Source: Content
Compiled
by Semiconductor Industry Observer (ID: icbank)
from
"
semiconductor-digest
", thank you.
Since its launch, the MEMS industry has developed rapidly.
Since the first products in the 1990s, sensors have evolved from measuring pressure, to measuring acceleration (acoustic/mechanical quantities), then to rotational sensing (gyroscopes) and visible light management (DLP), and then to optical sensing beyond visible light (microbolometers, thermopiles), microfluidics, RF, ultrasound, multispectral, etc.
We are now moving towards an era of more holistic sensing, where many parameters are monitored by many different sensors (e.g. sensor hubs). Additionally, MEMS is moving towards an era of more processing/computing because it brings a higher quality to the data.
As shown in Figure 1, the global MEMS market size reached $11.5 billion in 2019. This includes the following MEMS devices, which have a wide range of applications, including: inkjet heads, pressure, microphones, inertial, IR, Si microfluidics, optical, RF, oscillators, environmental, ultrasonic, micro speakers, etc.
Revenue is expected to decline 5.2% this year due to the negative impact of COVID-19 in certain markets:
figure 1
-
The automotive MEMS market will be affected due to the sharp drop in demand for cars during the pandemic lockdown and due to economic insecurity. In addition, the prolonged closure of automotive factories has exacerbated the negative impact.
-
In the defense and aerospace sectors, traditional MEMS sensors (pressure, inertial) used primarily in civil aviation will be affected due to the paralysis of air travel earlier this year, and a reduction in new aircraft orders. MEMS used in defense applications will not be affected much, as government programs will continue and procurement budgets remain unchanged.
-
In consumer applications, the impact depends on the end system. Mobile demand appears to remain low in 2020 as people worry about their jobs, and economic insecurity prompts many to postpone upgrades. But PC/laptop/tablet demand has been strong in the first half of 2020 as users have purchased new systems to cope with stay-at-home demands, both professionally and for education/leisure.
-
Revenues from medical, industrial, and telecommunications will not be affected much. On the contrary, medical and industrial MEMS are expected to grow in 2020 due to their applications in the pandemic: microfluidics for PCR diagnostics, pressure and flow meters for ventilators to help patients with severe respiratory diseases in the ICU, thermopiles in temperature guns are also booming, because fever is one of the main symptoms of COVID-19, which makes microbolometers in thermal imagers help screen people with fever, which brings a boom in related sensors.
From 2021 onwards, the MEMS market will resume growth and the market is expected to reach nearly $17.7 billion by 2025, with a CAGR of 7.4% from 2019-2025. The largest growth relative to market size is expected to come from consumer, industrial and medical MEMS. Yole's previous analysis in calendar 2019 predicted a total of about $18 billion by 2024, so this forecast has been pushed back by one year due to the slowdown caused by COVID-19.
With low prices, where is MEMS going?
Over the past few years, there has been price pressure in the MEMS market, particularly in consumer and automotive applications (Figure 2).
figure 2
Historically, waves of ASP erosion have occurred across a variety of devices, first in automotive (inertial, pressure), then in consumer (inertial, pressure, microphones), followed by RF connectivity (BAW filters) and broad adoption in other consumer products (for example, inertial MEMS and microphones in wearables). We are now at a price point where prices cannot go much lower ($0.4-$0.5).
In fact, some MEMS (such as inertial, microphones, etc.) are already as low as 0.1-0.2 USD, but will still require some active sensing area (Si), and these costs will appear together with the baseline costs. In addition, in addition to the BOM, the company still needs some profit, even if it is very low. So the question is, where do we go next? Various scenarios are possible:
-
First, due to supply and demand adjustments caused by COVID-19, ASPs may temporarily increase slightly driven by supply allocations, then fall back to pre-COVID levels and continue to decline slowly.
-
There is a possibility of a new and stronger wave of ASP erosion due to trends in industrial, medical, and consumer IoT.
-
ASP will increase again due to the addition of features combined with sensors, such as more software, data pre-processing, edge computing, edge AI, etc.
Yole believes that players will try to extract more value by adding functionality/processing to sensors, so the third scenario or a mix of the first and third scenarios is the most likely.
The uniqueness of each MEMS device will drive its future
Across the market, AV, EV, ADAS, AR/VR, Industrial IoT and connected factories, healthcare transformation and telemedicine are impacted by the electronics megatrends. MEMS covers a broad spectrum, including multiple types of MEMS devices related to different end markets/applications.
