There are as many companies active on the femtocell backhaul integration with IP core as there are Femto cell manufacturers, with Kineto, Sonus and Tatara systems topping the list.
Sonus has recently announced partnerships with three femtocell manufacturers: ip.access, 3way and RadioFrame.
My worry is that some companies in the IP core space are re-inventing the wheel so to speak which threatens the main selling story for femtocells: the low introduction cost and the potential savings to operators. I'm no core expert, but I would guess that the existing framework(s) are sufficient to enable Femtocell integration with core networks without any major investment or development effort.
5 Jul 2007
Femtocell Forecasts
Cellular News published an interesting short article suggesting that femtocell shipments will reach 36million in 2012, which in my view a rather conservative estimate. The estimate is contingent on a couple of big operators tidying up their messy business models. If operators grasp the big potential offered by the femto concept, then one can easily predict an estimate closer to 100 million plus. Nevertheless I perfectly understand why other experts do not see it this way.
An interesting quote in the article mentioned above:
"...no matter how attractive the service offerings a femtocell-based solution can bring, the sheer savings from backhaul and energy costs could equate to over $70 billion by 2012, outshining service revenues and providing enough financial incentive for carriers to actively support this solution".
Obviously, the operators that will gain the most are the ones that do not have yet a fully deployed 3G infrastructure.
An interesting quote in the article mentioned above:
"...no matter how attractive the service offerings a femtocell-based solution can bring, the sheer savings from backhaul and energy costs could equate to over $70 billion by 2012, outshining service revenues and providing enough financial incentive for carriers to actively support this solution".
Obviously, the operators that will gain the most are the ones that do not have yet a fully deployed 3G infrastructure.
20 Jun 2007
Femtocell Challenges
In a previous post I talked about the potential benefits of femtocells to the operators as well as consumers. In this post I'd like to investigate the potential hurdles in the road to viable femtocell solutions.
Looking at various analysts remarks I narrowed the list of potential show stoppers to the following items, all of which are not unresolvable, but one has to be aware of:
Cost:
This is the obvious one. There is no doubt that we are well into the cost reduction era. This is by far the number one item on every operator's agenda. The bottom line is that operators want to see a nice dollar savings associated with Femto cells, which increases pressure on femto cell development firms to reduce their price tag. Traditionally, pricing is based on a BOM plus margins. An alternative way for development firms is to price based on potential benefits/savings. My firm has recently been engaged in a similar pricing activity with a potential investor.
The cost challange means that there is a limit to the amount of effort and features developers will bundle in the femto cell box. Deciding what is "sufficient" feature list is a difficult call (check the technology life cycle model proposed by D. Normann in one of my previous posts)
Radio Planning and interference management:
Unlike Wi Fi, the Femto cell is likely to use the licensed spectrum e.g. UMTS band. Although it is my view that the femto network is better have its own carrier, many clients would like the femto cell concept to work on the same carrier frequency used by existing infrastructure be it a macro, micro or pico layer.
Femtocells are expected to coexist with the rest of the network by automatically configuring themselves. Operator or user intervention is undesirable...a femto cell box has to be fully plug and play without the need to setup,configure or optimise it in anyway. Therefore lots of effort is being spent on femto cell radio management features e.g. choosing the right carrier, automatic setting of power levels, protecting other infrastructure from harmful interference, handoff ...etc .
Core Network Integration:
I touched upon this in a previous post (Femtocell Architecture).
In a nutshell, a legacy architecture is too old fashioned, while on the other hand a full flat IP architecture with an IMS core is an expensive undertaking with no sign of becoming available soon.
Interim solutions relying on some kind of concentrator seem to attract some attention. I can't really comment much on this and I'll leave it to the core experts to make their minds.
With converged solutions such as femto cells there is also an increased expectation of a higher level of network intelligence and features at the edges of the network (as opposed to the traditional model of core intelligence).
End user perception:
The consumer will (and has to) ask: "what's in it for me". At the moment most of the promises revolve around cheaper calls as well as using the often over hyped word: "convergence". The fact is that most people do not know what this word means! The operators will find it challenging to convince the consumer that the femtocell is the way to go.
In addition, there is the issue of backward compatibility in terms of usage. Many people have Internet access at home and a big proportion use WiFi routers to have wireless coverage at home. Is the Femtocell going to replace that? if so, then there is a need to have a combined WiFi/Femto router to enable legacy equipment to access the Internet. Alternatively people will end up with a stack of routers, from different suppliers.... hardly a convergent solution!.
Therefore the extent of device convergence will play a significant role in any operator's femtocell strategy.
There are also health concerns about exposure to increased levels of RF waves.
Looking at various analysts remarks I narrowed the list of potential show stoppers to the following items, all of which are not unresolvable, but one has to be aware of:
Cost:
This is the obvious one. There is no doubt that we are well into the cost reduction era. This is by far the number one item on every operator's agenda. The bottom line is that operators want to see a nice dollar savings associated with Femto cells, which increases pressure on femto cell development firms to reduce their price tag. Traditionally, pricing is based on a BOM plus margins. An alternative way for development firms is to price based on potential benefits/savings. My firm has recently been engaged in a similar pricing activity with a potential investor.
The cost challange means that there is a limit to the amount of effort and features developers will bundle in the femto cell box. Deciding what is "sufficient" feature list is a difficult call (check the technology life cycle model proposed by D. Normann in one of my previous posts)
Radio Planning and interference management:
Unlike Wi Fi, the Femto cell is likely to use the licensed spectrum e.g. UMTS band. Although it is my view that the femto network is better have its own carrier, many clients would like the femto cell concept to work on the same carrier frequency used by existing infrastructure be it a macro, micro or pico layer.
Femtocells are expected to coexist with the rest of the network by automatically configuring themselves. Operator or user intervention is undesirable...a femto cell box has to be fully plug and play without the need to setup,configure or optimise it in anyway. Therefore lots of effort is being spent on femto cell radio management features e.g. choosing the right carrier, automatic setting of power levels, protecting other infrastructure from harmful interference, handoff ...etc .
Core Network Integration:
I touched upon this in a previous post (Femtocell Architecture).
In a nutshell, a legacy architecture is too old fashioned, while on the other hand a full flat IP architecture with an IMS core is an expensive undertaking with no sign of becoming available soon.
Interim solutions relying on some kind of concentrator seem to attract some attention. I can't really comment much on this and I'll leave it to the core experts to make their minds.
With converged solutions such as femto cells there is also an increased expectation of a higher level of network intelligence and features at the edges of the network (as opposed to the traditional model of core intelligence).
End user perception:
The consumer will (and has to) ask: "what's in it for me". At the moment most of the promises revolve around cheaper calls as well as using the often over hyped word: "convergence". The fact is that most people do not know what this word means! The operators will find it challenging to convince the consumer that the femtocell is the way to go.
In addition, there is the issue of backward compatibility in terms of usage. Many people have Internet access at home and a big proportion use WiFi routers to have wireless coverage at home. Is the Femtocell going to replace that? if so, then there is a need to have a combined WiFi/Femto router to enable legacy equipment to access the Internet. Alternatively people will end up with a stack of routers, from different suppliers.... hardly a convergent solution!.
Therefore the extent of device convergence will play a significant role in any operator's femtocell strategy.
There are also health concerns about exposure to increased levels of RF waves.
13 Jun 2007
Femto cell Architecture
The business case for indoor femto cells has gained increasing traction in the last few weeks, and lots of news comments about various femto cell development companies. For example, check here, here and here.
Although the femto cell concept is quickly gaining grounds towards complete and operational products, there is still some misunderstanding on how the overall network architecture will look like. I'll try to address this issue in this post.
A femto cell router is a small device, the size of any Wi-Fi router, which is in effect a miniature base station. The radio is a standard based radio such as UMTS/HSPA which the operator will likely require a license to operate on. The router connects to a DSL line. The idea is to enable the subscriber to make and receive mobile calls indoors, with low signal levels which has a number of benefits to the operator:
- Hopefully accelerate fixed line substitution.
- Reduce the cost of building a full macro layer network.
- Lock subscriber in to the operator and reduce the likelihood of churn.
- Provide a viable medium for content distribution (the ugly walled garden paradigm)
It is not my intention here to argue for or against these benefits. I will dedicate a full post on the business case for the femto cell and the potential cons sometime soon. So let us stay on the technology side and try to investigate how a UMTS based femto cell routers will be integrated with the operator's core network.
