|
|
FROM THE CHIEF STRATEGY OFFICER
|
|
|
|
|
|
|
|
|
|
|
|
Sensor Fusion and New Sensor Interface Developments Open Up Innovation
http://semico.com/content/sensor-fusion-and-new-sensor-interface-developments-open-innovation
published by Tony Massimini on Tue, 2014-11-11 23:36
Last week at the MEMS Executive Congress in Scottsdale, Arizona (Nov. 5-7, 2014) two separate announcements were made that will have long term impact on sensors. The MEMS Industry Group announced the first open source algorithm community for sensor fusion and the MIPI Alliance introduced a new sensor interface specification.
MIPI I3C
The I2C, also known as I Squared C, standard has been used extensively for sensor interface. Many sensor hub controllers, mostly microcontrollers, use I2C for connecting to sensors. But I2C has its limitations in terms of power, speed and scalability. SPI is another interface standard that is used for sensors, but this requires more pins.
MIPI is addressing the interface fragmentation and scalability issues with a new sensor interface specification, MIPI I3C. As that name implies it is backward compatible with I2C. But the new standard provides data throughput capabilities comparable to SPI. According to MIPI “the name MIPI SenseWire℠ will be used to describe the application of I3C℠ in mobile devices and the use of the I3C interface for mobile devices connecting to a set of sensors, directly or indirectly.”
This new standard has been developed because of the steadily growing proliferation of sensors in smartphones. A new standard was needed that could be scalable. MIPI has developed I3C with the participation of sensor vendors and other companies in the mobile ecosystem.
The I3C specification is scheduled for Working Group completion by the end of 2014. Ratification and approval is expected in 1Q 2015. Speaking with MIPI it is possible that I3C could be implemented in devices by the end of 2015.
MIG AIC
The Accelerated Innovation Community (AIC) is the first open source algorithm cooperative for sensor fusion applications. The inaugural AIC member is Freescale which has provided support and innovation to AIC.
AIC’s goal is to accelerate sensor algorithm development. It enables collaborative sensor data collection. The intent is to foster open innovation to spur sensor applications.
For many semiconductor companies, software is not their main strength. AIC will enable smaller players to move up the value chain. Larger players will be able to offload and share R&D costs. An open source community will encourage others to develop different use cases and ecosystems. This will lead to greater diversification, lower cost for software development, and shorten the R&D cycle.
To kick off AIC Freescale is offering its Open Source Sensor Fusion Library which includes C source library for 3-, 6-, and 9- axis Sensor Fusion. There is also a Sensor Fusion Starter Kit that includes a Kinetis based (ARM MCU) development board. Freescale also offers customization services through its software services.
Other companies have quickly added support to AIC. Coming on board are Analog Devices, Berkley Sensor & Actuator Center, Carnegie Mellon University, Kionix and NIST. PNI Sensor Corp. will contribute three algorithms: quaternion to heading pitch and roll; heart rate monitoring using PPG sensor; and step counting. Other MIG member companies are expected to join and provide further support over the next couple of months.
Semico Spin
The MIPI I3C standard and MIG AIC are completely separate developments that have no connection to each other. It is coincidental that they have been announced at the same time. Semico sees this as a serendipitous event. The two complement each other very well. MIPI I3C addresses the hardware side of Sensor Fusion and MIG AIC addresses the software side.
MIPI I3C offers a new topology that will enable more sensors to be designed in. While MIPI focuses on the smartphone and tablet markets, the technology is being leveraged into other applications. The standards are enabling other designs by expanding the capabilities of the ecosystem.
MIPI I3C will enable more flexible designs. Sensor data can be shared between an applications processor and a sensor hub controller. More than one sensor hub controller may be designed in sharing sensor data. Sensor Fusion algorithms can be partitioned among the application processor and sensor hub controllers for different applications. This will open up innovation for new applications.
MIG AIC will enable companies to address new applications which will utilize new designs developed around MIPI I3C. There will be more sensors, specialized algorithms which will become more complex. Many system designers are not familiar with sensors and sensor fusion. MIG AIC will provide tools and software they will need to proliferate sensors into new markets.
A few months ago there were several mergers and acquisitions impacting sensor fusion. In May 2014, Fairchild announced the acquisition of Xsens, the Dutch company known for motion tracking software. On June 24, 2014 Audience announced it would acquire Sensor Platforms. On July 7, 2014 InvenSense announced it was acquiring two companies, Movea and Trusted Positioning, Inc. Thus, the only remaining independent third party sensor fusion developer without a competing chip is Hillcrest Labs.
There was concern at the time that these developments might stifle sensor fusion development. Chip companies wanted to integrate more IP in their respective technologies. The recent announcement of MIG AIC opens up sensor fusion to many more developers. Semico believes that along with MIPI I3C, there will be accelerated development for sensor fusion across new markets.
