Chip-scale sensor systems today are growing increasingly complex and sophisticated. In fact, design engineers working for manufacturers of mobile devices, home appliances, automotive systems and elsewhere are benefiting from the remarkable extension of the usefulness and value of sensors in almost every kind of electronic device.
Various forces underpinning the changing role of sensors in electronics systems include the following:
- The extreme miniaturization of components of many sensor systems. In optical applications, for example, ams pioneered the fabrication of on-wafer interferometric light filters for multi-channel spectral sensors, and micro-optic lens arrays for miniature light sources and light detectors.
- The availability of advanced computing resources in non-computer devices – a capability that enables products such as smartphones to process and intelligently use vast amounts of sensor data.
- The expansion of the Internet of Things (IoT), which gives rise to a new class of connected, autonomous devices that need the ability to see, hear, feel and smell, and to report their “perceptions” to cloud-based monitoring and control systems.
- Climate change, which gives urgent impetus to OEMs’ energy-saving innovations, many of which depend on capturing more precise and relevant information about the real-world phenomena experienced or produced by electronics devices, such as heat, noise and vibration.
These forces are driving sensor and sensor system manufacturers to intensify and accelerate their development efforts. This results in a radical change the chip-scale sensor, which previously was a simple, single-function analog device that supported an application implemented via a microcontroller or other digital device. Now, the sensor is a more complex, intelligent system which itself implements an entire application - a sophisticated and valuable function such as heart-rate measurement, air-quality monitoring or color matching.
This is illustrated by referencing examples of new sensor systems. One example is the 3D imaging sensor solution, a system whose best-known use is user facial recognition in smartphones. At the heart of a facial recognition system is a classic analog optical sensor – a set of photodiodes that detect reflections from the user's face. But to work effectively in the real world, the system needs much more:
- A sophisticated LED light source, often of “structured light,” emitted in complex patterns that enable the generation of a detailed and accurate depth map of the face. The map consists of thousands of points that can overcome the effect of interference from ambient light and other sources.
- Very small, low-profile micro-optics to focus the light from the light source and capture reflections.
- A logic circuit and algorithms to convert the raw reflected light signals into a depth map, and to analyze the depth map to perform facial recognition.
For the facial recognition system to work properly, the hardware elements need to be precisely characterized so they work in unison. The system software is also an essential element of the sensor solution’s operation. In practice, the only viable way an OEM can implement such a complex system is by procuring it as a modular solution. This means the complete application - in this case, 3D depth mapping of the face - is implemented in the module.
The integration of modular sensor systems is an increasingly common feature of new OEM design projects. In areas where ams focuses, such as optical sensing, audio sensing, environmental sensing and imaging, OEM customers are rapidly adopting application-specific sensor systems for object detection and ranging, gas and pollution detection, mobile spectrometry and color analysis, X-ray scanning and color tuning of LED lights.
New and previously unimagined applications are a fascinating consequence of these sophisticated sensor solutions. Chip-scale spectral sensors, for instance, perform incredibly detailed and accurate detection of the spectral content of reflected light, and can be used in portable devices for color matching of fabrics, materials and coatings.
Color analysis is also a powerful tool in many other operations. One area is assessing the dirt levels in waste water produced by a washing machine. If the machine can detect when the clothes are clean, rather than continuing to the end of a set program it can end the washing cycle early, saving energy, water and time. This, and any other chip-scale spectrometry application, relies on the provision of a complete spectrometer module, including firmware for rendering raw photodiode signals as color coordinates.
The future of sensors is complex, sophisticated, application-focused and often supplied as a modular solution. Enough real-world examples already exist for the electronics industry to clearly see where the future is headed.
ams AG
+43 (0) 3136 5000