Basically, there are a number of ways, the first of which is the conventional hierarchy using an RNC . Just like any node-b is connected back to an RNC which is then in turn is connected to the core network, the femtocell routers can be treated like individual node-bs and connect to RNCs. This may appeal to big manufacturers who already have substantial deployments and many RNC on the ground. The downside to this approach is the limited chances of inter-operable devices if the operators chooses to diversify their suppliers. Although the Iub interface (base station to RNC) is standardised, the reality is most implementations are proprietary. Typically operators don't like to put all their eggs in one basket and would prefer to get solutions from various suppliers. The other downside of this approach is that most available RNC solutions are geared towards Macro/Micro type of deployments. In other words they are built in order to support relatively small number of cells with a huge number of subscribers in each cell. They don't scale very well to support Femto cell deployments with almost as many cells as there are subscribers, and a handful of subscribers per cell.
An alternative approach is to use UMA (or now called GAN). UMA was originally conceived to support dual mode cellular/WiFi-over-Internet type of connectivity. When the mobile is detected indoors within the range of pre-determined WiFi coverage, a UMA concentrator does the core network negotiation on behalf of the mobile, e.g. registering, and updating location ... etc. It is thought that a similar procedure can be used for femto cells. When the mobile is detected within the coverage of a femto cell, UMA kind of hand shaking takes place to ensure that: 1. the mobile is allowed to access the network at this particular femto cell, 2. the traffic between the mobile and the core network is forwarded accordingly. Using this method a UMA concentrator is required, so in a way, the solution is still hierarchical and it is certainly one of the criticisms of this method is that it is not a fully flat architecture. Kineto is one of the companies promoting this approach.
Yet another alternative approach is enabling femto cell connectivity through an IMS service. In this approach, the femto cell router talks SIP over the internet back to an IMS service which acts as a bridge between the femto layer and the rest of the network. Although this is considered the "flattest" approach, it is yet to understand how it will work in practice, the delay in working fully featured IMS platforms is one of the concerns of femto cell development companies.
Incidentally, my current clients are hedging their bets by supporting more than one approach and have developed partnerships with various companies in their respective fields to make the Femto architecture as flexible as possible.
Although the femto cell concept is quickly gaining grounds towards complete and operational products, there is still some misunderstanding on how the overall network architecture will look like. I'll try to address this issue in this post.
A femto cell router is a small device, the size of any Wi-Fi router, which is in effect a miniature base station. The radio is a standard based radio such as UMTS/HSPA which the operator will likely require a license to operate on. The router connects to a DSL line. The idea is to enable the subscriber to make and receive mobile calls indoors, with low signal levels which has a number of benefits to the operator:
- Hopefully accelerate fixed line substitution.
- Reduce the cost of building a full macro layer network.
- Lock subscriber in to the operator and reduce the likelihood of churn.
- Provide a viable medium for content distribution (the ugly walled garden paradigm)
It is not my intention here to argue for or against these benefits. I will dedicate a full post on the business case for the femto cell and the potential cons sometime soon. So let us stay on the technology side and try to investigate how a UMTS based femto cell routers will be integrated with the operator's core network.
Basically, there are a number of ways, the first of which is the conventional hierarchy using an RNC . Just like any node-b is connected back to an RNC which is then in turn is connected to the core network, the femtocell routers can be treated like individual node-bs and connect to RNCs. This may appeal to big manufacturers who already have substantial deployments and many RNC on the ground. The downside to this approach is the limited chances of inter-operable devices if the operators chooses to diversify their suppliers. Although the Iub interface (base station to RNC) is standardised, the reality is most implementations are proprietary. Typically operators don't like to put all their eggs in one basket and would prefer to get solutions from various suppliers. The other downside of this approach is that most available RNC solutions are geared towards Macro/Micro type of deployments. In other words they are built in order to support relatively small number of cells with a huge number of subscribers in each cell. They don't scale very well to support Femto cell deployments with almost as many cells as there are subscribers, and a handful of subscribers per cell.
An alternative approach is to use UMA (or now called GAN). UMA was originally conceived to support dual mode cellular/WiFi-over-Internet type of connectivity. When the mobile is detected indoors within the range of pre-determined WiFi coverage, a UMA concentrator does the core network negotiation on behalf of the mobile, e.g. registering, and updating location ... etc. It is thought that a similar procedure can be used for femto cells. When the mobile is detected within the coverage of a femto cell, UMA kind of hand shaking takes place to ensure that: 1. the mobile is allowed to access the network at this particular femto cell, 2. the traffic between the mobile and the core network is forwarded accordingly. Using this method a UMA concentrator is required, so in a way, the solution is still hierarchical and it is certainly one of the criticisms of this method is that it is not a fully flat architecture. Kineto is one of the companies promoting this approach.
Yet another alternative approach is enabling femto cell connectivity through an IMS service. In this approach, the femto cell router talks SIP over the internet back to an IMS service which acts as a bridge between the femto layer and the rest of the network. Although this is considered the "flattest" approach, it is yet to understand how it will work in practice, the delay in working fully featured IMS platforms is one of the concerns of femto cell development companies.
Incidentally, my current clients are hedging their bets by supporting more than one approach and have developed partnerships with various companies in their respective fields to make the Femto architecture as flexible as possible.
19 May 2007
Top 20 best selling handsets in UK
The list of top 20 selling handsets with contracts from Mobile Today shows the weakest performance of Motorola handsets this year.
In the top 20 there are: 7 SonyEricsson, 6 Nokia, 4 Samsung, 2 LG and only one Motorola made handset. The only Motorola handset in the list is the KRZR, ranking 14.
I'm not sure if analysts use this kind of market trends at all when they make investment decisions, but somebody has to question the reasons behind such a dismal performance.
Motorola has failed to find a followup to the best selling RAZR phone. The company was a victim of its own success with the RAZR by thinking that they found the right formula for a best selling handsets. Instead of developing new handset ideas with new concepts, they kept on releasing models that essentially copy the RAZR concept over and over.
I remember vividly how my casual suggestion of developing a new mobile platform that enables handset customisation was met with ridicule at Motorola.
Since the news that the ex head of mobile devices Ron Garriques is to join Dell came out a couple of months ago, the analysts suggested that one of his tasks may be to develop Dell-style user customised handsets. Well done Ed, another opportunity lost for Motorola.
In the top 20 there are: 7 SonyEricsson, 6 Nokia, 4 Samsung, 2 LG and only one Motorola made handset. The only Motorola handset in the list is the KRZR, ranking 14.
I'm not sure if analysts use this kind of market trends at all when they make investment decisions, but somebody has to question the reasons behind such a dismal performance.
Motorola has failed to find a followup to the best selling RAZR phone. The company was a victim of its own success with the RAZR by thinking that they found the right formula for a best selling handsets. Instead of developing new handset ideas with new concepts, they kept on releasing models that essentially copy the RAZR concept over and over.
I remember vividly how my casual suggestion of developing a new mobile platform that enables handset customisation was met with ridicule at Motorola.
Since the news that the ex head of mobile devices Ron Garriques is to join Dell came out a couple of months ago, the analysts suggested that one of his tasks may be to develop Dell-style user customised handsets. Well done Ed, another opportunity lost for Motorola.
10 May 2007
Moto CEO Tantrums pay off
So Carl Icahn failed to get a seat on Motorola board. Few months back Icahn raised some valid and very real concerns as an important shareholder of the company, which any sensible corporate interested in adding value to its shareholder should keenly take on board (almost literally!) and address sincerely and with an open mind.
I just could not believe how Ed Zander reacted to Icahn’s concerns: he simply wrote a letter to all other shareholders, effectively shooting down Icahn and attacking his character. As Moto’s politically correct HR policies put it, Zander is expected to “lead by example”, so what a great lesson to teach your subordinates on how SMALL you can get: “don’t take responsibility of trouble you brought on yourself, …do your dirty politics and you’ll be fine”. Frankly speaking, the whole thing read like a scripted version of a street brawl….
Since the late 90s, Moto corporate culture has encouraged managers, directors and executives to get away with such behaviour. So much energy and effort of Moto managers is wasted into: dirty politics and plotting against other Moto managers, trying to dodge blame and never take responsibility of actions, finding scapegoats, cronyism, empire building, ego inflation….etc.
Frankly speaking, I sympathise with Icahn’s concerns. I can not remember the last time Motorola has actually came up with something genuinely new or different. The company in general encourages conformity, so the few bright and imaginative people don’t have the space or encouragement to be creative. Terms like “creative”, “on the ball”, “has the acumen” in Motorola’s lingo are bestowed on the arrogant who, by definition, talk a lot but are not good listeners (essentially experts in BS), and those who can put together slide packs from thin air.
Motorola can go back to its glorious days, it just needs a fresh corporate culture that values new ideas and encourages long term success, not short-term agendas.
