Blog
Comment
mbarnett
Per Dr. Dietmar Wenzel, chair of the DigRF Working Group, DigRF v4 is not ...
Bert22306
Was looking for a cool graphic that showed which countries supported what, but ...
MIPI Alliance offers new specs for RF control
Janine Love
1/17/2012 12:18 PM EST
Today, the MIPI Alliance announced new DigRF and RFFE specifications, and yesterday, I had the opportunity to speak with Jim Ross, of Skyworks, and Dr. Dietmar Wenzel, of Intel about the changes. Both men are part of the MIPI Alliance.
The DigRF v4 v1.10 specification released today still defines the interface between the baseband and the RFIC, but it addresses the increased bandwidth handled by HSPA+ and LTE architectures in one common high-speed interface. Specifically, DigRF v4 v1.10:
"In order to fit different bandwidth requirements, the speed of the physical interface is scalable and can be done quickly with fast cycling," notes Wenzel. He reports that the spec allows the system to be tailored to a lower current consuming mode, optimizing for power. "It reduces time to market and speeds up designs for customers," he adds.
The companies working on the DigRF specification include Arasan Chip Systems Inc., Cadence Design Systems, Inc, Fujitsu Limited, Intel Corporation, Motorola Mobility, Inc.,Nokia Corporation, Panasonic Corporation, Research In Motion, ST-Ericsson, Synopsys, Inc., and Texas Instruments, Inc.
Introduced about a year ago, the RFFE standard aims to provide a consistent method to control RF front end devices, minimizing the need for package pins and routes on the board. RFFE v1.10 highlights include:
According to Ross, in the RFFE standard, the RFIC is the master for the interface, then up to 15 slave devices, such as antenna tuners, antenna switches, LNAs, PAs, and filters, attach to this interface. "It only requires a very select number of pins to control anywhere from 1 to 15 devices, and it doesn't grow in required pins for the number of devices," says Ross.
The primary contributors to the new RFFE spec include Analog Devices Inc., Fujitsu Limited, National Semiconductor, Nokia Corporation, NXP Semiconductors, Panasonic Corporation, Peregrine Semiconductor, Qualcomm Incorporated, RF Micro Devices, Skyworks Solutions, Inc., ST-Ericsson and WiSpry Inc.
Are they worried about LTE-Advanced?
By all accounts, there is great industry uptake of the MIPI Alliance recommendations, but what's next? LTE-A is a hot topic, so you can bet that the MIPI Alliance is already working on it. Wenzel notes that part of the challenge is that the LTE-A topologies are still being defined. He expects that the MIPI Alliance will be able to formally address LTE-A with DigRF in the next one to two years. Technical discussions are ongoing within the organization in order to identify clear requirements. Right now they expect that there will be different versions of RF transceiver topologies, and that needs to be addressed by the spec.
As for RFFE, Ross says that for now the group has identified some "nice to have" features, but haven't identified any show stoppers for LTE-A. The group is still looking at the evolving requirements. "We should satisfy LTE-A given what we know, but we are still investigating what might be a must have in the near future," says Ross. Given that this is still a fairly new specification (only a year and a half old), he notes that the group is still working with the user community to see what new features they could use to help their devices.
For more on the MIPI Alliance:
MIPI Alliance website
M-PHY benefits and challenges
MIPI M-PHY takes center stage
Unification in the RF front-end : the new MIPI standard
The DigRF v4 v1.10 specification released today still defines the interface between the baseband and the RFIC, but it addresses the increased bandwidth handled by HSPA+ and LTE architectures in one common high-speed interface. Specifically, DigRF v4 v1.10:
- supports LTE, HSPA+, WiMAX, 3.5G, 2.5G
- features multiple speed modes, fast cycling
- PHY uses MIPI M-PHY technology which provides a single link between baseband and RFIC
"In order to fit different bandwidth requirements, the speed of the physical interface is scalable and can be done quickly with fast cycling," notes Wenzel. He reports that the spec allows the system to be tailored to a lower current consuming mode, optimizing for power. "It reduces time to market and speeds up designs for customers," he adds.
The companies working on the DigRF specification include Arasan Chip Systems Inc., Cadence Design Systems, Inc, Fujitsu Limited, Intel Corporation, Motorola Mobility, Inc.,Nokia Corporation, Panasonic Corporation, Research In Motion, ST-Ericsson, Synopsys, Inc., and Texas Instruments, Inc.
Introduced about a year ago, the RFFE standard aims to provide a consistent method to control RF front end devices, minimizing the need for package pins and routes on the board. RFFE v1.10 highlights include:
- control for all types of RF front ends
- support for point-to-multi-point connectivity
- multi-mode/multi-band multiple antennas
- up to 15 slaves per bus
- common control interface
According to Ross, in the RFFE standard, the RFIC is the master for the interface, then up to 15 slave devices, such as antenna tuners, antenna switches, LNAs, PAs, and filters, attach to this interface. "It only requires a very select number of pins to control anywhere from 1 to 15 devices, and it doesn't grow in required pins for the number of devices," says Ross.
