We had quite a few interesting discussions in the Small Cells Global Congress, Operator Mindshare session. Here are some of the things that were discussed:
Licensed v/s Unlicensed deployments:
Many operators are now deploying WiFi in the unlicensed spectrum. This can help in the short term to alleviate the capacity problems but as more and more of this unlicensed spectrum nodes get deployed, they create interference between each other and make them unusable for anyone. An example was provided about Tokyo where in some areas, too many free WiFi hotspots means its unusable for anyone. One solution is to have one operator do all the logistics for the deployment and other operators can pay to use the service. Who (operator) would be the first one to go through the process of deploying everything first? Everyone would prefer wait and watch approach.
Providing free WiFi:
The consensus was that the free WiFi provided by operators don't give any additional benefit to them and there isn't much of a business case.
Consumer awareness for residential Femtocells:
Globally, not much effort is being done by the operator to make the end users aware of residential Femtocells and this is hampering the take-up A point was made about when Vodafone launched their product, Vodafone Access Gateway (VAG), it was perceived as negative thing because the ads show that if the coverage was poor you can install this to improve coverage. From a users perspective, it showed that the network had poor coverage. Still consumer awareness is important, how to do it?
Placement of Small Cells:
Where should the public small cells (metrocells) be placed. The Biggest challenges are:
* Site Acquisition is the biggest problem. - This is a bigger problem if lap posts are sought to deploy on public locations
* Power - Lamp posts are centrally switched off, so small cells on laamp posts may need alternative sources
* Power meter if used in a shared location
* Bullet proof (especially in the US)
* Backhaul - especially is non line of sight case.
* Health concerns (if visible)
* Visual appearance
Operators should be clearer in what they want. Right now the vendors are pushing the solutions that operators not necessarily need and not giving what the operators want. The Backhaul should be more flexible and future proof. It should be able to cater for upcoming technologies like Carrier Aggregation, CoMP, etc.
Shared v/s Dedicated carrier for 3G Small Cells:
Dedicated carrier is ideal but is not easily possible for most operators. When shared carrier is used it causes interference and handovers are not easy.
Interoperability in the new hardware equipment for support of small cells:
Certain vendors are still not creating the the networks that can interwork with other vendors equipment. As we are moving towards LTE, this seems to be a much bigger problem. Sprint for example has 3 completely different networks in the US with no interoperability between them. Standards are not helping either as they do not dictate implementation.
Some Interesting discussions on Case studies, Business Cases, etc.
* Deployed residential Femtocells
* Deployed for coverage purpose
* Dont have handover capability yet
* Want to be able to deploy Microcells/Small Cells on Highways, around 1-2Km radius
* Their typical Microcells use 40W output power
* The cost of deployment if Macro using cabinet, antenna, etc is roughly 100K per site.
Telefonica, O2 trials in UK
* To get access to council lamp posts, it was required that the bidder offer free WiFi
* O2 set a high bar by paying lot of money to the councils in London, but this is not a sustainable model
A Business case for carrier neutral WiFi on light pole in Lima, Peru
* Each light pole can have 3 different locations
* The retail business case is to get the user to usse the offering and maybe offer the operator services, tempting to move to this operator from current one
* There can be a wholesale case of selling the WiFi capacity in bulk to companies, organisations
Some interesting statistics thrown up:
* WiFi cell radius is 30m in South America
* 83% of people in US think that operators should provide free WiFi because of lousy coverage of the mobile network.
* The first 4000 customers of a WiMax operator were using an average of 750 MB per day, 22.5GB per month.
* Some fixed Internet operators are now thinking of putting a cap on unlimited offering at 350GB per month.
There were no consensus and conclusions for many items so feel free to write your opinion in the comments.
I have seen on Twitter some people think that HetNets (Heterogeneous Networks) is just a new name for the Hierarchical Cell Structures (HCS). The main difference between then is that while HCS requires all layers to have different frequencies, HetNets can use the same or the different frequency. In case the same frequency is used, there needs to be a way to manage interference between the different layers. In fact the term 'layers' is hardly used with HetNets as there is nothing strictly hierarchical with different types of cells that co-exist in a HetNet. Typically a HetNet comprises of Macro cells, Micro/Pico cells, other Small Cells (including Femtocells) and WiFi as well (if used to offload traffic).