I just could not believe how Ed Zander reacted to Icahn’s concerns: he simply wrote a letter to all other shareholders, effectively shooting down Icahn and attacking his character. As Moto’s politically correct HR policies put it, Zander is expected to “lead by example”, so what a great lesson to teach your subordinates on how SMALL you can get: “don’t take responsibility of trouble you brought on yourself, …do your dirty politics and you’ll be fine”. Frankly speaking, the whole thing read like a scripted version of a street brawl….
Since the late 90s, Moto corporate culture has encouraged managers, directors and executives to get away with such behaviour. So much energy and effort of Moto managers is wasted into: dirty politics and plotting against other Moto managers, trying to dodge blame and never take responsibility of actions, finding scapegoats, cronyism, empire building, ego inflation….etc.
Frankly speaking, I sympathise with Icahn’s concerns. I can not remember the last time Motorola has actually came up with something genuinely new or different. The company in general encourages conformity, so the few bright and imaginative people don’t have the space or encouragement to be creative. Terms like “creative”, “on the ball”, “has the acumen” in Motorola’s lingo are bestowed on the arrogant who, by definition, talk a lot but are not good listeners (essentially experts in BS), and those who can put together slide packs from thin air.
Motorola can go back to its glorious days, it just needs a fresh corporate culture that values new ideas and encourages long term success, not short-term agendas.
27 Apr 2007
New Milestone for this blog
Since the beggining of this year, the number of visitors to this blog has been steadily increasing. However, in the past 2 weeks alone, the number of visitors and page views have grown by close to 300%. Earlier this year there was a peak around the time just before and after 3GSM in Barcelone. At that time most of the visits were search hits for terms like: IMS, LTE and VoIP.
The last two weeks show a surge in number of search hits for "Femtocells" and "Infrastructure Sharing", which if anything reflects the frenzy in the industry about the expected shift in mobile network operation strategies.
Other interesting seach words on the blog include "Flat IP Architecture" and "Continuous Packet Connectivity".
Google Analytics reports indicate that most of the hits were from domains owned by the major players: Qualcomm, Siemens, Motorola, Lucent and Nortel.
21 Feb 2007
The drive for cost reduction in mobile networks
A while back I wrote about getting the next billion subscribers and discussed the importance of mobile networks that are cheap to build and run in low ARPU markets.
I take another look here at some of the activities involved in building and maintaining a network and identify ways to make those activities more efficient.
CapEx:
- Network architecture: Choosing the right architecture can make a big difference in terms of expenditure. For example, choosing an architecture that requires a lot of infrastructure or which restricts the options for choosing backhaul and interconnect suppliers can be costly.
- Network design: out of experience, network dimensioning and design activities tend to make pessimistic assumptions about the network layout, so most of the time networks tend to be over-dimensioned for the purpose they were built for.
- Hardware/Software features: Vendors tend to promote bundles of features with their products. Low ARPU operations rely on very few and basic features such as voice and text. Low ARPU operators can demand an almost "bare bones" system from their suppliers, without the bells and whistles.
- Installation/Integration/Commissioning: Having been involved in II&C of various trial networks, these activities can be very costly, typically because of lack of proper project management. Putting processes to eliminate multiple site visits can reduce cost of deployment. In a recent project I was involved in, due to an arrangement the operator had with the company that owns various site locations, accessing sites to install /upgrade /maintenance equipment cost the operator a standard £300 per site per day. Experienced II&C engineers can make a big difference here, getting sites (or nodes) up and running in few hours.
- Spectrum: in low ARPU markets, the business model is very sensitive to the assumptions you make about the business. Try to reflect the spectrum cost accurately and, depending on your business case, decide what is the absolute maximum you are willing to pay to get a license.
OpEx:
- Interconnect: Cost of interconnect is typically around 15% of the network OpEx. There is a huge opportunity to reduce cost of interconnect. There are many providers these days, so managing relationships with interconnect providers and exploiting competition between them can get you good deals. Unfortunately, this is less possible in low ARPU economies where market liberalisation has not been fully achieved yet.
- Site Rental: Site sharing and network sharing is definitely the next area to explore in network operations. Even big players like Vodafone and Orange have decided that network sharing is the way to go. In a small country like UK, there are five network infrastructure layers to essentially cover the same population. Imagine the savings if operators share sites on a larger scale, or even better: share infrastructure.
- Personnel: Personnel charges in labour, training, ..etc is typically around 20% of CapEx. Vendors who aim to build products for low ARPU markets have to design them with simplicity in mind. Some of the equipment I worked on has a ridiculous list of counter intuitive parameters that need to be set. the result is wasted time and effort. When it comes to network infrastructure, usability tend to be at the end of the list. If the effort wasted on fairly mundane activities were to be properly cost-ed, i am sure network operators will force vendors to change their approach.
- Asset Depreciation: This can be as high as 30% of network Opex!. Think of the product life and what you are going to do with it at the end of its life cycle. This is clearly linked to your business model. Financial consultants can often suggest ways to retrieve some of the depreciation charges in tax relief or other ways.
I take another look here at some of the activities involved in building and maintaining a network and identify ways to make those activities more efficient.
CapEx:
- Network architecture: Choosing the right architecture can make a big difference in terms of expenditure. For example, choosing an architecture that requires a lot of infrastructure or which restricts the options for choosing backhaul and interconnect suppliers can be costly.
- Network design: out of experience, network dimensioning and design activities tend to make pessimistic assumptions about the network layout, so most of the time networks tend to be over-dimensioned for the purpose they were built for.
- Hardware/Software features: Vendors tend to promote bundles of features with their products. Low ARPU operations rely on very few and basic features such as voice and text. Low ARPU operators can demand an almost "bare bones" system from their suppliers, without the bells and whistles.
- Installation/Integration/Commissioning: Having been involved in II&C of various trial networks, these activities can be very costly, typically because of lack of proper project management. Putting processes to eliminate multiple site visits can reduce cost of deployment. In a recent project I was involved in, due to an arrangement the operator had with the company that owns various site locations, accessing sites to install /upgrade /maintenance equipment cost the operator a standard £300 per site per day. Experienced II&C engineers can make a big difference here, getting sites (or nodes) up and running in few hours.
- Spectrum: in low ARPU markets, the business model is very sensitive to the assumptions you make about the business. Try to reflect the spectrum cost accurately and, depending on your business case, decide what is the absolute maximum you are willing to pay to get a license.
OpEx:
- Interconnect: Cost of interconnect is typically around 15% of the network OpEx. There is a huge opportunity to reduce cost of interconnect. There are many providers these days, so managing relationships with interconnect providers and exploiting competition between them can get you good deals. Unfortunately, this is less possible in low ARPU economies where market liberalisation has not been fully achieved yet.
- Site Rental: Site sharing and network sharing is definitely the next area to explore in network operations. Even big players like Vodafone and Orange have decided that network sharing is the way to go. In a small country like UK, there are five network infrastructure layers to essentially cover the same population. Imagine the savings if operators share sites on a larger scale, or even better: share infrastructure.
- Personnel: Personnel charges in labour, training, ..etc is typically around 20% of CapEx. Vendors who aim to build products for low ARPU markets have to design them with simplicity in mind. Some of the equipment I worked on has a ridiculous list of counter intuitive parameters that need to be set. the result is wasted time and effort. When it comes to network infrastructure, usability tend to be at the end of the list. If the effort wasted on fairly mundane activities were to be properly cost-ed, i am sure network operators will force vendors to change their approach.
- Asset Depreciation: This can be as high as 30% of network Opex!. Think of the product life and what you are going to do with it at the end of its life cycle. This is clearly linked to your business model. Financial consultants can often suggest ways to retrieve some of the depreciation charges in tax relief or other ways.
14 Feb 2007
The advent of Femto cells - cont'd
Both Epen and Martin made valuable comments on my last posting, so I thought that expanding some of the ideas is in place.
From the consumer point of view, reduced call prices are definitely an important aspect of Femto cells. As for the interference aspect, whether Femto cells will cause more or less interference to Macro cells is debatable. On one hand the signal will be contained within indoors. On the other hand it is likely there will be so many of them. Also, there has to be a reasonable signal penetration into (or out of) buildings to facilitate handover between the two layers.
Will DSL providers get a piece of the pie? It depends who the DSL provider is. As Martin suggested, there are many operators buying DSL assets and are thinking of service "convergence" on the long run. If you are a mere DSL provider, your business model is unlikely to be affected much with Femto cells. Somebody is paying for the DSL subscription (could be the consumer, but could be the operator too), and that's all that matters. If you are someone competing with mobile operators (e.g. BT), then it is likely they will put any hurdle they can think of in the way of Femto cells, unless they make some money out of it.
