While the term Internet of Things (IoT) is relatively new, the idea behind it is not. We have anticipated the arrival of the “intelligent ambient” since the 1960’s – if not before. 

The fictional worlds of the Jetsons, Star Trek and Dr. Who introduced technologies and capabilities that would anticipate needs and adapt accordingly, providing needed resources when required and maximizing comfort, convenience and well-being.

However, this age of techno-optimism quickly gave way to reality. We were technically limited by the size, power, sensitivity and accuracy of what we could build.

Fast forward nearly 50 years, and we’ve finally developed “smart” technologies that are beginning to realize the vision for a Smart Planet, Smart Cities, Smart Transportation, Smart Grid, Smart Medicine, and more. We are on the verge of a proliferation of electronics not seen since the introduction of the cell phone and laptop computer. We are on the threshold of the Internet of Things, with trillions of connected devices to address nearly every facet of our lives, within the next 15 years.

There are three aspects to the IoT – the central computing resource (cloud), the access points and hubs (gateways), and the distributed sensor nodes (swarm). Texas Instruments has a strong and broad product portfolio in all three areas – from high-performance digital signal processors and wireless connectivity technologies to ultra-low-power, high-precision analog signal conditioning, power management and integrated precision transducers.

Autonomous sensor node requirements
The IoT will include “swarms” of distributed sensor nodes for sensing and detecting the “real world.” Analog signal conditioning is a critical first technology enabler for the IoT, because the analog front end is where the electronic control system meets the real world. It translates the sensory inputs to the transducers into information that the system then acts upon.

In the IoT, each sensor node will be a self-contained, autonomous system that consists of the transducer, analog signal conditioning, radio, embedded processor, power management, and energy source. These nodes must be autonomous due to the sheer quantity of nodes that will be required. Each must be able to operate uninterrupted without external power sources or batteries for at least 10 years, which is no small feat. 

To achieve this, we must build microsystems with power requirements that are a few orders of magnitude lower than those we have today. And to do that, we must pay close attention to every transistor and interconnect in the system and optimize them for power and performance. 

We recognize the important role our analog front ends play and have demonstrated 10x reductions in power for the same capability, but that’s not enough. We’re working hard to achieve power reductions that are orders of magnitude lower while maintaining the performance requirements of the system. To do this, we’re making our analog front ends “smart”.  

Making sensor front ends “smart”
Smart sensor front ends adapt the input signals to compensate for poor response of the transducer, due to non-linearity, temperature variations and other non-idealities, thereby providing a linear response and reducing the complexity of the circuit design. Hence, smart sensor front ends can reduce the complexity and sensitivity needed from the sensors/transducers themselves.

With smart sensor front ends, we can optimize the system for more power reduction without compromising performance. These techniques enable us to achieve orders of magnitude lower power, in addition to making the sensor front ends easier to use, which will allow them to address more applications.

The key to achieving a trillion connected devices and making the IoT a reality lies in our ability to combine advances in low-power circuit design with system-level power optimization, and intelligent energy management. 

The journey continues……………

Want more?
To learn more about the important role sensor signal conditioning plays in a sensing system, check out this recent technical article from my colleague Arun Vemuri. In it, Arun discusses how to remove unwanted signals and achieve the SNR you desire with proper anti-aliasing design.