Despite 2018 being associated with loyalty programs offered by top consumer companies, the inkjet printhead consumer market is suffering from weak demand for home printers due to digitization. Demand for inkjet-based single-pass office printers, which compete with inkjet printers, remains low.
In the future, a small growth is expected due to the demand for large MEMS printheads in the commercial and industrial markets. In industrial applications or in situations where ink may clog the nozzles, ink recirculation is becoming increasingly important for inkjet printhead manufacturers as it improves print quality and printhead life.
The green mobility trend will lead to the need for more pressure sensors in cars (MAP, BAP, transmission oil pressure, GPF, DPF, etc.), partially offset by the move to more EVs (no ICE powertrains). Although the TPMS market was hit hard in 2019, industrial and medical applications will be the future growth drivers, such as process control and HVAC as well as blood monitoring equipment and respiratory equipment. To be sure, strong demand for average selling prices from automotive system and consumer product manufacturers puts equipment manufacturers under great pressure.
The popularity of voice personal assistants (VPA), such as Apple's Siri, Amazon's Alexa and Google Assistant, has pushed audio technology towards new systems that adopt VPA as a new human-machine interface. This interface can be seen as a gadget, such as for smartphones, but is more obvious for kitchen/smart home assistants or smart watches, where their accessibility or small size prevents the emergence of touch interfaces. Or for cars, where hands need to avoid touch screens while driving. The general trend of voice Internet will drive the MEMS microphone market.
Inertial sensors are becoming smarter. The performance is improving all the time, allowing the addition of features in the form of algorithms at the sensor level and for many use cases (step counting, gesture recognition, sports activity recognition). They also have high potential for health and fitness wearables and are gaining traction in the context of medical wearable applications for monitoring patient activity. Their trend from standalone sensors to more combined (IMU) continues.
Optical MEMS are still struggling (e.g. projectors and pico-projectors). However, MEMS micromirrors for LIDAR or other light projection applications seem promising, and many companies are betting on this technology, including AEYE, Luminar, Velodyne, Robosense, Benewake, Valeo, Infineon, Innoviz.
Due to COVID-19, there has been a surge in demand for thermal sensing/imaging MEMS (thermopiles and microbolometers) for fever checking applications. This surge in 2020 could lead to consumerization of thermal imaging/sensing (integrated in smartphones) if the technical/cost requirements are met and consumers are educated on this technology. Huawei released the Honor Play 4 Pro smartphone in June 2020, which integrates a thermal sensor for temperature measurement.
In the field of microfluidics, a growing number of promising companies are using silicon chips for point-of-care testing, which will increase the size of this market segment. New point-of-care products based on silicon/CMOS chips are being developed and are starting to hit the market (InSilixa, DNAe, Mobinostics, Qorvo Biotechnologies, Genalyte, Ontera), and this trend will certainly continue in the coming years. COVID-19 will also boost the market for microfluidics-based diagnostic tests.
5G will be increasingly adopted in smartphones, which will require RF MEMS-based 4×4 MIMO filters. Faster-than-expected penetration may lead to an acceleration in high-frequency BAW technology. This market is expected to continue to grow, especially due to the arrival of new frequency bands dedicated to 5G. MEMS timing (oscillators) are increasingly being adopted in phones from multiple manufacturers, and they will also be needed when deploying new 5G base stations.
Environmental (open) hubs including microphones, temperature, humidity and pressure sensors are increasingly of interest for smart home devices/products. Likewise, in addition to integrating gas sensors into cars, another possibility is to use them, for example, in various infrastructures at key points in cities. Cars can then connect to these infrastructures via V2X and in this way obtain real-time information on local air quality.
Finally, there will be a growing demand for future MEMS such as microspeakers, which can leverage their potential for low power consumption, smaller size, and increased sound directivity and be integrated into many consumer devices.
Traditional electrodynamic speaker technology has not changed much in more than 50 years, and MEMS can provide an alternative for miniaturization. Butterfly Networks is paving the way for the democratization of ultrasound and the consumerization of medical testing with its cMUT.
Furthermore, pMUT for ultrasonic fingerprint sensors in smartphones seems promising, although there is always competition with other technologies (optical fingerprint or 3D facial recognition to unlock the phone, etc.). The need for better privacy in a data-flooded world may mean that having multiple authentication solutions will be mandatory for redundancy reasons.
What’s next and the way forward?
MEMS is a mature technology, and no major breakthroughs have occurred in MEMS manufacturing over the past 10 years. The main technology trends are shifting; previously, they were about increasing yields, reducing costs, and increasing integration; now, MEMS devices are shifting toward higher precision, ultra-low power consumption, and embedded intelligence.
