We interviewed Nigel Toon, the new president and CEO of picoChip (the primary supplier of femtocell chipsets). He gave us his view of the current and future femtocell market, explaining picoChip’s strategy to retain dominant leadership. He believes that within 2 years, operators will be clamouring to get a femtocell into your living room and that when the first operators launch femtocells there will be tremendous competitive pressure for others to follow suit.
Nigel, first tell us a little bit about your background and how you came to picoChip?
I had run the European side of the business for Altera, the FPGA company, which was around 25% of the company’s world-wide turnover. Having helped to grow that business from a few million to nearly $400m, I “avoided” relocating to Silicon Valley after it went public. I co-founded Icera in Bristol, UK with three friends. We could see the need for high performance 3G data device chips, and developed one of the first HSPA products for use in USB sticks and datacards. Icera remain at the forefront of chipsets for high performance HSPA devices.
picoChip and Icera have several investors in common. After discussions towards the end of 2008, I joined picoChip in February – just in time for Mobile World Congress.
What markets are picoChip targeting these days? It used to focus on WiMAX and 3G UMTS femtocells, but now the range seems to have expanded to TD-SCDMA, LTE and even outdoor GSM.
picoChip originated with an innovative high performance multi-core DSP design which is ideal for mobile applications. In the early days, this was focussed on WiMAX and other wireless standards with heavy investment in the physical wireless layer. This same DSP technology can be used for many other wireless variants.
Mobile networks are reaching the limits of what can be technically achieved with the current macro network topology. HSPA is very close to Shannon Law limits – you just can’t squeeze many more bits through the same spectrum. Operators have two solutions for delivering more data capacity:
a) Introduce very small cells which are very efficient and capable
b) Add more wireless spectrum, which is expensive and fraught with regulatory problems
picoChip enable very small cells through two product lines:
picoXcell. Our “gold standard” optimised chips for femtocells.
picoArray. Provides flexibility for rapid development of new products, addressing all major standards including GSM, HSPA, CDMA and WiMAX but with a strong focus today on LTE.
Our roadmap includes new products for both residential and enterprise femtocells. We will offer lower price points, enterprise femtocells that support over 16 channels, improved handover and range enhancements.
Where do you see the femtocell market situation today – are we about to see huge launch and rollout programs from many operators, or will it be a slow takeup?
Within 2 years, I believe operators are going to be clamouring to get a femtocell into your living room.
For example, one major converged carrier told us that today they have about an 80/20 split in their data traffic between their fixed line and their wireless networks – they believe that this will switch within the next 5 years to 80/20 wireless/fixed. That is not because the data traffic on fixed line is reducing, it continues to grow – but because wireless data is growing much, much faster.
In the medium term, I believe wireline POTS (Plain Old Telephone System) will be switched off in favour of all-IP and broadband only fixed line. There won’t be DECT/Cordless phones in your house – they will be replaced by personal mobile phones using femtocells instead.
So we expect over the next 12 to 18 months that most major mobile network operators in most developed countries will launch a femtocell service. Major carriers are launching this year, and once one launches there will be tremendous pressure on their competitors who will have to follow.
The current issue in operators’ minds is around the business model – the technical issues have been more or less solved.
In the long term, I could see operators offering free femtocells with a monthly fee for “all you can eat” voice and data at your home. Operators will want to encourage their customers to offload traffic from the macro network onto femtocells to reduce the cost of delivering data.
WiFi versus Femtocell - How do you see these competing?
3G has several technical benefits over WiFi:
A better link budget leads to better coverage in the home
Power control algorithms in 3G deliver much improved battery life
Better voice quality
Reduced interference because 3G works on controlled/restricted spectrum
Operators are going through a major transition to All-IP. I believe that many of the traditional RAN vendors have seriously underestimated the future data capacity required for mobile networks and that we will see a major a major shift to small cells and new network architectures.
I view picoChip as the “arms dealer” for femtocells – we supply the chips and the software that are making this market happen and today we have the vast majority of design wins and volume. Our plan is to:
Deliver the features that our customers need so that we can continue to lead in market share and make sure they have no reason to use anyone else’s chip
Win new customers as the market expands
We are driving the industry roadmap with aggressive introduction of new features.
The picoChip roadmap cycle is based around:
Major new silicon design every two years. This can include introduction of the latest fabrication technology. We use a 65nm process today and we will evolve to 40nm next year
Annual minor release of new silicon.
Software releases every six months.
Our designs are pin-compatible where possible, so that (for example) our 8 channel chip can be used in existing 4 channel designs. We provide other pinout variants for specific applications. Timing, sync and location functions are being addressed.
What is your main sales line today?
Our volume shipping product today is the PC202. Products based on the PC302 (a system-on-a-chip design) are in development now at customers and will be shipping in high volume before the end of the year.
Who are your competitors?
Qualcomm is clearly going to make a significant impact in the market when they introduce their chipsets next year. But picoChip have been in production for 4 years and have acquired a significant amount of practical extra knowledge in this field. For example, picoChip have IOT’d (Interoperability Tested) with over 100 different handsets and our solutions are being used in over 40 commercial carrier trials and launches.
How has picoChip's strategy changed since you took the helm?
It’s not so much a change in direction – more that we have a stronger focus and accelerated execution.
There is a faster pace of innovation, we’ve become more explicit and directed about what we do.
I’ve also spent a lot of time seeing customers (and spent a lot of time on aeroplanes recently).
How can you compete with the larger silicon vendors?
I'd like to compare picoChip with Atheros, who design WiFi chipsets. They had a similar strategy, and have been able to achieve a huge market share in WiFi, ahead of the major chip companies, by driving the pace of innovation – keeping ahead of their competitors by being nimble and innovative – through leading technologies such as 802.11n.
How do you see the femtocell market evolving
We’ve been impressed by how the network operators have insisted on a standards-based approach. This will enable multiple vendors of femtocell access points within the same carriers.
There's no reason why the majority of these suppliers can't base their designs on the picoChip solution - just look at the DSL modem market where Broadcom dominate.
And finally... What keeps you awake at night?
Jetlag - I've been making a lot of international trips in recent weeks!