Sensors: What’s in YOUR Smartphone?
Sensors: What’s in YOUR Smartphone?
In the growing functionality of high end devices, the spread of basic smartphones, and the emergence of new device types – wearables, etc. – sensors are a $60.4B market that is growing at a 9.2% CAGR. We categorize them into 5 groups – physical, optical, radio, electrical and chemical – each serving specific applications and requiring particular design and manufacturing disciplines. As such, competition amongst sensor manufacturers stays within these silos. For sensor functions already well embedded into portable devices, suppliers are challenged to minimize cost, footprint and power draw, often integrating multiple functions of a type into a single solution. Emerging functions, such as gesture control or biometric ID, have drawn approaches using different sensor types, although economics and performance will eventually drive standardization. New devices are a growth market for existing sensor types, but may also create opportunity for new sensor components suited to specific applications. While we are bullish on new portable device opportunities for sensor components, we note that many longstanding sensor products are commoditized, particularly within well established traditional end markets, such as automotive and industrial automation.
The proliferation of electronic devices is driving sensor demand. Sensors are components that measure and convert natural phenomena into digital data that can be analyzed by electronic devices. Traditionally, the sensor market has been primarily automotive and industrial, but as personal devices have proliferated, they have catalyzed significant growth. Sensors are growing as a percentage of the BOM for high end smartphones, while the rapid growth of low end smartphones drives unit volume growth for basic sensors. The emergence of new device categories, such as wearables and home automation, and potential breakthroughs with autonomous vehicles and commercial drones may be considerable future opportunities for sensor makers. We project better than 10% overall CAGR for sensors over the next 5 years, based on these drivers.
Sensor demand segmented into 5 distinct silos. Sensors fall into 5 very different categories, based on the type of stimulus to which they are designed to respond: Physical, Optical, Radio, Electrical and Chemical. Each of the 5 types favors specific design methodology and manufacturing processes that has kept competition amongst sensor makers confined by those types. While integrating multiple functions within a type to a single module is a common strategy to reduce cost, footprint and power draw, functions are typically NOT combined across types. As such, the companies playing in sensors usually specialize in just 1 or 2 categories. Today, optical and RF ICs are the biggest parts of mobile device BOM, but physical and electrical sensors are projected as the fastest growing segments.
Physical sensors are increasingly based on MEMS. The most common physical sensors measure orientation, acceleration and vibration, functions built into all higher end smartphones. State of the art designs include all three sensors into a single component built as a microelectromechanical system (MEMS) – a tiny, moving machine built on a silicon chip. Increasingly, MEMS is also used for high performance microphones to process sound, cancel noise and measure proximity based on ultrasonic waves. MEMS components are also used by high end cameras for image stabilization. 11.2% projected growth in physical sensors will come from the increased precision of integrated components for high end devices, the proliferation of lower end devices with rudimentary components, and from the emergence of new device types like wearables.
Cameras, gesture control, and biometrics to drive optical sensors. The camera quality war in high-end smartphones drives more, and more sensitive, sensors. Stereoscopic focus could add a second sensor to the primary camera. Gesture controls could require multiple forward cameras, which could also support facial recognition or retina scans for ID. Other applications include infrared proximity sensors, ambient light meters, and UV meters. Components for these applications have converged on silicon CMOS, and like physical sensors, will see demand growth from increasingly sophisticated parts for high end products and commodity cameras for the entry level. We note that autonomous driving systems and commercial drones will rely on batteries of highly sensitive optical sensors.
Radio requirements continue to get more complicated. As 3G and 4G networks roll out into new spectrum worldwide and as the standards evolve to higher performance, radio component makers are asked to design integrated antennas, filters, and power amps to fit the broadening connectivity into compact and power efficient components. This integration brings down prices, partly offsetting volume growth, but offers opportunity for differentiation, particularly as high frequencies require special material (GaAs, SiG) and mixed signal process expertise. However, simpler RF elements for WiFi, Bluetooth, GPS and other standards are now integrated into digital parts, and as CMOS improves, cellular RF will likely eventually be integrated as well, an enormous risk for RF makers.
Capacitive touch sensors looking for differentiation. Electrical sensors are used for “touch” applications, like displays or fingerprint readers, usually combining an electrostatic charged surface and a semiconductor controller to interpret changes in the electric field as it is touched. Demand for touch screen modules continues at a 10%+ CAGR, but commoditization for basic displays, which make up most of the demand and growth, has driven down margins for many suppliers. New applications, such as fingerprint scanners or “no touch” proximity sensors, are opportunities for differentiation at the high end while strict price/performance will continue to rule at the low end.
Chemical sensors are a nascent opportunity. Chemical sensors, which measure the presence of specific molecules, have not been introduced into common consumer devices. Future applications could be breathalyzers, air quality testing, or blood analyses, although it is not clear if any of these have broad enough appeal for inclusion in general purpose devices, particularly since chemical sensors are specific to particular substances. Wearables, home devices and peripherals appear to be more appropriate venues for these sorts of applications, and thus, will likely have a much smaller end market.
Winners will dominate niches and aggressively integrate. In each sensor type, winners are emerging with scale and ability to integrate functionality. We see QCOM as uniquely able to integrate sensor functionality, beginning with RF, into reference designs – a substantial advantage, while SKYW is gaining share in RF near term. We also see substantial room for differentiation and integration in MEMS physical sensors – a boon for INVN, and KN. Optical sensors are a growth market without pure plays for investors. Touch screens are commoditizing, but SYNA is innovating and consolidating share.
For our full research notes, please visit our published research site.