The internet of things is a system that connects objects and users via the internet. “Things” are made much simpler using microelectromechanical (MEMS) technology. MEMS devices come in a variety of flavors, including magnetometers, microphones, oscillators, sensors, and switches. Their small form factor, low power consumption, and manufacturing scalability make MEMS devices a good fit for cost-conscious IoT applications.
MEMS can also improve traditional technologies used in IoT applications. One example is energy harvesting, in which power can be generated by changing factors in the environment such as pressure, temperature, sound, and vibration. These specialty MEMS components can enhance IoT devices with their unique features, such as more compact size and low operating power. Essentially, MEMS in IoT will be the new trend in simplification and accuracy for smart devices.
Let’s look at a few recent unique MEMS products and technologies for the IoT.
Intelligent sensor processing unit
STMicroelectronics has succeeded in combining signal-processing and artificial-intelligence algorithms onto MEMS sensors. This combination injects local decision-making with an added bonus of saving space and power. The intelligent sensor processing unit (ISPU) combines a digital signal processor (DSP) suited to run AI algorithms and MEMS sensors on the same silicon.
By combining the DSP and AI algorithms on MEMS sensors, ST said the ISPU can reduce power consumption by up to 80%. The company also calculates a 5× to 6× power savings over system-in-package approaches in sensor-fusion applications and a 2× to 3× savings in run mode.
The proprietary ultra-low–power DSP, with an enhanced 32-bit reduced instruction set computing (RISC) machine, can be programmed in C. It will allow quantized AI sensors to support full-to-single-bit–precision neural networks for higher accuracy and efficiency for tasks such as activity recognition and anomaly detection by analyzing inertial data.
MEMS microphones
The SiSonic surface-mount MEMS microphones from Knowles Corp. offer small sizes and low profiles with the UltraMini (<11.5 mm) and Slim UltraMini (<8.5 mm) footprints, together with several mounting options. A bottom port feature also enables a thin design.
The MEMS microphones also offer increased output capacities, along with new digital audio options, that help eliminate noise. New MaxRF models are said to eliminate GSM/TDMA burst noise and provide wideband RF noise suppression, and the integrated designs offer differential or switchable gain.
These microphones can also communicate with Knowles’s IntelliSonic software and special porting designs that provide a precisely customized sound. They also offer several performance modes (sleep, low-power, and standard) for voice-trigger applications when entering a low-power, high-signal-to-noise–ratio sensing mode, said the company.
Knowles also developed an AI-enabled true wireless stereo (TWS) development platform. The scalable development platform, consisting of the SiSonic MEMS microphone arrays, voice vibration sensors, and a choice of speaker driver assemblies, enables advanced emerging audio use cases. The system contains wireless left and right channels with a detachable ear bud capability for testing of different configurations. This TWS platform also features premium Knowles small speaker drivers for discreet devices, targeting long battery life. The system also contains a balanced armature tweeter and dynamic woofer for high-definition audio performance.
Knowles’s AI-enabled TWS development platform (Source: Knowles Corp.)
Energy harvesting for wireless sensor nodes
The Fraunhofer Institute for Silicon Technology (ISIT) and the Mechanical and Medical Engineering, Hochschule Furtwangen University in Germany have proposed a zero-power wake-up scheme for energy-efficient sensor applications based on a piezoelectric MEMS energy harvester featuring wafer-level–integrated micromagnets. This technology is presented in the Fraunhofer ISIT study “Broadband Zero-Power Wakeup MEMS Device for Energy-Efficient Sensor Nodes.”
Electromagnetic MEMS, which are based on electromagnetic vibration, are critical devices in energy harvesting, especially in wireless industrial nodes and remote locations. These electromagnetic vibration energy harvesters (EM-VEH) can power remote wireless sensor nodes for the IoT.
Because the method of energy generation for these MEMS devices is from mechanical vibration, energy production is enabled via the motion of a silicon spring positioned between two neodymium-iron-boron (NdFeB) magnets. These are the most powerful, strongest rare-Earth magnets.
This EM-VEH includes a silicon spring, two miniature NdFeB magnets (where one is a tuning magnet and the other is a transducing magnet), and a planar copper micro-coil.
The working principle of the electromagnetic energy harvester is Faraday’s law of electromagnetic induction. This method enables higher power density. The design uses inductors with high volume, very thin-film ferromagnetic material, and magnets.
To demonstrate zero-power standby, a battery-powered microcontroller was awakened by the MEMS harvesting device. Wake-up times in the order of tens of milliseconds can be achieved.
Magnetometer
Memsic magnetometers are new MEMS sensors that are based on anisotropic magnetoresistive (AMR) technology. They can be used in a range of applications, including consumer electronics, automotive, and smartphones.
The technical advantages of a Memsic magnetometer are higher accuracy with measured resolution up to 1 degree, high signal-to-noise ratio, high sensitivity, and low noise. In addition, the AMR sensor technology does not change with temperature compared with Hall sensors that need temperature compensation, and they offer more stable performance that is less affected by environmental changes compared with Hall sensors that have a high zero offset, said the company.
MEMS-based 3D depth camera technology
OQmented has designed an ultra-compact 3D depth-sensing camera, which provides a cost-effective solution for upgrading mobile or stationary cameras with complimentary RGB-D technology.
The biaxial MEMS laser scanner is designed around a patented structured light projector that enables accurate high-resolution scans across an adjustable large field of view. This is unlike conventional low-resolution infrared dot projectors.
OQmented’s LiDAR camera can project dynamically changing infrared patterns by applying its patented Lissajous laser-scanning technology, which is the key to frame rates in the kilohertz range. Concentrating all laser energy of an eye-safe IR laser in a single spot, which is dynamically scanned by the biaxial MEMS mirror, is crucial for overcoming the typical depth range and resolution limitations of standard 3D LiDAR cameras with stationary IR dot projectors.
Adaptive ZeroPower Listening MEMS microphone
The VM3011 from Vesper Technologies, Inc. is the company’s first intelligent digital piezoelectric MEMS microphone with Adaptive ZeroPower Listening technology. This power-saving architecture extends battery standby life by 10×.
This microphone’s Adaptive ZeroPower Listening technology will automatically learn the acoustic characteristics of the environment in real time and allow the system to ignore background noise. The microphone will wake only in response to keywords or other desired acoustic triggers. This will allow the system to hibernate over 90% of the time and extend battery life.
Vesper’s VM3011 intelligent digital piezoelectric MEMS microphone (Source: Vesper Technologies, Inc.)
Applications include IoT products, along with portable smart speakers, smartwatches, and other battery-powered systems. They can also be used in automotive, medical devices, and security cameras.
Conclusion
MEMS is a massively successful technology for the IoT, as modern, mature MEMS manufacturing makes it possible for small, low-cost, high-performance devices. We will see more improvements in devices and applications added to the IoT in the future as MEMS grows and matures with new and enhanced solutions.
Advertisement
