Best Practices When Developing an IoT Device

     blog image

    Insights | Oct 04, 2018


    To develop a killer IoT device that meets or exceeds expectations in this fast-moving space, requirements such as data management, connectivity, energy efficiency, security, platform selection, and scalability must be addressed to stay ahead of future needs. And even for companies that have been developing in-house software for years, IoT development poses new challenges. Keeping those requirements and challenges in mind, here are some best practices to consider when developing an IoT device.

    Data Management

    One of the first decisions to make is what to do with the large volumes of data that will be generated from your device and the system in which it operates. Do those bits and bytes ever leave the edge? Where should they be stored – and processed? The decisions you make about data management directly influence your choices around communications protocols, energy management, and security. One of the first things you need is a thorough IoT System Architecture that accounts for the most efficient handling of your data.

    Connectivity

    Your options are nearly endless here based on use cases and budget so there’s no need to focus on the pros and cons of individual networking options and communication protocols yet. Remember that IoT System Architecture we suggested when talking about data management in the paragraph above? Mapping out your whole system of connected devices is the first step to selecting the most appropriate connectivity protocol for each piece of your IoT puzzle. As you consider the communication needs of each device, gateway, and data system in your IoT System, you may find that you have the need for a variety of different wired and wireless protocols to enable fast, efficient, affordable, and scalable communication and data management. One other practice that is often overlooked is to consider what happens to your device or system during the temporary loss and restoration of connectivity. You may even consider developing an “offline first” device or system that functions without the noticeable loss of features should connectivity become an issue.

    Energy Efficiency

    Even the best designed smart, connected IoT device becomes a very fancy brick if the batteries die. Energy efficiency and well-metered consumption thus become some of the most important features of an IoT device. Your connectivity protocols, as you’ve diligently mapped out in your IoT System Architecture, may be the primary driver of your energy needs. Your device may be passive, have limited function and spend most of its days sleeping or napping. But it may also be always-on, performing multiple tasks while processing and sharing data. Low-end edge devices might be perfectly suited for BLE and low power consumption. But when you run into requirements for always-on devices that are crunching and collecting a large bandwidth of data, your energy requirements may entirely be driven by connectivity requirements and cost.

    Security

    As more and more devices are interconnected, vulnerabilities at the device, server, or software level need to be addressed with encryption and authentication. While developing an IoT device, the best opportunity to implement security protocols should be during the design phase at both the hardware and software level. Mapping out your IoT System Architecture and doing some in-house hacking to stress-test your security should give you some confidence that your device and system is secure.

    A thorough solution that is scalable could potentially include a PKI (Public Key Infrastructure) platform with IAM (Identity and Access Management), backed by RBAC (Role-Based Access Control).

    Platform Selection

    Newer embedded development kits and accompanying IDEs have made it easier and less expensive to build out prototypes and low-volume solutions. You might choose an Arduino, Raspberry Pi, or an NXP Rapid IoT Prototyping kit simply for the ease of use in setting up a pilot project or small deployment of a handful of devices. Understand in advance whether your hardware prototype can scale from there without a total redesign – especially if your plans call for secure network monitoring and controls for thousands of devices. With the right IoT IDE platform, you can move all the way from prototype to large scale deployment using the same hardware and software platforms.

    Scalability

    Your IoT device and its associated end-to-end system may be your company’s initial entry into IoT – part of a broader portfolio of applications and capabilities you will build out over time – one that encompasses everything in your world that is connected, monitored and controlled. A flexible IoT software platform allows you to scale your current efforts, hit the ground running on new projects, and keep your IoT portfolio flexible for change as new software, hardware and communication protocols enter the market. A standards-based IoT platform gives you all of the tools you need to remotely manage diverse devices through security updates, analytics visualization, and the integration of new data sources. The built-in interoperability of an IoT software platform enables you to seamlessly scale to millions of endpoints.

    To summarize, developing an IoT device can get complex real quick. If you start with an IoT System Architecture to map out your needs for data management and connectivity, that can drive your options for energy consumption. Never forget the importance of security at every level. Lastly, an IoT platform can help you not only move quickly from prototype to production, but also enable you to scale to millions of devices while remaining flexible for the future.