Femto cells will definitely face fierce competition. If mobile operators loose the battle, at least they can claim that they tried, instead of sitting back and watching the market slip away. They may well loose, but they will loose more by not trying.
There are many handsets with WiFi capabilities around. Whether someone would want to use Skype and ditch the conventional mobile network is a matter of personal preference. For the time being, only technology suave people with love of experimentation do this.
There are also many hurdles that internet based VoIP (e.g. Skype) has to pass before it is adopted on a full scale for mobility. For example, seamless call transfer between the home cell and Macro layer can not happen if you use a simple VoIP routed over the internet: the mobile operator has to be involved somehow. This makes the Femto solution more appealing, because the call is managed easily and seamlessly.
It is not perceived that wireless calls over WiFi will pick up on a large scale, simply because powerful operators (who by the way spent billions on spectrum) will not allow it to happen. So the issue of mobility support is intertwined with the spectrum license and the regulatory aspects associated with it. For example it is not permitted to support mobility on a system using the fixed wireless broadband spectrum in UK.
All in all, here is a summary of all the factors that will influence the Femto cell strategies going forward:
Indeed... only time will tell.
From the consumer point of view, reduced call prices are definitely an important aspect of Femto cells. As for the interference aspect, whether Femto cells will cause more or less interference to Macro cells is debatable. On one hand the signal will be contained within indoors. On the other hand it is likely there will be so many of them. Also, there has to be a reasonable signal penetration into (or out of) buildings to facilitate handover between the two layers.
Will DSL providers get a piece of the pie? It depends who the DSL provider is. As Martin suggested, there are many operators buying DSL assets and are thinking of service "convergence" on the long run. If you are a mere DSL provider, your business model is unlikely to be affected much with Femto cells. Somebody is paying for the DSL subscription (could be the consumer, but could be the operator too), and that's all that matters. If you are someone competing with mobile operators (e.g. BT), then it is likely they will put any hurdle they can think of in the way of Femto cells, unless they make some money out of it.
Femto cells will definitely face fierce competition. If mobile operators loose the battle, at least they can claim that they tried, instead of sitting back and watching the market slip away. They may well loose, but they will loose more by not trying.
There are many handsets with WiFi capabilities around. Whether someone would want to use Skype and ditch the conventional mobile network is a matter of personal preference. For the time being, only technology suave people with love of experimentation do this.
There are also many hurdles that internet based VoIP (e.g. Skype) has to pass before it is adopted on a full scale for mobility. For example, seamless call transfer between the home cell and Macro layer can not happen if you use a simple VoIP routed over the internet: the mobile operator has to be involved somehow. This makes the Femto solution more appealing, because the call is managed easily and seamlessly.
It is not perceived that wireless calls over WiFi will pick up on a large scale, simply because powerful operators (who by the way spent billions on spectrum) will not allow it to happen. So the issue of mobility support is intertwined with the spectrum license and the regulatory aspects associated with it. For example it is not permitted to support mobility on a system using the fixed wireless broadband spectrum in UK.
All in all, here is a summary of all the factors that will influence the Femto cell strategies going forward:
- License: Is it needed, who owns it and what does it cover? What spectrum?
- Mobility: Will seamless communication (handover,routing of incoming and outgoing calls) be part of any competing solutions?
- Content: who has control of it, and how much will it cost to access it?
- Backhaul: who controls DSL backhaul?
- Regulatory: how will the spectrum be allocated and what are the conditions attached to it by the regulatory bodies?
- HW costs: What volumes? Economies of scale? Cost in handset or router?
Indeed... only time will tell.
9 Feb 2007
The advent of Femto Cells
Home base station solutions, also called Femto cells have recently attracted a lot of attention. The idea is to give the mobile user a small box similar to wireless routers, which provides mobile cellular coverage at home. But why would mobile operators want to do that?
In order to answer this question, one has to look at what mobile operators are competing against. Fixed line carriers and DSL service providers are seriously threatening the mobile business by offering new products centered around Fixed Mobile Convergence, which in essence utilises wireless access technology to capture mobile users when at home, and divert their mobile originated voice and data calls to the fixed line network.
Mobile operators found themselves in a predicament: the traditional Macro cellular networks are not able to compete with what DSL service providers can offer, neither from a cost structure point of view, nor in terms of access speeds. (of course the mobile operators are partly to blame because they picked the wrong battle by trying to compete on the basis of access speeds rather than mobile centric applications).
A home base station or femto cell can help mobile operators in many ways. First they can compete directly with other home access solutions which rely on other types of access technology such as WiFi or WiMax. This is because low cost mobile handsets are widely available, whereas dual mode handsets with WiFi capabilities are pricey and still limited in variety.
The femto cell will also enable the operator to cater for subscribers where the service is most needed. Compare this with traditional macro deployment where the infrastructure is first built in places where subscribers are "likely" to appear and even before subscribers start using the network. A femto cell also circumvents the issue of outdoor-to-indoor propagation which typically kills the capacity of 3G and 3.5G systems.
The deployment costs associated with a femto cell are marginal in comparison with Macro cell deployment. This is because the customer is providing the "real estate" to keep the box, the power, as well as the backhaul (e.g. DSL subscription). I don't assume here that Femto cells will replace the macro layer, nevertheless the extent of investment in the macro layer will be considerably less.
Femto cells will also enable the operator to offer creative and disruptive tariffs to directly compete with fixed line. Home zone offers can also be more effective than traditional home zone offers using macro cells which extend for kilometers sometimes (because they rely on a cell ID parameter).
So where is the catch? I will discuss the downside in a future post.
In order to answer this question, one has to look at what mobile operators are competing against. Fixed line carriers and DSL service providers are seriously threatening the mobile business by offering new products centered around Fixed Mobile Convergence, which in essence utilises wireless access technology to capture mobile users when at home, and divert their mobile originated voice and data calls to the fixed line network.
Mobile operators found themselves in a predicament: the traditional Macro cellular networks are not able to compete with what DSL service providers can offer, neither from a cost structure point of view, nor in terms of access speeds. (of course the mobile operators are partly to blame because they picked the wrong battle by trying to compete on the basis of access speeds rather than mobile centric applications).
A home base station or femto cell can help mobile operators in many ways. First they can compete directly with other home access solutions which rely on other types of access technology such as WiFi or WiMax. This is because low cost mobile handsets are widely available, whereas dual mode handsets with WiFi capabilities are pricey and still limited in variety.
The femto cell will also enable the operator to cater for subscribers where the service is most needed. Compare this with traditional macro deployment where the infrastructure is first built in places where subscribers are "likely" to appear and even before subscribers start using the network. A femto cell also circumvents the issue of outdoor-to-indoor propagation which typically kills the capacity of 3G and 3.5G systems.
The deployment costs associated with a femto cell are marginal in comparison with Macro cell deployment. This is because the customer is providing the "real estate" to keep the box, the power, as well as the backhaul (e.g. DSL subscription). I don't assume here that Femto cells will replace the macro layer, nevertheless the extent of investment in the macro layer will be considerably less.
Femto cells will also enable the operator to offer creative and disruptive tariffs to directly compete with fixed line. Home zone offers can also be more effective than traditional home zone offers using macro cells which extend for kilometers sometimes (because they rely on a cell ID parameter).
So where is the catch? I will discuss the downside in a future post.
10 Jan 2007
Advanced Cellular Technologies Episode 6: MIMO
MIMO stands for “Multiple Input Multiple Output”. A MIMO system consists of multiple antennas at both the transmitter and the receiver.
A scattering channel can be problematic because it causes interference and hence sub-optimal reception. Many methods have been introduced to counter the effect of interference, such as interference cancellation, phased array antennas and many of the diversity techniques mentioned in this document. However, it is possible to exploit the spatial diversity aspects of the channel, especially when it is highly dispersive and causes a reasonable degree of scattering. The MIMO concept is designed to take advantage of the channel dispersive nature instead of counteracting it.
There are various techniques that work with a MIMO system. Spatial multiplexing for example works by splitting the data stream to multiple streams (equal to number of transmitting antennas). If the channels are totally independent and uncorrelated it is possible to increase the overall data rate by multiple folds.
Another method is to employ space-time coding where the transmitted data is sent over multiple antennas at different times. With a MIMO system, the combined effect is a reduced power requirement (signal to noise ratio), or an increased throughput if traded off with power.
The scope of MIMO techniques varies from one implementation to another, and a combination of: spatial multiplexing, beamforming, space time coding, array antennas … etc. can all be considered part of a MIMO system implementation.
Theoretically, a NxN MIMO system can increase the throughput in the system by N folds compared to a simple “single input single output” system.
