IoT and the Evolution of Smart Sensors

How cheap MCUs are turning ubiquitous sensors into the next IoT frontier.

The symbiosis of sensors and chips is leading to smarter and smaller devices, which facilitate more accurate and efficient data collection and processing; and boast self-calibration, self-diagnosis, and self-test capabilities. This fusion of sensing and signal-processing functions is redefining the sensor landscape that has traditionally been tied to discrete elements like thermocouples and accelerometers.
Traditional sensors like thermocouples, resistance temperature detectors (RTDs), and strain gauges have largely been constrained by issues such as system noise, signal attenuation, and response dynamics. Moreover, these sensors produce low-level analog outputs that require precision signal conditioning to safeguard the critical information they are acquiring in the first place.

ADI's smart sensor solution for healthcare wearables

Enter low-cost MCUs and DSPs, enabling multi-sensing and signal-processing capabilities locally inside the smart sensor packages. Microcontrollers, for instance, come integrated with communications interfaces, excitation sources, analog-to-digital (ADC) and digital-to-analog (DAC) converters, discrete I/Os and embedded signal-processing features.

Subsequently, the integration of local compute power transforms an ordinary sensor into a smart sensor, the one that is able to carry out complex calculations on measured data locally within a sensor module. A smart sensor usually comprises of a sensing element, analog interface circuit, ADC, and a bus interface.

Anatomy of Smart Sensors

For a start, a smart sensor allows signal conditioning to occur much closer to the sensor, or even within the sensor package itself, and that preserves signal integrity and ensures signal isolation in harsh industrial environments. Next, a smart sensor can use the local computer power to process and interpret data locally; make decisions based on the physical parameters measured; and communicate accordingly.

That allows these sensors to decide which data to process locally and which data to deliver to the cloud via the Internet of Things (IoT) infrastructure. Smart sensors can create boundary conditions without a human operator. Next, IoT apps can analyze data locally and trigger action or the alarm can be sent via the cloud to the end-user in the form of e-mail, text message, or app notification.
A new level of sophistication that smart sensors enable in fault alarms and process efficiency can transform many industries: industrial, automotive, healthcare, smart building, etc. For instance, the ability to analyze and process data locally can optimize high-speed machining by pinpointing present and future failures and predicting how long parts will last.

Helium Green is a smart sensor for environment monitoring

A smaller amount of data that is more accurate will also enhance the efficiency of cloud and IoT infrastructure. Case in point: almost 80 percent of the power consumed by sensors comes from the wireless links used to transmit data to the edge and cloud. So, a more deliberate and controlled data transfer to the cloud leads to greater power efficiency, a major consideration in the IoT designs.
In retrospect, smart sensors mark an evolution at the intersection of electronics and sensing devices. The cost of integrating electronics has been a major challenge for developers, but with the advent of cheap MCUs with integrated data converters, smart sensor recipe is increasingly becoming more affordable.

Then, there are security issues regarding how to safeguard the sensor data. Here, a new breed of secure MCUs and specialized security co-processors is gradually overcoming a lot of safety concerns.

Finally, smart sensors are bound to comply with a plethora of communications standards such as Wi-Fi, Bluetooth, and ZigBee. Again, the electronics part, mostly comprising of cost-effective MCUs and MPUs, comes to rescue with multiple peripherals supporting a diverse array of communication standards.