Boosting functionality through system-in-package technology

04 March 2019

Embedded systems are now moving into a new phase of complexity and functionality. Even systems that most would consider to be relatively simple are reaching the point where more sophisticated control is needed to maintain their position in the market. This has a knock-on effect on the criteria development teams need to use when they select hardware platforms.

We can see the effect in many fields, but consider a security system as an example. On the surface two electronic door locks may look equally simple, however the functionality they contain can differ dramatically. A traditional design would employ a comparatively simple microcontroller unit (MCU) that interfaces to a peripheral for reading electronic keycards (using a magnetic strip or via an NFC mechanism). These are the types of system that will give way to designs that are, internally speaking, far more advanced and intelligent though. They provide greater functional scope, such as the ability to unlock a door based on face, fingerprint or voice recognition – or potentially combinations of those biometrics.

The core function and logic of the lock remains the same – only unlocking the door if the key (in whatever form that takes) is presented – but the level of sophistication required for the design is much greater. Although it must be noted, that there still needs to be adherence to cost and power constraints of course. This more advanced lock may need to take input from a video camera source, possibly along with a microphone and a fingerprint reader, depending on how many of the biometrics options available are used. High performance digital signal processing (DSP) will be mandated to handle the incoming data from these devices and process each stream intelligently.

Read the full article in the March issue of DPA