A scattering channel can be problematic because it causes interference and hence sub-optimal reception. Many methods have been introduced to counter the effect of interference, such as interference cancellation, phased array antennas and many of the diversity techniques mentioned in this document. However, it is possible to exploit the spatial diversity aspects of the channel, especially when it is highly dispersive and causes a reasonable degree of scattering. The MIMO concept is designed to take advantage of the channel dispersive nature instead of counteracting it.
There are various techniques that work with a MIMO system. Spatial multiplexing for example works by splitting the data stream to multiple streams (equal to number of transmitting antennas). If the channels are totally independent and uncorrelated it is possible to increase the overall data rate by multiple folds.
Another method is to employ space-time coding where the transmitted data is sent over multiple antennas at different times. With a MIMO system, the combined effect is a reduced power requirement (signal to noise ratio), or an increased throughput if traded off with power.
The scope of MIMO techniques varies from one implementation to another, and a combination of: spatial multiplexing, beamforming, space time coding, array antennas … etc. can all be considered part of a MIMO system implementation.
Theoretically, a NxN MIMO system can increase the throughput in the system by N folds compared to a simple “single input single output” system.
5 Jan 2007
Advanced Cellular Technologies Episode 5: Interference Cancellation for W-CDMA
In a conventional WCDMA system, each user is detected non-cooperatively where other users’ signals are treated as noise. This basic scheme has a low implementation complexity but a limited overall spectral efficiency due to multiple-access interference (MAI).
Higher spectral efficiency can be achieved by reducing the MAI. In contrast to conventional RAKE receivers, the Multi-user detection (MUD) receiver does not treat intra-cell user signals as noise, but tries to eliminate their interfering effect.
There are two main variants of The MUD receiver structures: interference cancellation (IC) and joint detection (JD). The first category involves a subtractive scheme where the receiver produces estimates of the interference and then subtracts them from the received signal. The second category (JD) involves linear transformation of the received signal using, depending on the algorithm, information regarding the signals' time offsets, codes, amplitudes and phases.
The major advantage of this scheme is its relatively low complexity compared to other MUD techniques. Interference cancellation receivers belong to the class of non-linear suboptimal MUD techniques and their complexity is a linear function of the number of users compared to polynomial or exponential complexity for other types of MUD receivers. In effect, IC enhances the interference limited uplink capacity of WCDMA cellular systems.
Higher spectral efficiency can be achieved by reducing the MAI. In contrast to conventional RAKE receivers, the Multi-user detection (MUD) receiver does not treat intra-cell user signals as noise, but tries to eliminate their interfering effect.
There are two main variants of The MUD receiver structures: interference cancellation (IC) and joint detection (JD). The first category involves a subtractive scheme where the receiver produces estimates of the interference and then subtracts them from the received signal. The second category (JD) involves linear transformation of the received signal using, depending on the algorithm, information regarding the signals' time offsets, codes, amplitudes and phases.
The major advantage of this scheme is its relatively low complexity compared to other MUD techniques. Interference cancellation receivers belong to the class of non-linear suboptimal MUD techniques and their complexity is a linear function of the number of users compared to polynomial or exponential complexity for other types of MUD receivers. In effect, IC enhances the interference limited uplink capacity of WCDMA cellular systems.
4 Jan 2007
Advanced Cellular Technologies Episode 4: Evolution of UMTS/HSDPA
3GPP is constantly thinking of ways to improve the performance of UMTS to compete with new emerging technologies that promise enhanced performance over legacy systems.
There are two evolution philosophies within 3GPP. One philosophy is start from scratch and design a new system with new architecture and air interface which do not have any legacy constraints. Another philosophy is to fine-tune existing architecture and introduce advanced features to the existing air interface to enhance its performance to the best level possible while maintaining backward compatibility. The driver for this second approach is to capitalize on existing infrastructure investment. While 3GPP is devising strategies to evolve UMTS on the long term to achieve a set of target performance goals, 3GPP is also thinking of ways to develop the existing UMTS standard to reach its full potential.
Near-term 3GPP enhancements
The near-term enhancements to UMTS are addressed in 3GPP Rel’6 and Rel’7. These primarily focus on the issues directly related to market deployment and the customer experience and to fine-tune and incrementally improve the performance of UMTS earlier standard releases to ensure that UMTS/HSDPA products and services are of high quality, perform correctly and are as fully featured as intended. Some of the goals of these enhancements are:
- Call set-up delay reduction for improved end-user experience.
- Signalling flow optimization.
- Support of real-time packet services such as VoIP.
- Increased system efficiency and overhead reduction.
Medium-term HSDPA evolution: HSPA+
HSPA+ is a manifestation of the evolved HSPA philosophy where the existing HSPA implementation is pushed to its limits by introducing various improvements to increase the efficiency of the system while maintaining backward compatibility. Some of these improvements are:
- Higher order modulation
- The use of advanced receiver such as Equalizers and IC.
- Possibly, the introduction of MIMO receivers and Receive diversity.
The main driver for HSPA+ is to capitalize on existing HSPA investment in infrastructure by focusing on backward compatibility and upgrade simplicity.
Long Term Evolution (LTE) of 3GPP
The main objective for LTE is to enhance the 3GPP standard to become a highly competitive packet-based radio access technology. 3GPP promises huge increases in performance and capacity with LTE. From a performance point of view, the main goals of LTE are:
Flexible spectrum usage with scalable system bandwidth from 1.25 MHz up to 20 MHz
Increased spectrum efficiency and peak data rates at cell edge. Target peak rates of 100 Mbps/DL and 50 Mbps/UL.
Reduced latency for both user and control plane: less than 10ms round trip delay for user plane between UE and the serving RAN node, less than 100ms transition time for control plane between inactive state and active state.
LTE philosophy is more “revolutionary” than HSPA+ in scope. In order to achieve the ambitious goals set for it, LTE takes a fresh look at system architecture and air interface access without the constraints of legacy systems. Therefore system architecture will noticeably change with a new radio access layer.
Although many aspects of LTE are still under discussion in 3GPP, some of the main attributes are:
- OFDM-based air interface (OFDM=Orthogonal Frequency Division Multiplexing).
- Flat IP system architecture. (Often this is described as SAE: System Architecture Evolution, and is a separate study item in 3GPP).
- Higher level modulation and state of the art receiver technology.
3 Jan 2007
Advanced Cellular Technologies Episode 2: Steered Beams
Steered beams are essentially the combination of two technologies: Beam-forming and beam steering.
Beam-forming is achieved by using phased array antennas to create a narrow beam directed at each mobile in the downlink. The antenna elements form a phased array antenna that have very small spacing (usually half a wavelength). By virtue of the small spacing, transmitted signals from the array antennas are added constructively in one particular direction and are added destructively (or nullified) in all other directions. The net effect is a narrow pencil-shaped beam. In contrast, transmit diversity antennas have to be spaced at multiple wavelengths to reduce the correlation between the propagation paths and produce a diversity gain.
Beam steering is achieved by altering the signal phases that feed each of the antenna array elements. This alters the direction in which the signals are added constructively. Feedback information from the mobile terminal is required to continuously alter the phases in such a way that makes the beam follow the mobile as it is moving.
2 Jan 2007
Getting the next billion subscribers - enabling low cost infrastructure
At the end of 2005, the number of global subscribers exceeded the 2 billion mark. Vendors were preparing themselves for the challenge of adding the next billion subscribers. The most optimistic forecasts estimated that the next billion subscribers will be added by the end of 2010.
What happened in fact is that more than 500 million subscribers have already been added by the end of 2006, 41% of which are in AsiaPac area alone. The estimates have changed now, adding the next 500million is expected in the coming 18 months!
In the long term, there is no doubt that developing economies will drive growth in the wireless industry over the next 10 years (should rather use the politically correct term: "emerging markets" instead of "developing countries". The term "developing countries" itself is a milder way of saying "under-developed countries"). Although the telecomm vendors have been talking for a while about the huge opportunities in these emerging markets, they have done very little in the way of efficiently enabling it.
The wireless/mobile business in the merging markets typically exhibit low ARPU, therefore the focus should be on coming with ways to enable a low ARPU business models. There are so many things that can be done by both vendors and operators to reduce cost of deployment and running costs of mobile networks. The emphasis has to shift towards building "no frills", low cost networks that copy the Ryan Air business model of cost competitiveness in every respect with an integrated and holistic approach to cost competitiveness.
Instead, both vendors and operators alike continue to invest huge sums to develop advancements that will offer increasing speeds and better performance, which suggests they learnt very little from the Personal Computer story.
When PCs were first introduced, the driver was providing more technology and better performance, and this was the motivation for Moore's law. With time, the technology matured, and consumers demanded cheaper products, and they were less concerned with the technology content of their PCs. In his book the Invisible Computer Donald Norman argues that as technology develops it reaches a transition point where it offers excess quality that customers tend to be uninterested in. The transition point what differentiates technology-rich, high-performance products from commoditized products that offer just the right level of technology and reliability. (This paradigm perhaps explains why processor makers made the shift to multi-core processors). For today's telecomm vendors to succeed, they need to come up with commoditized, stable, and standard products that can achieve economies of scale, not feature rich ones.
(Graph from Donald Norman's article: the life cycle of technology)
Most handset makers have certainly worked hard to bring about low-cost handsets that can sell below the $30 mark. Qualcomm is already talking about a 3G handset to sell below $120 and prices are in decline. Infrastructure cost reduction however is not on everybody's agenda. The only exceptions to this that I could find are Nokia's announcement in 3GSM last year that they intend to build an inexpensive and “robust” 3G base station, as well as Motorola's “Reach” GSM product line that enables low cost infrastructure.
But low cost infrastructure is not the only part of the puzzle. As I mentioned above, there should be a holistic approach that encompasses virtually everything: planning, deployment, operation, and servicing.
Cheap infrastructure: By providing the very basic features and capabilities, and commoditizing the product to achieve economies of scale. Vendors can leverage some of the commonalities between platforms for example or rely on pure ASIC instead of DSP.
Cost reductions in supply chain: Horizontal integration in the industry will help vendors get cheaper parts if they have large enough economies of scale. Using commoditized products with low variation, operators can source fairly standard ancillaries such as antennas and cables at low prices.
Network Design and deployment: A slight relaxation of network design parameters can save the operator millions in infrastructure equipment. There are various concepts that operators can also benefit from, such as Network Pre-optimisation (this is a concept I invented three years ago) . Operators can also save money by forcing vendors to simplify their products for testing and deployment purposes.
Low cost transmission networks: by using low cost technologies (e.g. over microwave) and Infra structure sharing
Both vendors and operators alike have to think how to reduce network running costs and how to reduce subscriber churn in the network.
I barely touched the surface here. There are hundreds of things that can be done to make mobile telephony as cheap as air.
What happened in fact is that more than 500 million subscribers have already been added by the end of 2006, 41% of which are in AsiaPac area alone. The estimates have changed now, adding the next 500million is expected in the coming 18 months!
In the long term, there is no doubt that developing economies will drive growth in the wireless industry over the next 10 years (should rather use the politically correct term: "emerging markets" instead of "developing countries". The term "developing countries" itself is a milder way of saying "under-developed countries"). Although the telecomm vendors have been talking for a while about the huge opportunities in these emerging markets, they have done very little in the way of efficiently enabling it.
The wireless/mobile business in the merging markets typically exhibit low ARPU, therefore the focus should be on coming with ways to enable a low ARPU business models. There are so many things that can be done by both vendors and operators to reduce cost of deployment and running costs of mobile networks. The emphasis has to shift towards building "no frills", low cost networks that copy the Ryan Air business model of cost competitiveness in every respect with an integrated and holistic approach to cost competitiveness.
Instead, both vendors and operators alike continue to invest huge sums to develop advancements that will offer increasing speeds and better performance, which suggests they learnt very little from the Personal Computer story.
When PCs were first introduced, the driver was providing more technology and better performance, and this was the motivation for Moore's law. With time, the technology matured, and consumers demanded cheaper products, and they were less concerned with the technology content of their PCs. In his book the Invisible Computer Donald Norman argues that as technology develops it reaches a transition point where it offers excess quality that customers tend to be uninterested in. The transition point what differentiates technology-rich, high-performance products from commoditized products that offer just the right level of technology and reliability. (This paradigm perhaps explains why processor makers made the shift to multi-core processors). For today's telecomm vendors to succeed, they need to come up with commoditized, stable, and standard products that can achieve economies of scale, not feature rich ones.
(Graph from Donald Norman's article: the life cycle of technology)
Most handset makers have certainly worked hard to bring about low-cost handsets that can sell below the $30 mark. Qualcomm is already talking about a 3G handset to sell below $120 and prices are in decline. Infrastructure cost reduction however is not on everybody's agenda. The only exceptions to this that I could find are Nokia's announcement in 3GSM last year that they intend to build an inexpensive and “robust” 3G base station, as well as Motorola's “Reach” GSM product line that enables low cost infrastructure.
But low cost infrastructure is not the only part of the puzzle. As I mentioned above, there should be a holistic approach that encompasses virtually everything: planning, deployment, operation, and servicing.
Cheap infrastructure: By providing the very basic features and capabilities, and commoditizing the product to achieve economies of scale. Vendors can leverage some of the commonalities between platforms for example or rely on pure ASIC instead of DSP.
Cost reductions in supply chain: Horizontal integration in the industry will help vendors get cheaper parts if they have large enough economies of scale. Using commoditized products with low variation, operators can source fairly standard ancillaries such as antennas and cables at low prices.
Network Design and deployment: A slight relaxation of network design parameters can save the operator millions in infrastructure equipment. There are various concepts that operators can also benefit from, such as Network Pre-optimisation (this is a concept I invented three years ago) . Operators can also save money by forcing vendors to simplify their products for testing and deployment purposes.
Low cost transmission networks: by using low cost technologies (e.g. over microwave) and Infra structure sharing
Both vendors and operators alike have to think how to reduce network running costs and how to reduce subscriber churn in the network.
I barely touched the surface here. There are hundreds of things that can be done to make mobile telephony as cheap as air.
Advanced Cellular Technologies Episode one: GRAKE
In the next few posts I will provide a high level description of a number of new cellular technologies that are either being implemented or are strong candidates for future generation cellular systems.
Generalised-RAKE, or G-RAKE, has attracted some attention recently, because of the gain it provides over traditional RAKE receivers, while being very similar in architecture and operation.
An important property of CDMA systems on the downlink is that signals within the same cell are transmitted using orthogonal waveforms so that they will not interfere with each another. The so-called “orthogonal codes” separate users from each other and they work best when they are perfectly synchronised and aligned in time.
The multipath effect causes the received signal to be spread in time, therefore the signal is received with duplicates or “echoes” due to channel reflections. This scattering effect causes the signal to be dispersed in time, which causes the orthogonal codes to partially loose their perfect time alignment (orthogonality). Loss of orthogonality leads to what is called: Intra-cell or same-cell interference, which is the interference that users of one cell cause to each other. (This is in distinction to other-cell or inter-cell interference, which is interference coming from neighbouring cells)
A traditional RAKE receiver works by combining the signal duplicates or echoes constructively to achieve a good SNR (signal to Noise ratio). Interference, whether from the same-cell or neighbouring cells, is treated as white noise.
The Generalised-RAKE (G-RAKE) effectively distinguishes between inter-cell interference, which is a pure white noise, and intra-cell interference, which can be mitigated by equalizers. The combining of signals in a Generalized RAKE receiver tries to reverse the adverse effect of intra-cell interference. Unlike the conventional RAKE receiver, G-RAKE tries to match to the channel as well as "whiten" the interference.
The G-RAKE gives a typical improvement of 1-3 dB in SNR for a moderate increase in complexity. G-RAKE provides significant improvement in BER performance, which translates to higher data throughput, as well as increased capacity.
There is a debate about the benefit of G-RAKE in comparison with pure channel equalisation approach. As usual, there is always a trade-off between complexity and performance. The performance improvement also depends on various channel parameters (e.g. speed, reflections, multi-path ...etc).
Generalised-RAKE, or G-RAKE, has attracted some attention recently, because of the gain it provides over traditional RAKE receivers, while being very similar in architecture and operation.
An important property of CDMA systems on the downlink is that signals within the same cell are transmitted using orthogonal waveforms so that they will not interfere with each another. The so-called “orthogonal codes” separate users from each other and they work best when they are perfectly synchronised and aligned in time.
The multipath effect causes the received signal to be spread in time, therefore the signal is received with duplicates or “echoes” due to channel reflections. This scattering effect causes the signal to be dispersed in time, which causes the orthogonal codes to partially loose their perfect time alignment (orthogonality). Loss of orthogonality leads to what is called: Intra-cell or same-cell interference, which is the interference that users of one cell cause to each other. (This is in distinction to other-cell or inter-cell interference, which is interference coming from neighbouring cells)
A traditional RAKE receiver works by combining the signal duplicates or echoes constructively to achieve a good SNR (signal to Noise ratio). Interference, whether from the same-cell or neighbouring cells, is treated as white noise.
