The Internet of Things is unleashing an unimaginable growth in the number of connected devices. In fact, Arm predicts that 1 trillion devices will be connected by 2035, many of which are sensor systems that are designed to sense and interact with the world around them.
The world around us is analog, and analog sensors are often used to provide the sensing mechanism to the microcontroller. Analog circuits are often big, bulky and can add significant costs to a device. Using modern microcontrollers with digital signal processing (DSP) software means it is possible to convert many analog circuits into equivalent software. And DSP is now more accessible than ever, with the Digital Signal Controllers from Arm that allow the use of one processor for both general purpose and DSP processing. Arm's Digital Signal Controllers also provide various performance points by using a familiar, easy-to-use programming model.
In this blog, I will cover how product manufacturers can decrease costs, area and design time, while increasing flexibility and adaptability by replacing analog components with DSP software.
DSP extensions in Cortex processors perform the signal processing required in many every day devices
The first and most obvious benefit to converting an analog circuit into software is that the bill of material costs (BOM) decrease. The reason is that there are now fewer components per board that need to be purchased. In many cases, the reduction in materials is for components such as resistors and capacitors. Surface mount components can be inexpensive, on the order of a penny or less each, so it’s important to weigh the total cost savings over the life of a product to determine if it will be worthwhile. This can be done by taking the cost reduction per board, multiplying it by the volume per year and then multiplying by the number of years the product will be in production.
It’s not uncommon when working with analog circuits and filters to constantly tune the circuits to get the best performance. A minor board re-spin can result in new electrical characteristics for the board that suddenly changes parasitic capacitance or inductance, resulting in a detuned analog circuit. This adds engineering time for teams who then need to modify the analog circuits so that they work as expected again. Converting analog circuits into DSP algorithms not only removes this risk, but also adds in the flexibility to tune and adjust the behavior as needed from software.
Decreasing the BOM cost has a fringe benefit that also allows developers to reduce the dimensions of their product. Fewer analog components, which can sometimes be bulky, means that the board size can be minimized. Decreasing the product dimensions has many cost related benefits to manufacturing such as:
These are all benefits that are often overlooked, yet can make a big impact on the bottom line.
Converting analog circuits to software can help to decrease the design cycle time. There are several reasons why this is true: First, there are quite a few tools available that allow software designers to simulate and generate the DSP algorithms that they need to replace the analog circuitry. This is usually far faster than the time it takes to tune a circuit through circuit simulation and bench testing. Second, if changes need to be made, they can be made in the software which can be done in minutes versus having to re-spin boards or make hardware modifications.
In some products, you never know what conditions will be encountered in the field by the end users. Using DSP algorithms allows a team and even the end user to make on-the-fly adjustments to adapt to conditions in the field without having to make a bunch of hardware modifications. Field adaptability can even simplify the supply chain and manufacturing process by removing the number of product variations that need to be built to meet various customer’s needs.
As we have seen in this blog, there are many benefits to replacing analog circuits with digital signal processing algorithms.It’s important that a development team learn how to evaluate whether they should use an analog circuit or DSP software. The decision may require carefully weighing the benefits to make sure that they will outweigh the additional cost and complexity of the software. If the scale tips towards the DSP software, then the replacing analog components makes a lot of sense.
If you are interested in learning more about how to choose between analog circuits and DSP software, attend our free webinar: How to choose between analog hardware or digital signal processing software, which will be streamed live on January 24th, 2019 and will be available on-demand afterwards.
Register for the webinar