This recent whitepaper from 4G Americas is an excellent source to understand more about HetNets
The above Dilbert strip is a reality for a lot of us, everyday. There are many bugs in each product and we only have a limited amount of time and resources that can be dedicated to fixing bugs. The '5 stages of bug fixing' was presented in one of the Cambridge Wireless events not long ago. I am not sure if allowed to mention by whom, so here it is anonymously.
The five stages of Bug Fixing
If you think about it from the Tester's point of view, this is what he prays for everyday. See the old post here.
Couple of days back, I posted some videos that show technology advancements for the mobile phones. Here is a presentation by Peter Whale from Qualcomm in a recent Cambridge Wireless event about how Sensors and Context-engine will make the future devices much more intelligent then they already are.
Couple of months back, there was this Dilbert strip on the big data.
Apparently, Social networks, M2M devices and many other sources of data, keeps on generating data all the time. This data can provide us with a lot of useful information if proper analytics can be done on it. This is a real challenge in guess. There will also be security and privacy implications that may decide how and what can be used and by whom.
Here is a simple introductory video by Intel explaining what Big Data is:
M2M is going to be big. With the promise of 50 Billion devices by 2020, the networks are already worried about the overloading due to signalling by millions of devices occurring at any given time. To counter this, they have been working on avoiding overloading of the network for quite some time as blogged about here.
The feature to avoid this overload is known as Extended Access Barring (EAB). For E-UTRAN, in Rel-10, a partial solution was implemented and a much better solution has been implemented in Rel-11. For GERAN a solution was implemented in Rel-10. The following presentation gives a high level overview of EAB for E-UTRAN and GERAN.
In Rel-11, a new System Information Block (SIB 14) has been added that is used specifically for EAB. Whereas in Rel-10, the UE would still send the RRCConnectionRequest, in Rel-11, the UE does not even need to do that, thereby congesting the Random Access messages.
The following is from RRC 36.331 (2012-09)
Here is my attempt to explain the difference in overload control mechanism in Rel-8, Rel-10 and Rel-11. Please note that not actual message names are used.
As usual, happy to receive feedback, comments, suggestions, etc.
The following 14 MTC Features have been identified during the 3GPP Release-10 timelines:
Packet Switched (PS) Only
Small Data Transmissions
Mobile Originated Only
Infrequent Mobile Terminated
Priority Alarm Message (PAM)
Location Specific Trigger
Network Provided Destination for Uplink Data
Group Based MTC Features
In Rel 10, 3GPP will focus on the general functionality required to support these features:
Overload control (Radio Network Congestion use case, Signalling Network Congestion use case and Core Network Congestion use case)
The following specifications are associated with the MTC work
Spec - Specifications associated with or affected by MTC work 22.011 - Service accessibility 22.368 - Service requirements for Machine-Type Communications (MTC); Stage 1 23.008 - Organization of subscriber data 23.012 - Location management procedures 23.060 - General Packet Radio Service (GPRS); Service description; Stage 2 23.122 - Non-Access-Stratum (NAS) functions related to Mobile Station (MS) in idle mode 23.203 - Policy and charging control architecture 23.401 - General Packet Radio Service (GPRS) enhancements for Evolved Universal Terrestrial Radio Access Network (E-UTRAN) access 23.402 - Architecture enhancements for non-3GPP accesses 23.888 - System improvements for Machine-Type Communications (MTC) 24.008 - Mobile radio interface Layer 3 specification; Core network protocols; Stage 3 24.301 - Non-Access-Stratum (NAS) protocol for Evolved Packet System (EPS); Stage 3 24.368 - Non-Access Stratum (NAS) configuration Management Object (MO) 25.331 - Radio Resource Control (RRC); Protocol specification 29.002 - Mobile Application Part (MAP) specification 29.018 - General Packet Radio Service (GPRS); Serving GPRS Support Node (SGSN) - Visitors Location Register (VLR); Gs interface layer 3 specification 29.060 - General Packet Radio Service (GPRS); GPRS Tunnelling Protocol (GTP) across the Gn and Gp interface 29.118 - Mobility Management Entity (MME) - Visitor Location Register (VLR) SGs interface specification 29.274 - 3GPP Evolved Packet System (EPS); Evolved General Packet Radio Service (GPRS) Tunnelling Protocol for Control plane (GTPv2-C); Stage 3 29.275 - Proxy Mobile IPv6 (PMIPv6) based Mobility and Tunnelling protocols; Stage 3 29.282 - Mobile IPv6 vendor specific option format and usage within 3GPP 31.102 - Characteristics of the Universal Subscriber Identity Module (USIM) application 33.868 - Security aspects of Machine-Type Communications 36.331 - Evolved Universal Terrestrial Radio Access (E-UTRA); Radio Resource Control (RRC); Protocol specification 37.868 - RAN Improvements for Machine-type Communications 43.868 - GERAN Improvements for Machine-type Communications 44.018 - Mobile radio interface layer 3 specification; Radio Resource Control (RRC) protocol 44.060 - General Packet Radio Service (GPRS); Mobile Station (MS) - Base Station System (BSS) interface; Radio Link Control / Medium Access Control (RLC/MAC) protocol 45.002 - Multiplexing and multiple access on the radio path
Here are couple of presentations I have extracted the above information from:
Last year I blogged about Femtocell as a Service (FaaS) but since we are talking in terms of Small Cell, we can use the term SCaaS that can form a superset to FaaS. This is from a presentation by Prof. Simon Saunders, in the Cambridge Wireless event.
We had over a hundred delegates in this event and as one of the speakers admitted privately, they were expecting to see around a dozen people and were a bit overwhelmed by the number and caliber of the people. The delegates included small cell vendors, chipset vendors, test & measurement vendors, operators, industry analysts, regulators, etc. It was a lovely day to be in Cambridge with the sun shining the whole of morning and the afternoon to show us the best of the Downing college.
The event was kicked off by Prof. Simon Saunders, formerly the founding chairman of Small Cell Forum who talked about the long journey the small cells (or Femtocells as they were then known) have gone through, the benefits and the road ahead.
This was followed by a talk from Julius Robson of CBNL, who was also representing NGMN. The focus of the talk was on backhaul.
Nick Johnson, CTO of ip.access was the next speaker who started his presentation with humorous note. His presentation was titled "Building the World’s Largest Residential Small-Cell network" but as he said he was very tempted to change the title to “How to Screw Up the World’s Largest Residential Small Cell Deployment”. His talk had lots of real life examples on where and how things can go wrong and how to make sure they dont. If they do, what lessons should be learnt. Some of these problems have been faced by me too in various test scenarios. It was a very interesting talk.
After the break we heard a presentation from Steve Brown of Telefonica O2 UK. The talk was a bit familiar for me (and my blog readers) as I have already blogged on similar information in the past. It was though new information for the audience and could see that they were enjoying this information. A lot of questions were asked after the presentation and also in the panel discussion at the end. There is some interesting new information that I will blog later on.
The final talk was by Iris Barcia of Keima who talked about "Small Cell Network Design".
Finally we had a panel discussion with lots of interesting questions. Once the discussions finished there were people discussing and debating issues among themselves for a long time. I am looking forward to the next event in January in London on the topic "Lets get real!" where we are hoping to be able to hear from some more operators/vendors on the deployment and rollout issues. More details available on the Cambridge Wireless page here.
Some of the interesting findings from the conference include:
TD-LTE is gaining momentum, and its beyond WiMAX operators and China mobile, many APAC operators are considering it for unpaired spectrum and to efficiently meet the asymmetric capacity requirements of mobile broadband which is mainly download
Software defined radio and self-organizing networks are proving critical to manage operational costs
Single RAN is proving the best way to manage network performance
Signaling is in a mess - what is the good of standards when it creates such a mess?
IMS gaps continue - what is the good of standards when it doesn't meet basic migration needs?
