Old-fashioned radio

From fitness trackers to smart thermostats, connected products are increasingly common, part of the ever growing Internet of Things. But if the user has difficulty connecting to a product in order to make the most of all the fantastic functionality that is available, they are left with a negative impression. Worse still, if they can’t connect at all, the product is rendered useless. The ability to make and maintain a connection between devices is heavily affected by their radio design, regardless of whether the radio standard being used is Bluetooth, Wi-Fi, or something else. How do we ensure that the radio performance doesn’t hinder a consumer's enjoyment of their new device?

Throughout my career I have heard RF engineering described as “Black Magic”, something that is impenetrable to all but its foremost practitioners. In recent years it has become easier than ever to create wireless products, with a proliferation of single-chip radios, ceramic chip antennas and PCB antenna reference designs. It seems as though there may no longer be any need for the magic at all; has the need for RF engineers to be involved in connected product development come to an end?

You may not need to be a radio engineer to create a functioning radio, but how do you know whether the product is going to operate over the range required? What if it is operating between different rooms in the home or is going to be body-worn? We don't have to accept these factors as unknowns, where we just cross our fingers and hope for the best — by considering the requirements at an early stage in the design they can be addressed in a methodical and scientific way.

For example, if there is a requirement for a wearable product to be able to communicate over a certain range in an indoor environment, statistical models can be used to understand the likelihood of a connection and the performance needed from the circuit design and the antenna design in order to meet the requirement. This information can then be fed into a simulation, where the positioning of the antenna within the product and the proximity of the body can be investigated in order to understand what trade-offs need to be made in order to achieve the required range.

Whether it is the choice of radio standard, the selection of antenna solution and its positioning relative to other electrical and mechanical components, or the selection of component sizes and trace widths, it is possible to make decisions throughout the design process that impact whether a product will give high quality and consistent performance across the expected use cases.


The anechoic chamber in our Boston office

Part of the reason I have enjoyed working on RF design so much at Cambridge Consultants is the wide range of facilities we have to help the design process and enable me to make informed design decisions. Being able to work on the design of a novel antenna in HFSS (High Frequency Structural Simulator), test some prototypes in the anechoic chamber in our Boston office (just meters from my desk!) and plot the 3D radiation pattern of the assembled product on a phantom in our Satimo Stargate chamber in the UK is incredibly powerful. It’s also very satisfying (and often a bit of a relief) to find that a custom antenna solution behaves just as well in the real world as in simulation!

As much as I enjoy people describing my work as black magic, this image of RF engineering is actually harmful. Radio design can and should be tackled with good engineering practice, just like all the other challenges we face at Cambridge Consultants, and the end result will be a superior product.

Brett Wakley