The sun shone on the last day of Mobile World Congress in Barcelona, leaving a warm feeling on those who've worked (and played) hard throughout the week. With almost 50,000 attendees, this is very much the industry central event, covering a very diverse range of the community. Were femtocells as prominent this year as before?
You can call it MWC or whatever, but it will still be the GSM family event for me - I didn't see any signficant presence for WiMAX, CDMA or other technology family - the focus remains clearly around the mainstream GSM/HSPA/LTE path. Perhaps it should be called 4GSM next year. It remains primarily a networking event - there is a conference, but perhaps only 10% of delegates attend; there is a very large exhibition, but we saw far fewer trucks with extensive hardware this year. Instead the focus is on meetings and networking - often putting a face to those you've only spoken to on the phone or by email throughout the rest of the year.
What was the main theme of the event this year?
I asked a few people this and concluded that unlike other years, there wasn't a single dominant theme.
- For some, it was applications. There were plenty to see. An announcement of 24 operators coming together to create a single platform to compete with Apple, Google etc shows how the industry is reacting to the success of these industry giants. The Telstra CTO pointed out how hard it is to support applications across more than 1000 different handsets
- Backhaul (the transmission between cellsites and switching centres) was another hot topic - this has become the primary bottleneck in many overloaded data networks. It's estimated that ATT Wireless will be spending over $8Billion this year alone upgrading their backhaul to match the increased demands from their data traffic. (They could give a femtocell to every customer for free for that kind of money).
- A voice standard for LTE. This new 4G radio technology is already commercially launched, but there still remains some debate about how voice service will be provided. The name VoLTE (not to be confused with VoLGA) is the new name for a simplified IMS solution supported by many operators, previously known as OneVoice. Acme-Packet launched a low cost integrated solution for this, with a price tag of around $2-3 per subscriber.
- Femtocells remained a hot topic though, with most of the femto stands heavily subscribed throughout. The industry seems to have moved on from skepticism and is now figuring out how to take these to market. Investors seem to be taking an extremely active interest.
- How to make money from data. I heard ATT state that 41% of their data traffic originates from 2% of their subscribers (and that some individuals use up to 60GB [not a misprint] per month on their mobile devices), operators have figured out that unlimited data plans aren't viable in the long term. So plenty of discussion about what commercial and technical solutions might fit here - offloading through femtocells being one.
There were also a few (shocking for me) announcements this week, including:
- Verizon will allow (encourage?) Skype to be used on its 3G data network. Positioned as being cheap for international calls, this is an enormous step-change for an industry which makes the bulk of its revenue from voice today. Unlike other Skype partnerships (e.g. with UK operator 3), this will use Skype natively as a VoIP application.
What were the headlines for femtocells?
A common demand from operators (and the public) is for extremely affordable femtocells. Ubiquisys were claiming that the $100 price barrier has now been broken, with their G3 model having been cost engineered by a Far Eastern volume manufacturer using their Femto-Engine design and Percello chipset.
I've already reported how both picoChip and Percello are well positioned for the coming anticipated ramp up in production:
- picoChip have already exceeded their 2009 sales volume and anticipate orders for a million femtocell chipsets by the summer
- Percello showing their 3rd generation chipset which integrates the dynamic RAM, so being virtually hacker-proof and can be paralleled up to provide up to 48 channels.
System Integrators are also expanding their range in what I referred to as Happy Families earlier this week:
- NEC have added Airvana to their eco-system, alongside the Ubiquisys products already commercially deployed
- Alcatel-Lucent have expanded their range of femtocells on offer to include some new vendors, believed to be from the Far East.
- Meanwhile, Sagem (who already supply through Alcatel-Lucent the Vodafone SureSignal product) have expanded their range to include smaller, lower power units as well as integrated modem and set-top box options
Femtocell capacity was also a strong theme, intended mainly for the enterprise market
- Airvana were flexing their muscles too, with an inspired "piano" style demo of 16 concurrent calls on their UMTS femtocell, which has now completed interoperability testing with Kineto Wireless as part of its entry to the NEC eco-system above.
And showing that femtocells are already well positioned for LTE
- picoChip/Continuous Computing/Cavium demonstrated a compact (but working) LTE femtocell. While there's no suggestion that these could be used in place of the large outdoor LTE sites (which need to be multi-sector, high power, longer range etc.), its clear that they will be of great use in high density "not-spots".
- Lime microsystems were touting their new RF chip, which can be configured to handle almost any cellular frequency from 400Mhz to 4GHz and is compatible with both 3G and 4G.
Apologies for any omissions to the above list - I didn't have time to get around everything and there were a lot of press releases during the week.
On a more practical note, Wi-Fi simply didn't work in my hotel some days and was virtually unusable in the harsh environment of a busy exhibition area - my iPhone detected dozens of hotspots, none of which I could use. I could still make calls and receive email over 3G, which has been engineered to handle this kind of load in a more controlled way. Unfortunately I left my phone on 3G data roaming overnight, which ran up a huge bill just checking my email regularly. For me, this re-inforces how cellular technology is a very much better technical solution to meet the demands of burgeoning data traffic levels, but there remains some restructing of (roaming) data pricing to be done. With a sub-$100 price point, femtocells are setup to move from being a practical solution for areas of poor coverage today into a significant answer to the problem of wireless data capacity (and price).