The Generalised-RAKE (G-RAKE) effectively distinguishes between inter-cell interference, which is a pure white noise, and intra-cell interference, which can be mitigated by equalizers. The combining of signals in a Generalized RAKE receiver tries to reverse the adverse effect of intra-cell interference. Unlike the conventional RAKE receiver, G-RAKE tries to match to the channel as well as "whiten" the interference.
The G-RAKE gives a typical improvement of 1-3 dB in SNR for a moderate increase in complexity. G-RAKE provides significant improvement in BER performance, which translates to higher data throughput, as well as increased capacity.
There is a debate about the benefit of G-RAKE in comparison with pure channel equalisation approach. As usual, there is always a trade-off between complexity and performance. The performance improvement also depends on various channel parameters (e.g. speed, reflections, multi-path ...etc).
18 Dec 2006
VoIP over UMTS: The Finale - Embrace the trend or block the threat?
This post is a try to summarise and distill a theme from the previous posts about VoIP over UMTS, without getting much into the technical detail, but rather by proposing a view of the telecomm market and motivations for VoIP.
First, let us get the definitions out of the way:
VoIP over cellular:
Carry the voice conversations of mobile users using VoIP by routing them over a packet-switched network, instead of the traditional dedicated, circuit-switched voice transmission lines.
The Market landscape:
Fixed Line broadband prices are falling. Fixed operators offer a huge bandwidth at cheap prices. With a WiFi extension it is possible to become semi-mobile. Fixed line operators realised that they have to diversify their business instead of relegating themselves to become ISPs and ISP carriers. Therefore came up with the Fixed-mobile Convergence concept. They want to win part of the mobility market by putting together compelling UMA offerings. Mobile operators undoubtedly will fight back, probably with home/enterprise collapsed architecture base stations.
Mobile operators see DSL as a headache, because suddenly they have to compete to offer high throughput. They have been fighting an uphill battle with WiFi (and soon WiMax) since they shifted their focus to “high throughput”. Mobile operators are generally unimaginative and still think in terms of “talk”, “text” and recently “broadband”, instead of capitalising on their greatest assets: mobility and location information.
The Proliferation of Wireless broadband has changed the mobility landscape. There are WiFi hotspots on every corner. WiFi Social networks are emerging. (FON): a company that gives you a customised router and allows you to share your broadband with other FONers. Businesses can also make money.
From the regulatory point of view, there is an increasing pressure on reducing mobile roaming charges. Also, it is still ambiguous how mobile data termination charges will be collected, at the moment there aren’t any, which makes VoIP an appealing solution.
Advent of “convergence” has put vendors in a predicament. Now they are trying to sell “standardised” architecture to very different types of customers.
Whenever the industry gets excited about new technology, financial institutions and banks become skeptical because they want to see some ROI on existing investments before committing any new ones.
If they don't embrace VoIP, mobile operators face a threat from it. Enterprise VoIP is already perceived to be a major threat. At the moment they are resist the trend and offering differentiated services to their enterprise customers, e.g. extension dialling
Shift in Markets structure:
The industry is moving away from a vertically integrated arrangement, where operators (think vodafone) own the infrastructure, the frequency, the transport, and applications, the portals, the content and the handsets. The future will be different, companies will own infrastructure to carry multiple types of traffic and a myriad of access technologies (fixed +wireless). Other companies will specialise in content provisioning, others will concentrate on marketing and user acquisitions.
It will be interesting to see how incumbents will shift away from current business models. There is space to diversify revenue streams: For instance, new technology will enable operators to challenge traditional promotion channels. Location information is a great asset and many interesting applications can be wrapped around it.
At the end of the day social re-engineering can open up new business models and allow the industry to get into new businesses. For example, adopting “work from home” ethos will allow telcos to compete with many industries such as: Real estate, car industry, and transportation service providers!.
VoIP over UMTS: motivation and market drivers:
First, let us get the definitions out of the way:
VoIP over cellular:
Carry the voice conversations of mobile users using VoIP by routing them over a packet-switched network, instead of the traditional dedicated, circuit-switched voice transmission lines.
The Market landscape:
Fixed Line broadband prices are falling. Fixed operators offer a huge bandwidth at cheap prices. With a WiFi extension it is possible to become semi-mobile. Fixed line operators realised that they have to diversify their business instead of relegating themselves to become ISPs and ISP carriers. Therefore came up with the Fixed-mobile Convergence concept. They want to win part of the mobility market by putting together compelling UMA offerings. Mobile operators undoubtedly will fight back, probably with home/enterprise collapsed architecture base stations.
Mobile operators see DSL as a headache, because suddenly they have to compete to offer high throughput. They have been fighting an uphill battle with WiFi (and soon WiMax) since they shifted their focus to “high throughput”. Mobile operators are generally unimaginative and still think in terms of “talk”, “text” and recently “broadband”, instead of capitalising on their greatest assets: mobility and location information.
The Proliferation of Wireless broadband has changed the mobility landscape. There are WiFi hotspots on every corner. WiFi Social networks are emerging. (FON): a company that gives you a customised router and allows you to share your broadband with other FONers. Businesses can also make money.
From the regulatory point of view, there is an increasing pressure on reducing mobile roaming charges. Also, it is still ambiguous how mobile data termination charges will be collected, at the moment there aren’t any, which makes VoIP an appealing solution.
Advent of “convergence” has put vendors in a predicament. Now they are trying to sell “standardised” architecture to very different types of customers.
Whenever the industry gets excited about new technology, financial institutions and banks become skeptical because they want to see some ROI on existing investments before committing any new ones.
If they don't embrace VoIP, mobile operators face a threat from it. Enterprise VoIP is already perceived to be a major threat. At the moment they are resist the trend and offering differentiated services to their enterprise customers, e.g. extension dialling
Shift in Markets structure:
The industry is moving away from a vertically integrated arrangement, where operators (think vodafone) own the infrastructure, the frequency, the transport, and applications, the portals, the content and the handsets. The future will be different, companies will own infrastructure to carry multiple types of traffic and a myriad of access technologies (fixed +wireless). Other companies will specialise in content provisioning, others will concentrate on marketing and user acquisitions.
It will be interesting to see how incumbents will shift away from current business models. There is space to diversify revenue streams: For instance, new technology will enable operators to challenge traditional promotion channels. Location information is a great asset and many interesting applications can be wrapped around it.
At the end of the day social re-engineering can open up new business models and allow the industry to get into new businesses. For example, adopting “work from home” ethos will allow telcos to compete with many industries such as: Real estate, car industry, and transportation service providers!.
VoIP over UMTS: motivation and market drivers:
- There is a big hype behind VoIP over cellular, largely because of the attention it attracts in fixed networks and internet world.
- There is a perception that VoIP over cellular will be cheap. This may be true if the operator already has an under-utilized packet core network for example. Nevertheless, offering operator controlled VoIP is not necessarily cheap. Having to run an IMS network to control VoIP introduce what is called “IMS Tax”, where the mobile operator has to deal with increased traffic for the same number of calls, and loose one of their greatest advantages: the ability to carry CS voice efficiently.
- There is also a view that VoIP over cellular is inevitable because once LTE is out CS voice is dead.
- If an operator wants to deploy IMS capabilities, VoIP can be bundled with other added value services and experiences, which hopefully will make a good compelling offer.
- There is a perception that VoIP offers some advantages in terms of capacity , quality and efficiency.
- Convergence: “VoIP fits well with the all IP convergence concept”. The fact of the matter is that convergence is not exactly “all IP”. Convergence is : "the process of replacing several legacy networks (boxes) based on old technology with a single network for all telecomm services, and for all telecom access (fixed + wireless)". So the theme is: a simplified, scalable and integrated transmission and core network.
What is the right VoIP over UMTS operator strategy?
There are two generic strategies which are opposite to each other, and a spectrum of possibilities in between. Essentially, the operator can either offer VoIP is a cheap substitute of voice calls, with limited investment in infrastructure and very much in line with the traditional internet provisioning model, or alternatively, offer a differentiated and high quality voice which is part of a more compelling and complete offering.
Cheap:
- The Premise: Subscribers want to do cheap voice calls.
- Provider mentality: “We give the pipe, you do the skype”.
- Follows the broadband ISP model by providing access to the internet and no additional differentiated services.
- Best effort voice calls.
- User experience is not seamless: standalone applications for VoIP and users have to know how use them.
- Ubiquity to the internet user, who can access their VoIP provider accounts from any terminal using any type of access. (unless blocked by a scrupulous operator).
Premium:
- Premise: Subscribers want to do reliable voice calls.
- Walled Garden mentality: “My network, my revenue”. Operator wants to charge for your VoIP calls.
