Yet it’s widely accepted that nearly 75% of Internet of Things (IoT) projects fail. Not because of an overly stretched design engineer or flawed design, but because of the business case, or lack of one. With all the tools available it’s easy to get IoT-connected, but not always when looking at the business value.
Design engineers can leverage three key trends when designing for the Internet of Things (IoT) space to greatly reduce time to data visualisation, enabling engineers to focus on the activities that make their design significant and valuable.
Rapid prototyping
The first is rapid prototyping. The key here is to not over-engineer - you have to keep in mind the most viable product. In order to get there quickly, engineers should leverage resources, such as a reference design library (RDL) and take a modular-based approach.
There are already complete ecosystems created such as Arduino, Raspberry Pi or Adafruit Feather for engineers to build into. From there they can select the appropriate add-on technologies like shields, hats, feather wings, Click boards, XBee, etc. or expand via connector ecosystems like Grove or Qwiic to create an initial proof of concept
There’s no need for every engineer to be an RF expert. Communications have been simplified with modules allowing for easy implementation of everything from Bluetooth and Wi-Fi to NB-IoT and LoRa.
Design engineers don’t even have to understand the differences between those technologies – there are people out there that have already done that work.
A simplified design flow reduces the overall time to data visualisation. Free tools, for example, Digi-Key’s Scheme-It and DK IoT Studio allow designers to produce professional schematic diagrams and deploy fully-completed cloud IoT solutions.
Figure 1: The DK IoT Studio allows design engineers to create, deploy and manage a complete embedded-to-cloud IoT solution
Design leverage
IoT design requires knowledge across multiple disciplines. There’s a hardware component, a software component and a cloud component, and typically those involved with each of these components are specialty engineers.
So, if an engineer specialises in hardware, they might not be as strong at the software and cloud pieces, and vice versa. That is why it’s always a team approach in IoT - because of the multiple disciplines involved.
But that’s also part of the challenge if you’re at a small start-up coming up with the next ‘great’ idea – you don’t necessarily have that full team of specialised engineers to hand.
For companies in that position, there are many online communities sharing knowledge to further everyone’s designs, even Amazon and Google put their code on open source and engineers just need to pull in the right APIs to make everything work.
For example, if you want voice recognition with Alexa or Google Home, it’s easy to do. You don’t have to be an expert in voice recognition, you can just pull the right APIs and make a call to them and you seamlessly have voice recognition.
Online communities have set up forums and project repositories to share knowledge, and code repositories like GitHub bring together millions of developers to share and build improved solutions.
Leveraging open source communities as well as hardware design environments often help remove barriers to system design and allow for quicker implementations.
By designing in a tool like the open source EDA KiCad environment, engineers can be exposed to resources for everything from basic design creation tutorials to a robust community support network.
Should even greater expertise be required, there are also design service providers, a network of design firms offering fee-based development, prototyping, manufacturing and systems integrations services.
Advanced deployment
Beyond getting a single prototype up and running, when ready to scale to a production environment or full-fledged deployment, designers have to bring in the security elements and consider adding a cloud services layer to the mix. Most designers can get to the single prototype stage quickly, but scaling takes 10 times more bandwidth.
When getting to the scaling stage, a lot of the layers to be added require more consideration like data services, remote management and maintenance.
One of the latest trends has been sending updates over the air (OTA), similar to how we receive phone updates every few months. Those types of provisioning tools are important to consider when talking about a large-scale deployment.
Designers also have to consider testing as a critical component of scaling and advanced deployment requirements, especially when you involve wireless and RF, which can be a major pain point. If possible, designers should consider trying to partner with testing agencies to facilitate and improve the hit rate on testing.
Of course, security is of the utmost importance and is not always obvious. For example, take creating a beer microbrew application where you’re testing fermentation levels and temperature and all of those other elements. If someone puts that inside a restaurant and it’s all connected to the same network, someone may try and tap through your application to get the credit card data from the main system.
These are the types of things that you need to consider in order to ensure your environment is secure. You also need to ensure your data services and elements are very robust, and that your remote management is accessible at all times.
Trends for innovation
Innovation no longer just comes through the company, whether it’s a Fortune 500 manufacturer or an innovative start-up.
Many companies are now hosting community hackathons and other idea drivers to come up with feature enhancements and other community-sourced ideas.
For example, GE’s FirstBuild sponsors hackathons looking at ways to expand the value proposition of GE equipment. A recent winning team used an Arduino board interfaced to a standard GE oven to create a coffee roasting machine.
Distributors, like Digi-Key, are able to provide a breadth of product from suppliers who are able to provide different pieces of the solution. Distributors are able to filter out the noise and take a look at specific IoT design requirements.
Distributors will not only have the products, but will be able to support the entire ecosystem.
Digi-Key also has many tools to help ensure the success of IoT solution designers.
The Startup Survival Guide has been designed to help companies navigate each step along the Design Roadmap from Ideation and Concept, past Prototyping and Design, all the way to Production, Marketing through to Post-Sales Support.
A Design Dashboard tool provides resources and tasks along the roadmap journey, allowing designers to track each step.
Another example is Scheme-It, which is an ideation tool. Finally, it is able to offer a range of more traditional tools like the BOM Manager to help organise all your component needs.
For those engineers just starting to design for the IoT, Digi-Key has created a 6-part video series with Adafruit that talks about all the elements of an IoT design.
For hands-on learners, Adafruit 4450 is essentially a mini smart home, but also a great reference tool for IoT and a way to familiarise yourself with the DK IoT Studio. This kit includes complete step-by-step instructions for use within the DK IoT Studio.
The IoT Studio is a no coding required environment that enables an IoT designer to select from up to 18 different MCU platforms and marry them to over 100 different sensors and communication modules and breakout boards to specify their IoT design needs.
Each environment includes the embedded design, a mobile app, data aggregation, data transmission, cloud instance and data visualisation. There is simply no easier way to get an IoT rapid prototype up and running. Of course, you can also leverage all of the code generated for you in your own IoT design.
In this day and age, companies really don’t have to focus on re-inventing the wheel. It’s all about leveraging communities and the elements of design that already exist.
Author Details: Robbie Paul is Director of IoT Business Development at Digi-Key Electronics
