this would be out hopefully in time for the Release-11.

The following from 3GPP Docs:

The current 3GPP specifications for LTE/SAE security support a flexible algorithm negotiation mechanism. There could be sixteen algorithms at most to support LTE/SAE confidentiality and integrity protection. In current phase, 3GPP defines that there are two algorithms used in EPS security, i.e. SNOW 3G and AES. The remaining values have been reserved for future use. So it is technically feasible for supporting new algorithm for LTE/SAE ciphering and integrity protection.

Different nations will have different policies for algorithm usage of communication system. The current defined EPS algorithm may not be used in some nations according to strict policies which depend on nation’s security laws. Meanwhile, operators shall implement their networks depending on national communication policies. To introduce a new algorithm for EPS security will give operators more alternatives to decide in order to obey national requirements.

Picture: Zu Chongzi
Picture Source: Wikipedia

Some work has been done to adapt LTE security to national requirements about cryptography of LTE/SAE system, i.e. designing a new algorithm of EPS security, which is named ZUC (i.e. Zu Chongzhi, a famous Chinese scientist name in history). Certainly the new algorithm should be fundamentally different from SNOW 3G and AES, so that an attack on one algorithm is very unlikely to translate into an attack on the other.

The objective of this work item is to standardise a new algorithm in EPS. This will include the following tasks:
To develop new algorithms for confidentiality and integrity protection for E-UTRAN
To enable operators to quickly start to support the new algorithm
Not to introduce any obstacle for R8 roaming UE

The following issues should at least be handled in the WI:
Agree requirement specification with ETSI SAGE for development of new algorithms
Delivery of algorithm specification, test data and design and evaluation reports

The algorithm is provided for 3GPP usage on royalty-free basis.

The algorithm shall undergo a sequential three-stage evaluation process involving first ETSI SAGE, then selected teams of cryptanalysts from academia and finally the general public.

The documents related to the EEA3 and EIA3 algorithm could be downloaded from here.

If you are new to LTE Security, the following can be used as starting point: