The Internet of Things (IoT) comprises billions of devices, and each of these nodes has the potential to be compromised, even before installation, if
The Internet of Things (IoT) comprises billions of devices, and each of these nodes has the potential to be compromised, even before installation, if it incurs physical damage while being shipped from the manufacturer to the installation location. The environment it is deployed in can also have an impact, especially if it is hot or humid or there is a significant amount of vibration present. Depending on where devices are deployed, there is always the potential for people to physically interfere with them, or execute a security breach by some other method. However, mesh networks are eliminating the “weakest link in the chain” issue as the multiplicity of nodes and various paths provide a high degree of redundancy – unless one of the end nodes fails.
MarketsandMarkets has estimated that there will be over 17 billion operational IoT nodes by 2023, although other estimates put this number even higher. Whichever number is correct, the task of ensuring security for each and every node and overall network integrity is a massive challenge. Part of the solution is for these nodes to become more intelligent, gaining the ability to check for internal malfunctions and perform other important self-checks as well as ensuring integrity for any data received or transmitted. If an IoT node included a basic accelerometer then, should a drop occur during transport, the severity of the fall could be captured and the microcontroller alerted. Other sensors could determine the orientation of the node as it hit the floor, and the microcontroller could process this data to determine the likelihood that the node had been damaged.
The node could then send a message via an IoT gateway to let the manufacturer, shipping company or customer (or all three) know about the damage and the need for replacement. In order to deliver this functionality, designers need to consider all aspects of the system, especially the MCU, sensors, power management, wireless connectivity and memory. Central to this is the MCU, which must support the ability to self-diagnose so that this functionality can be implemented.
Depending on the sophistication of the node, how much power is available and the amount of data requiring processing, most IoT nodes will use either a 16-bit or 32-bit MCU. If the node is at the end of the mesh or is used as a gateway, it will almost certainly use a 32-bit MCU to process the more sophisticated algorithms that are necessary. Many will include encryption – AES-256 is quite common.
16-bit MCUs that are Power Efficient
The PIC24E family of 16-bit MCUs from Microchip comprises general-purpose 16-bit MCUs that offer high performance, enhanced code density and can deliver 70MIPS throughput from a single 3.3V supply. The fully featured devices include op-amps, motor control, enhanced ADCs, DMA channels for fast data throughput and USB OTGs, and have CAN bus communication capability. Suitable for rugged applications, they can operate in temperatures up to 150°C and offer memory options to 536kB, while occupying footprints as small as 5mm x 5mm.
Texas Instruments offers over 25 different package configurations with varied peripheral configurations in its MSP430 series of ultra-low-power MCUs. Devices in the series incorporate powerful 16-bit RISC CPUs, Flash memory (up to 512kB) and RAM (up to 64kB). There are also a number of methods to maximize code efficiency, including 16-bit registers. They can be woken into active mode in under 6µs, due to oscillators that incorporate digital control.
COMMENTS