Musical MEMS?
Written by: Stephen Whalley, Chief Strategy Officer, MEMS Industry Group
I had the pleasure of attending the 8th Annual Body Computing Conference on October 3rd 2014. This was the second year I have attended, and once again it did not disappoint. In one jam-packed day, this conference spearheaded by Dr. Leslie Saxon, Executive Director USC Center For Body Computing, brought together digital health rock stars of innovation from startups to the traditional establishments, investors, academics, athletes, and the general healthcare and technology supply chain. While I could point out a number of interesting new devices, software, APPs and services that were announced at the conference, I’d like to give a brief mention of just one.
SingFit is a musical therapy mobile app. It’s actually a bit of a stretch to mention it at all as it uses little to no MEMS technology. I highlight it though as it won the Body Computing and Skullcandy SLAM contest. The win highlights a growing trend in using music as a way to help patients comply with their therapies. It is also fun and shows great results. Rachel Francine and Andy Tubman developed and created the SingFit app to find new solutions for everything from autism and depression to chronic pain and Parkinson’s disease, by using the world’s oldest medicine, music, in a 21st century fashion. SingFit digitizes the evidence-based music therapy technique of lyric prompting, which enables practically everyone, including those with dementia, autism and traumatic brain injuries to sing on a regular basis in order to achieve therapeutic goals. The videos that Andy showed at the event were truly inspiring. They motivated me to think about what could be done if more MEMS and sensors were used in this way.
The Body Computing Conference topics and general industry landscape point to MEMS/sensors being front and center of the mobile health and handheld/wearable device discussion. During the conference, various speakers mentioned the current and future use and impact of MEMS/sensors throughout the day.
While there have been tremendous advancements in MEMS over the past two decades to meet the demanding needs of high-volume automotive and consumer electronics, we are still in our relative infancy when it comes to small form factor, low-power, low-cost mobile biosensors being applied in wellness and medical applications to deliver an easy-to-use consumer experience. As one analyst and panel discussion pointed out, the future is not wearable…it’s invisible. Is the MEMS/sensor industry capable of delivering on this future anytime soon? When will implantable sensors, skin tattoo sensors and sensor-based clothing actually be a reality for the masses? Lots more work to do here whether you believe it’s upon us already or will happen for the next generation.
The ‘more work to do’ aspect has got to involve closer cooperation between the healthcare industry and MEMS/sensor technologists. While Dr. Saxon ‘s work and conference are a bright spot in bringing technological innovation into healthcare settings, and AliveCor is a notable success story here, we are just scratching the surface of the opportunity and the challenges still to come.
The good news is that many individuals, companies and industry bodies are coming together to discuss and debate the issues. We need to move quickly to not just observe these challenges but to join and co-create the future of digital health. MEMS Industry Group has formed a healthcare working group to focus on what we can do to better serve not only the needs of our members and industry but to see how we can better serve the needs of the healthcare industry and our co-creation partners there. There will be a panel and topic table session on this at the upcoming MEMS Executive Congress US 2014, November 5-7, Scottsdale, AZ. Come sing, dance and co-create with us!
Karen’s Blog – Pittsburgh IMAPS Workshop
Packaging means a lot of different things to a lot of different people. Webster’s dictionary defines package as a “group or a number of things, boxed and offered as a unit.”
For my school-age daughters, packaging means figuring out how to maximize the components of their lunch into these bento-box-like containers I bought at Target in hopes that it would simplify their packaging and assembling process (at low cost and decent performance, mind you). Two months into the school year the packaging appears to be weathering extreme temperatures (cold fridge to hot dishwasher), drop-tests (I am sure you need no explanation here) and what I can only describe as a “cram test” (how many Oreos can you fit inside without the box breaking or my parents noticing).
But if you are in the microelectronics/MEMS industry, when you hear the word packaging your mind goes to the various MEMS packages that can contain a multitude of electrical and mechanical components that are inter-connected to the outside world for devices such as MEMS microphones, airbag accelerometers, gyros, RF MEMS and the list just goes on and on.
I had the pleasure to learn more about the challenges and opportunities affecting MEMS packaging at a recent International Microelectronics Assembly and Packaging Society (IMAPS) workshop held in my hometown of Pittsburgh and at my alma mater, Carnegie Mellon University (CMU). Presenters included our host, Gary Fedder, CMU’s Director of the Institute for Complex Engineered Systems (ICES); Maarten de Boer, CMU Associate Professor, Mechanical Engineering; Brett Diamond, MEMS Development Manager, Akustica; Erdinc Tatar, CMU Graduate Student; and yours truly.