Furthermore, the MEMS market is facing several challenges. On the one hand, increasing development costs for new generations of MEMS devices require more. For this we need an ever-expanding market to spread these costs and minimize any associated risks. On the other hand, focusing only on the sensor itself (at the sensor/hardware level) and its inherent performance is not enough to open new markets.
To get more value from sensors and other types of data (inertial, audio, environmental, etc.), a move toward more sophisticated algorithms, software, and computing levels is needed. For example, one goal is to make smartphones and wearables smarter so that they can anticipate user needs. Thus, various inputs can be fused, such as GPS and inertial/pressure sensors (location), calendars, travel appointments, motion sensors (activity levels), etc.
From their perspective, MEMS sensor manufacturers are seeking differentiation while trying to move up the value chain and remain cost-effective at the same time. To do this, they can follow one of three paths (Figure 3):
-
Find new applications and use cases by exploring new markets;
-
Aggregating capabilities through improved algorithms and software applied to existing use cases;
-
Add in extra processing and ultimately computing, and that could translate into more Si's sold to the same customer.
image 3
Of course, these are not the only options. Some players are already following different strategies in their pursuit of value (Figure 4). Bosch is an example of a company that uses Qualcomm’s software development platform to improve its algorithms for existing customer use cases. Examples include gyroscopes for automotive applications with improved angle detection algorithms and environmental sensors with improved sensing algorithms.
Figure 4
Audio players like Knowles would bet more on data processing near the sensor and additional intelligence at the edge. This is what they implemented on the Google Pixel Phone. The Pixel 3 incorporated 3 microphones from Knowles (costing $0.25 each), but in the Pixel 4 they managed to win a design-win for a standalone audio processor outside the application processor (costing $1.5), ultimately managing to increase value with this customer by 200%.
TDK has partnered with Qualcomm to develop a kit (RoboKit) that uses a fusion algorithm in conjunction with Qualcomm’s AI engine. STMicroelectronics (ST) has an inertial sensor product (iNemo, which is a MEMS IMU) that includes an embedded machine learning core in the MCU.
Finally, there are startups working on edge AI that don’t necessarily make sensors but are partnering with MEMS players (Infineon with Aspinity and Syntiant, STMicroelectronics with Cartesiam) to leverage sensor data. Processing and intelligence at the edge, especially for MEMS microphones and inertial, is becoming increasingly important.
The previous examples point to one thing: collaboration among the various players in the ecosystem, from software to hardware to the system level, is essential to enable the industry to move forward. In the end, a holistic sense is key, especially now that end users want more complete solutions rather than standalone components.
To escape the commoditization cycle and declining sensor value, MEMS vendors can reposition themselves, either by changing their business models or through M&A, to move up the value chain to more data processing and ultimately computing, which can be done first in the cloud and eventually at the edge (Figure 5).
Figure 5
A very fresh example supports this view: mCube (an inertial MEMS company) recently acquired Kinduct, a company that provides complete analytical platform solutions mainly for sports players (but not only), which utilizes inertial sensor data processed in the cloud and can provide actionable insights to customers. Kinduct has a rich list of customers, such as Olympic committees, big names in the NFL and NBA, etc. mCube understands the current problem and targets this growth relay point. Actionable insights related to ultra-sensitive wearable devices and other smart products around us are becoming more and more important.
What is certain is that the additional intelligence derived from MEMS (and other sensor) data will drive more innovation at the sensor and system level and open up new opportunities for the MEMS industry.
★ Click
[Read original text]
at the end of the article
to view the original link of this article.
*Disclaimer: This article is originally written by the author. The content of the article is the author's personal opinion. Semiconductor Industry Observer reprints it only to convey a different point of view. It does not mean that Semiconductor Industry Observer agrees or supports this point of view. If you have any objections, please contact Semiconductor Industry Observer.
Today is the 2527th content shared by "Semiconductor Industry Observer" for you, welcome to follow.
Semiconductor Industry Observation
"
The first vertical media in semiconductor industry
"
Real-time professional original depth
Scan the QR code
, reply to the keywords below, and read more
ICCAD2020|
Storage|Wafer|Huawei
|Automotive Chips|UMC|Qualcomm|RF|Packaging and Testing
Reply
Submit your article
and read "How to become a member of "Semiconductor Industry Observer""
Reply
Search
and you can easily find other articles that interest you!
Click to read the original text to view this article
Original link!