The primary contributors to the new RFFE spec include Analog Devices Inc., Fujitsu Limited, National Semiconductor, Nokia Corporation, NXP Semiconductors, Panasonic Corporation, Peregrine Semiconductor, Qualcomm Incorporated, RF Micro Devices, Skyworks Solutions, Inc., ST-Ericsson and WiSpry Inc.
Are they worried about LTE-Advanced?
By all accounts, there is great industry uptake of the MIPI Alliance recommendations, but what's next? LTE-A is a hot topic, so you can bet that the MIPI Alliance is already working on it. Wenzel notes that part of the challenge is that the LTE-A topologies are still being defined. He expects that the MIPI Alliance will be able to formally address LTE-A with DigRF in the next one to two years. Technical discussions are ongoing within the organization in order to identify clear requirements. Right now they expect that there will be different versions of RF transceiver topologies, and that needs to be addressed by the spec.
As for RFFE, Ross says that for now the group has identified some "nice to have" features, but haven't identified any show stoppers for LTE-A. The group is still looking at the evolving requirements. "We should satisfy LTE-A given what we know, but we are still investigating what might be a must have in the near future," says Ross. Given that this is still a fairly new specification (only a year and a half old), he notes that the group is still working with the user community to see what new features they could use to help their devices.
For more on the MIPI Alliance:
MIPI Alliance website
M-PHY benefits and challenges
MIPI M-PHY takes center stage
Unification in the RF front-end : the new MIPI standard
|
|
|
|
|
Navigate to related information
Bert22306
1/17/2012 4:09 PM EST
I can't tell if this is truly an all-encompassing standard, as it sounds like on the surface, or whether they are deliberately leaving out the CDMA, cdma2000, and EV-DO variants of 2.5G and 3.5G.
It seems odd to me that Europe continues to be an island of nothing outside of the GPRS/EDGE and the ITU variant of WCDMA, respectively for 2.5G and 3G +. I can't tell if DigRF is also that way, or if it's just the way the various options are being identified in these articles.
Sign in to Reply
mbarnett
1/18/2012 4:31 PM EST
Per Dr. Dietmar Wenzel, chair of the DigRF Working Group, DigRF v4 is not limited to 2G and 3G. It also supports the requirements of today’s LTE implementations and is prepared to support the requirements of future LTE-A/4G systems as well.
DigRF v4 is a completely new generation of BBIC to RFIC interface compared to previous versions of DigRF, i.e. DigRF v1.12 for 2.5G and DigRF v3.09 for 2.5G/3.5G. DigRF v4 is very flexible and configurable. Of course, it can be used for other standards like CDMA, TD-SCDMA, etc. as well.
Sign in to Reply
jackOfManyTrades
1/18/2012 2:20 AM EST
A very strange comment, Bert.
A year or so into my career, I remember reading that you could use a GSM phone in 70 different countries, but couldn't use a single phone all 50 US states. Hardly an "island of nothing".
I was a member of the standardisation committee of an earlier version of this standard. I shall never forget the session (in 2005, I think) where we were talking about the width of a PCB track and a US member said something like "5 or 6 mils" and all us Europeans said "what - 5 or 6mm?". It turned out the US member meant thousandths of an INCH!
Sign in to Reply
Bert22306
1/18/2012 4:42 AM EST
I made two points that I think are factual: (1) that this DigRF only addresses the GSM 2G and 34G radio interfaces (and further checking proved this to be the case), and (2) that Europe only supports GSM.
Both are true. In the US, you have been able to use GSM for many years now, all over. But in addition to that, there's also a widespread other standard, commonly but ambiguously referred to as CDMA. And other standards that have since fallen by the wayside.
Other countries also support this so-called CDMA standard, along with GSM, (Russia, Korea, Australia, and others), but not Europe. In Europe, it's GSM only.
Parenthetically, the reason I object to "CDMA vs GSM" is that this differentiation at best applied during 2G days, where GSM still used TDMA for the radio link. With 3G, everyone went to wideband CDMA, even though the GSM variant is slightly different, and incompatible with, the Qualcomm design.
Sign in to Reply
Bert22306
1/18/2012 5:02 AM EST
Was looking for a cool graphic that showed which countries supported what, but couldn't find it this time.
Not sure about Australia, however it looks like 40 countries do provide so-called CDMA cellular, in addition to GSM, but none of them are in Europe.
Also, I meant 3G in my previous, not 34G. That was fat finger typing, not that I returned from the distant future.
Sign in to Reply