- Guaranteed quality.
- Seamless user experience: user does not know his calls are carried over VoIP
- User ubiquity undermined by walled garden, perhaps have to pay for ubiquity.
- Differentiate the voice offering by value added services. VoIP becomes part of a complete compelling offer. (not voice centric).
Conclusion:
On the long term, VoIP over cellular is inevitable. However, there are serious issues to be addressed by the cellular mobile operators if they want to still exist few years down the line. They have to make up their mind what business they want to be in, and have to adapt to the new world of horizontal integration. It is very likely that the term "mobile operator" will diminish or disappear with new revenue streaming and business models emerging.
Contact me to get a power point presentation slides on VoIP over UMTS strategy)
5 Dec 2006
VoIP Episode 8: The future of mobile VoIP
I'm linking to an article by Malik Kamal-Saadi on telecoms.com because it has a very good treatment of the subject and in essence summarises most of the VoIP/cellular strategic issues I've touched upon in my previous posts. (I don't know why they did not acknowledge the author's contribution on the telecoms.com site. I've received the written report which clearly shows his name).
One of the interesting points that Malik talks about is the high performance requirements of VoIP cellular handsets. Incidentally, both the new 3X series phones offered by Hutchinson 3 in UK (Nokia N73 and Sony Ericsson W950i walkman) can run a Skype client. Nevertheless there is no doubt that having to support HSDPA/HSUPA with all the R6 and R7 enhancements together with enough processing power to support VoIP calls will definitely be a challenge for UE chip makers and vendors alike.
The threat of VoIP to traditional mobile voice revenue may be an inevitable outcome. In my opinion, VoIP's threat is starting to materialise when competing against the roaming chunk of the voice revenue, where business people for instance find cheap or free WiFi networks to camp on and make international VoIP calls. Apart from that, it does not seem that VoIP is giving operators much worry. Three for example has kept the traditional (line rental) pricing schemes, and introduced the X-series broadband package which the user has to pay for extra. The package already includes a large number of free minutes and TXT allowance. It does seem to me that Three has found a clever way to push their ARPU up without cannibalising their existing revenue, at least in the short term.
One of the barriers of VoIP over cellular is its lack of seamlessness. Not any user will appreciate that to run a VoIP call you have to start a different application than the usual button pressing you do when you call someone the usual way. The fact that the voice call is carried over IP should be transparent to the end user. Handset user interface software has to integrate both experiences in one unified package to give the end user the seamless feel. To make an analogy, just like the misconception that mobile Email is simply a pop3 client on your handset disregards usability issues and user experience, so does the flawed idea that mobile VoIP is simply a Skype or Vonage client.
© Copyright belongs to the author. Do not cite or quote without the express permission of the author.
One of the interesting points that Malik talks about is the high performance requirements of VoIP cellular handsets. Incidentally, both the new 3X series phones offered by Hutchinson 3 in UK (Nokia N73 and Sony Ericsson W950i walkman) can run a Skype client. Nevertheless there is no doubt that having to support HSDPA/HSUPA with all the R6 and R7 enhancements together with enough processing power to support VoIP calls will definitely be a challenge for UE chip makers and vendors alike.
The threat of VoIP to traditional mobile voice revenue may be an inevitable outcome. In my opinion, VoIP's threat is starting to materialise when competing against the roaming chunk of the voice revenue, where business people for instance find cheap or free WiFi networks to camp on and make international VoIP calls. Apart from that, it does not seem that VoIP is giving operators much worry. Three for example has kept the traditional (line rental) pricing schemes, and introduced the X-series broadband package which the user has to pay for extra. The package already includes a large number of free minutes and TXT allowance. It does seem to me that Three has found a clever way to push their ARPU up without cannibalising their existing revenue, at least in the short term.
One of the barriers of VoIP over cellular is its lack of seamlessness. Not any user will appreciate that to run a VoIP call you have to start a different application than the usual button pressing you do when you call someone the usual way. The fact that the voice call is carried over IP should be transparent to the end user. Handset user interface software has to integrate both experiences in one unified package to give the end user the seamless feel. To make an analogy, just like the misconception that mobile Email is simply a pop3 client on your handset disregards usability issues and user experience, so does the flawed idea that mobile VoIP is simply a Skype or Vonage client.
© Copyright belongs to the author. Do not cite or quote without the express permission of the author.
3 Dec 2006
Review of article: What exactly is IMS?
Just read this article by Ian Poole in IET Communications, which is a good overview of IMS architecture. (I don't know why I pay them subscription fees for the magazine when you can download all the articles for free!).
I am not convinced that the need for IMS stems from the desire to "converge". The main objective for IMS is to give the operators a mechanism by which they can charge for any user activity and control the content distributed over their network. With the introduction of 3G data access the operators realised that the future is bleak if they simply offered mobile data speeds for users to access the (mostly) free Internet content. Therefore they wanted a mechanism by which they offer an alternative or additional content they can generate revenue from and control what content the user has access to. Flexible content distribution platform and flexible service provisioning are by-products and not the core issues.
With IMS revenue generation may not be focused on the end user. It could well be based on a sharing revenue with 3rd party content providers or by selling to marketers or retailers alike access to customers.
I am not convinced that the need for IMS stems from the desire to "converge". The main objective for IMS is to give the operators a mechanism by which they can charge for any user activity and control the content distributed over their network. With the introduction of 3G data access the operators realised that the future is bleak if they simply offered mobile data speeds for users to access the (mostly) free Internet content. Therefore they wanted a mechanism by which they offer an alternative or additional content they can generate revenue from and control what content the user has access to. Flexible content distribution platform and flexible service provisioning are by-products and not the core issues.
With IMS revenue generation may not be focused on the end user. It could well be based on a sharing revenue with 3rd party content providers or by selling to marketers or retailers alike access to customers.
1 Dec 2006
Advanced Cellular Technologies Episode 3: Continuous Packet Connectivity (CPC)
CPC is a large 3GPP R7 work item. What is the motivation for it, what does it do and what should we expect out of it? We'll find out in this post.
Motivation:
When a handset is activated it is usually put in Idle state until the user decides to transmit or receive data, at which point the system transfers the handset from the Idle state to a "connected" state. If the user is inactive for a number of seconds, the handset is put again in Idle state in order to save network resources. This turns out to be problematic with certain types of applications that require continuous connectivity, such as IM, hence the motivation for Continuous Packet Connectivity. The vision is to make the mobile bradband experience similar to the DSL experience in terms of network response time.
The main concern with CPC is the amount of battery power required to maintain connectivity. There are various concepts and schemes that, together, will enable CPC. Some of these concepts are:
DPCCH gating:
DPCCH is a control channel which is typically always on. This will change such that control channels have a discontinuous pattern, which reduces the amount of interference and increases battery life.
Modified HS-SCCH operation:
HS-SCCH is a channel that carries HSDPA signaling in the downlink. This channel typically consumes large amount of power which eats up a part of the cell capacity. Therefore 3GPP came with various suggestions to reduce its effects by restricting it, or eliminate it altogether.
CQI reporting reduction:
CQI stands for Channel Quality Indicator, which is a message sent by the terminal to the bas station reporting the channel quality so that the base station can decide what to send to the terminal. Reducing the rate of CQI signaling will reduce overheads. This means that more bandwidth is available for traffic packets instead of overhead.
Motivation:
When a handset is activated it is usually put in Idle state until the user decides to transmit or receive data, at which point the system transfers the handset from the Idle state to a "connected" state. If the user is inactive for a number of seconds, the handset is put again in Idle state in order to save network resources. This turns out to be problematic with certain types of applications that require continuous connectivity, such as IM, hence the motivation for Continuous Packet Connectivity. The vision is to make the mobile bradband experience similar to the DSL experience in terms of network response time.
The main concern with CPC is the amount of battery power required to maintain connectivity. There are various concepts and schemes that, together, will enable CPC. Some of these concepts are:
DPCCH gating:
DPCCH is a control channel which is typically always on. This will change such that control channels have a discontinuous pattern, which reduces the amount of interference and increases battery life.
Modified HS-SCCH operation:
HS-SCCH is a channel that carries HSDPA signaling in the downlink. This channel typically consumes large amount of power which eats up a part of the cell capacity. Therefore 3GPP came with various suggestions to reduce its effects by restricting it, or eliminate it altogether.
CQI reporting reduction:
CQI stands for Channel Quality Indicator, which is a message sent by the terminal to the bas station reporting the channel quality so that the base station can decide what to send to the terminal. Reducing the rate of CQI signaling will reduce overheads. This means that more bandwidth is available for traffic packets instead of overhead.
Subscribe to:
Posts (Atom)