The SS7 guys have reinvented themselves as the Diameter guys
Business model innovation - LTE is not just for mobile devices, LTE is for quad play and an interesting array of business applications
The 3G network of many operators is congested - forcing the move to LTE
CSFB (Circuit Switched Fall Back) works
VoLTE testing / roaming / network issues remain - given voice remains by revenue the core service, our industry should be ashamed we're having so many problems with VoLTE
A belief on OTT partnering, but not quantification on the OTT's willingness to pay for QoS (Quality of Service)
Many operators have a question mark on the use of WiFi off-load - its not a technology issue rather one of economics and customer experience, LTE-A and small cells in hotspots appears to be the focus.
Briefly reviewing the slides shown below:
LTE Data Points
96 Commercial LTE deployments mainly in the 1.8 and 2.8GHz bands
APAC has 40% of LTE subscribers, likely to be the high growth region
Drivers for LTE: Throughput, efficiency and low latency
TD-LTE: 12 commercial deployments, 24 contracts and 53 Trials
Streaming video dominates traffic on handheld devices, with YouTube being the top traffic generator at 27% of peak traffic
South Korea Data Explosion
South Korea has seen OTT explode, Kakao Talk 51 mins of usage per day
20 times smartphone growth in 2 years (28M in June 2012, 53% penetration)
60 times mobile data growth to 37TB per month in 2 years, 32% is from LTE devices
LTE subs use 2.9GB per month compared to 3G sub on average use 1.2GB
LTE subs reached 10M, 141% monthly growth
Customer drive for LTE is speed (37%) and latest device (31%)
Challenge Jan 2010 and Jan 2012 ARPU fallen from $48-$35 while data use risen from 180MB to 992MB
Focus beyond voice, messaging and data into VAS: virtual goods (Korean thing), ICT (Information and Communication Technology) and cloud services / solutions (focus on enterprise)
HK CSL Migration to LTE
3G is congested, LTE is not
Key is LTE devices available, unlike the early 3G days
Migrating customers away from unlimited plans to family and shared plans that deliver value
LTE sub uses 2-5 times the data of 3G subs
Average speed seen is 20 Mbps
Using Software Defined Radio, Single vendor RAN, Self-Organizing Networks
Migration to LTE-A, small cells and WiFi where appropriate
Starhub's migration to LTE (they launched LTE at the event)
50% of voice traffic is still on 2G
Using AMR to re-farm 2G spectrum to LTE
Site access is critical - drive to software defined radio to avoid site visits
NTT DoCoMo's VoLTE Evolution
70% devices in portfolio are now LTE
All smartphones support CSFB
Drive to VoLTE is simply to switch off 3G voice (2G already off)
BUT IMS has missing functionality / standards - migration from 3G to VoLTE is not easy - example of failing in standards on basic issues
Yes: Example of innovative converged 4G operator in an developing market that uses web principles for service delivery
Role of Mobile Identity in BYOD (Bring Your Own Device)
BYOD is as significant a trend if APAC as any other market
Provides a nice review of the approaches in managing BYOD
LTE Quad-Play in Emerging Markets: TD-LTE case study
Smartphone growth implications: Review of the signaling problem and mitigation strategies across 3G and LTE. Highlights challenge current standards process
Even though I have known tracking area for a long time, the other day I struggled to explain exactly what it is. I found a good explanation in this new book 'An Introduction to LTE: LTE, LTE-Advanced, SAE and 4G Mobile Communications By Christopher Cox'. An extract from the book and Google embed is as follows:
The EPC is divided into three different types of geographical area, which are illustrated in Figure 2.6. (see Embed below).
An MME pool area is an area through which the mobile can move without a change of serving MME. Every pool area is controlled by one or more MMEs, while every base station is connected to all the MMEs in a pool area by means of the S1-MME interface. Pool areas can also overlap. Typically, a network operator might configure a pool area to cover a large region of the network such as a major city and might add MMEs to the pool as the signalling load in that city increases.
Similarly, an S-GW service area is an area served by one or more serving gateways, through which the mobile can move without a change of serving gateway. Every base station is connected to all the serving gateways in a service area by means of the S1-U interface. S-GW service areas do not necessarily correspond to MME pool areas.
MME pool areas and S-GW service areas are both made from smaller, non-overlapping units known as tracking areas (TAs). These are used to track the locations of mobiles that are on standby and are similar to the location and routing areas from UMTS and GSM.