To say that my presentation was different from the others is a gross understatement – I talked about the potential for MEMS and sensors in the expanding world of Internet of Things (IoT) as well as an overview of MEMS/sensors standardization and the proactive role that MEMS Industry Group (MIG) and my partners/members/colleagues are playing in addressing the remaining challenges to commercialization. You can access my presentation on the MIG resource library webpage (no password required).
As the others’ presentations are not posted (at least to my knowledge) I figure I’d give you a quick synopsis of what I learned and heard. Gary basically gave an overview about how amazing and fantastic CMU’s engineering, robotics and computer science departments are and that CMU is now partnering and working with universities and centers around the globe. Literally. They even have two programs going on in China.
Maarten’s presentation on the “Effect of Gas Environment and Materials on Electrical Contact Reliability in Micro- and Nanoswitches” was illuminating as I am somewhat familiar with the work that GE Global Research is doing on RF MEMS switches and am aware of the incredible market potential for this area (I wrote a featured blog on this topic for GE’s “Edison’s Desk” earlier this year). Maarten and his colleagues at CMU are taking this a bit further, by looking into different materials and applications at the nano scale.
Brett’s presentation on “Challenges in the Design, Manufacturing, and Usage of MEMS Microphones” was really impressive as it gave a very in-depth view of the true challenges of packaging a device that by design needs to be open to the environment. No small task and it was equally exciting to hear Brett hint at the future applications and integrations with their MEMS mic’s (I will not repeat them here at the risk of disclosing something I shouldn’t). But let’s just say that the market applications for MEMS microphones are just at the beginning – the potential is really big.
Erdinc’s presentation on “Environmental and Packaging Effects on High-Performance Gyroscopes” revealed why so many engineers love their work in the lab – as they are able to tinker and explore with new materials and processes. It’s another reason why I love my work in MEMS/sensors – because there is still an opportunity for “new science.”
MIG helped sponsor the event by providing snacks (including some great chocolate cookie/pie things that melted in my mouth) for the attendees to enjoy while attending the workshop and to facilitate networking. What I learned at the workshop confirmed what I suspected before – packaging is in the eye of the beholder – and at the end of the day what really matters is that the package is at a cost that is reflective of its application and performance expectations. Therefore, it’s important to communicate those expectations from both the user and supplier’s perspectives.
Packaging means a lot of different things and if done well it can mean the difference between success and failure. Or in my daughters’ case, deciding on how many Oreos to fit into the package before it fails and Mom finds out.
Sensory Shanghai
September 18, 2014/ MEMS INDUSTRY GROUP
Written by: Stephen Whalley, Chief Strategy Officer, MEMS Industry Group
It was over 10 years ago that I last visited Shanghai and oh my, how things have changed, most visibly, the skyline. Looking across the Huangpu River from The Bund back then, I clearly remember the ‘Pearl’ TV tower and a few tall buildings and thought how impressive it looked. Now, the view is an even more sumptuous feast for the eyes, day or night, and it keeps on growing and evolving. So too does the connectivity of the buildings and the people that live and work in Shanghai as the Internet of ‘Things’ brings it all together locally and globally.
Shanghai circa 2001
Shanghai 2014
I was in Shanghai to co-host the inaugural MEMS Industry Group (MIG) Conference Shanghai, September 11-12th, with our local partners, the Shanghai Industrial Technology Research Institute (SITRI) and the Shanghai Institute of Microsystem and Information Technology (SIMIT). The theme was the Internet of Things and how the MEMS and Sensors supply chain needs to evolve to address the explosive growth in China.
As one of our featured speakers, Jérémie Bouchaud of IHS pointed out, China-based smartphone vendors such as Lenovo, Huawei, ZTE, Xiaomi Technology and Oppo collectively make China the third largest handset manufacturer globally behind Apple and Samsung. With each handset having a dozen or more MEMS and sensor components, it’s easy to see why China’s government, research organizations, OEMs and investment funds are paying attention to this market. They also see where it’s going beyond these mobile consumer electronic devices of today. With the rise of smart cities, buildings, farms, homes, vehicles, wearable devices and more — where clusters of sensors abound — the hockey-stick growth predictions are gathering momentum.
With that background, over 150 attendees from China and 15 other countries came together to hear and discuss what’s needed to meet this opportunity. New technologies and products, fabs, funding and city infrastructure were all covered throughout the conference. It was clear that the international MEMS/sensors suppliers do not want to blink and miss the opportunity (as some already have, and to their detriment) and that the local MEMS/sensors community wants to ensure they do all they can to build a self-sustaining supply chain in and around Shanghai. MEMS Industry Group members can access all the great presentations here.
For the time being, the potential and high stakes for all sides means that the Shanghai skyline will be a familiar sight for locals and visitors to cast their eyes, and ‘inter-nets,’ over. And just as the skyline has grown dramatically, it appears the MEMS and sensors industry in China is on the rise too.
To view photos of the inaugural event, visit our Flickr page!
MEMSblog 