We spoke with Dr. Ebrahim Bushehri, CEO and co-founder of Lime Microsystems. Their new single chip RF front end considerably simplifies the overall femtocell chip count and can operate in 3G and 4G air interfaces and frequency bands. We asked him what is so unique about their approach, how they can enter the market and for his views of overall femtocell market takeup.

About Lime Microsystems

Lime Microsystems was formed in 2005 and operates as a fabless chip design company, focussed on the RF front-end chipset required for femtocells. The company currently employs around 20 people, ranging from design, applications and business development. Staff are split between offices in Guildford UK and Lithuania.

After arranging funding in 2006, the company embarked on an R&D program with an initial two chip design for RF front end component of 3G femtocells which it demonstrated in 2008. This was shrunk down to a single integrated LMS-6002 digital chip design announced in 2010. The chip is planned for commercial production later in the year.

What is unique about the Lime Microsystems RF chipset for femtocells?

Unlike some competitive designs, the chip has an extremely wide applicability. It can handle any frequency between 375MHz and 4GHz and common mobile RF technologies including UMTS, LTE and WiMAX, programmed via a set of internal registers. The adaptability of a femtocell using it is limited only by the external filter and power amplifier circuitry which might differ (or be duplicated) depending on the range of frequencies/bandwidths to be supported by the same design.

Working directly with baseband chipsets from companies such as picoChip and Percello, the RF front-end is an essential component of a femtocell. The factors in selecting a vendor include not just unit cost, but also its flexibility.

The challenge for us and any RF transceiver vendor is to be cost competitive and simplify the femtocell vendor supply chain.

Your chip supports many different technologies. Which one takes priority?

We’ve focused on 3G UMTS and LTE from the outset. We expect future LTE femtocells to be dual mode, covering 3G and 4G. Designing a new RF transceiver chip into a Femtocell is not trivial and requires close collaboration with the baseband vendors for proper interfacing and RF calibration.  On this basis, having a single Transceiver IC, covering multiple standards and frequency bands minimizes the need for supporting too many RF chipsets and simplifies the overall design.

We’ve announced reference designs with both picoChip and Percello.

At this time, ADI and Maxim are the key RF transceiver suppliers in this space.

How will you gain entry into the market?

Some femtocell vendors take full control of both baseband and RF, developing their own platform from scratch. Others prefer to use a reference design, reducing time to market. We see the opportunity to gain entry into market through working closely with the former and providing reference designs to the latter.

Ultimately, the market wants a very small board design with an optimum BoM – and it wants it quickly. Tight integration between the hardware components will help reduce the overall production cost.

We’ve paid a lot of attention to the production side of our business. We’ve partnered with Jazz Semiconductor as our volume manufacturer. Our packaging and test supplier will be announced soon – it’s a large, well-known player. Both of these partnerships give us the capability to ramp up production to large quantities with high quality very quickly.

Can you provide any forecast of market size for femtocells in the years ahead?

We’ve worked on the femtocell technology for four years, having come from a macrocell basestation background. We are fully committed to this concept and think femtocells should be seen as another part of the infrastructure. Understandably, network operators have wanted to conduct a period of thorough testing and integration which has accounted for slow initial takeup. We expect to see a healthy deployment in 2010, with more than a million femtocells sold worldwide. 2011 will see this ramping up to 5 million femtocells, with further market expansion thereafter.

Comments (4)

Calvin Wang said:

	Looks good Qs: 1. Does it support TD-SCDMA? 2. Will it be expensive? 3. Any data sheet to share?
	report abuse vote down vote up

March 10, 2010
Votes: +0

ThinkFemtocell said:

	It has to be commercial @Calvin: The chipset will support any mobile radio technology, including TD-SCDMA. It will have to be priced to be commercially attractive - as with all chips, higher volumes mean lower prices. Product details can be found on their website at http://www.limemicro.com/products.php
	report abuse vote down vote up

March 25, 2010 | url
Votes: +0

Tom said:

	They don't have TD-SCDMA sulution yet, from their website. The LMS6002D is a fully integrated multi-band, multi-standard single-chip RF transceiver for 3GPP (WCDMA/HSPA and LTE), 3GPP2 (CDMA2000) and WiMAX applications.
	report abuse vote down vote up

May 15, 2010
Votes: +0

ThinkFemtocell said:

Confirmed it does handle TD-SCDMA @Tom: I checked directly with Lime about this, since it's not obvious that a chipset designed for FDD might also work for TDD. They clarified as follows: "We do support TD-SCDMA. The transceivers designed for TDD usually have one PLL switching between transmit and receive. In our case, we have two PLLs to cater for the FDD on one hand and a setup arrangement to provide the required Phase and Frequency accuracy in TDD mode." I'd add that since TD-SCDMA is likely to be used in many odd/unusual frequencies (since it doesn't require paired up/down spectrum), their chipset is highly attractive because it can match such a wide range of frequencies.

report abuse vote down vote up

May 16, 2010 | url
Votes: +1