Best Practices for Designing IoT Edge Devices

Presented by Jacob Beningo

Development teams are always under pressure to deliver faster and at lower costs, but this is becoming more challenging as system complexity has risen exponentially with features for IoT and Machine Learning. The increased complexity can easily handcuff a development team and lead to not just longer development cycles with higher costs but also lower quality products.

In this session, we will explore best practices for developing real-time embedded systems that will help the modern developer stay on track and produce a quality product within their development cycle. We will explore best practices ranging from how to properly architect a system for scalability, how to manage a development cycle, secure and test a system. We will also discuss best practices for using frameworks and open-source software.

Go to Session

Tips and Tricks for Developing Products using the ESP32

Presented by Jacob Beningo

The ESP32 Wi-Fi / Bluetooth module has found its way into not only hobbyist applications but also professional IoT devices. This low-cost device with its open source software and strong ecosystem has many interesting characteristics that make it extremely attractive for use in a design. In this session, we are going to look at the nuances of designing products using the ESP32 and provide tips and tricks for successfully launching a product with it.

Topics covered in the session include:

• How to select the right ESP32 module
• Partitioning the Wi-Fi and Bluetooth capabilities with application software
• Advanced processor features, multiplexing and other capabilities
• How to leverage the open source software ecosystem
• Hardware design tips and tricks

Go to Session

Hawkbit and SWUpdate for OTA Updates of Linux IoT Devices

Presented by Mohammed Billoo

Linux-based IoT devices don't have a flat Flash structure that allows for a straightforward OTA update process. There are a multitude of bootloader, kernel, and root file system combinations that must be taken into consideration, along with the different failure modes. The combination of Hawkbit and SWUpdate aim to standardize the OTA update process for IoT devices running Linux. This session will provide an overview of Hawkbit and SWUpdate, the different components that can be updated, and the different failure modes that are supported. A case study will be presented to demonstrate how to set up Hawkbit and SWUpdate to support OTA updates for a Linux-based IoT device. Finally, ongoing work by the presenter to address current shortcomings will be discussed.

Go to Session

Easy CI/CD for Embedded IoT Devices

Presented by Dustin Reynolds

Continuous Integration and Continuous Delivery are a great fit for embedded IoT devices, and it has never been easier to get started with it. CI/CD lets you perform automated code quality and functionality tests against your firmware and software prior to it being deployed to your fleet.  

I will build a container-based CI/CD pipeline using Concourse CI using a popular IoT based embedded platform.  I will review the design of the CI/CD pipeline, showing all the details to build your own and conclude with a demo showing the full workflow from source code commit to automated update using the CI/CD pipeline.

Go to Session

Batteryless Sensing Systems Powered by Light

Presented by Andres Gomez

Over the last decade, energy harvesting has seen significant growth as different markets adopt green, sustainable ways to produce electrical energy. Even though costs have fallen, the embedded computing and internet-of-things community have not yet widely adopted energy harvesting-based solutions. This is partly due to a mismatch between power density in energy harvesters and electronic devices which, until recently, required a battery to be functional. This mismatch is especially accentuated in indoor environments, where there is comparably less primary energy available than in outdoor environments. In this talk, I will focus on photovoltaic harvesting, which has little requirements and a high energy density. Furthermore, I will present a design methodology based on source-load decoupling that can optimize energy flow in dynamic environments. To demonstrate its general applicability, I will present studies on the availability of indoor-based energy and a batteryless smartcard designed for efficient operation using only indoor lighting to sense environmental data and emit Bluetooth beacons.

Key Learning Points:

• Energy Harvesting Technologies
• Maximum Power Point Tracking for Photovoltaic Cells
• Cold Start of Harvesting-Based Systems
• Low Power System Design
• Batteryless Sensing Systems

Go to Session

Developing Rugged IoT Devices by Integrating Off-the-Shelf Hardware

Presented by Sai Yamanoor

The availability of IoT hardware prototyping platforms (e.g.: an Arduino shield or the ‘Feather’ form factor) enables the construction of ‘quick and dirty’ prototypes. The effort required to convert the prototype into a product is of a higher magnitude. There are several factors to consider including product form factor, power options, target costs, etc. Problems in connected hardware development tend to surface either during the later phases of development or during pilot manufacturing of the product.

The talk is going to be structured around a hardware design example whose target unit cost is supposed to be sub $100. The following is discussed:

• Utilizing rapid prototyping tools like 3D printing and a low-cost 24-hour turn-around PCB fab to effectively iterate hardware design
• Criteria to consider for selecting off-the-shelf hardware for your product
• Estimating product development costs to avoid surprises in your budget
• Product lifecycle management including planning for hardware obsolescence & parts approaching end-of-life
• Things to consider while integrating off-the-shelf hardware & third-party tools e.g.: recurring costs for cloud services, platform fee, billing structure, etc.

Go to Session

Envisaging Real-time Industrial and Intelligent Applications leveraging the Edge AI Paradigm

Presented by Pethuru Raj Chelliah

There are a bevy of digital transformation technologies and tools emerging and evolving fast in the recent past in order to fulfill the digital life dream. The strategically imperative combination of edge/fog computing and artificial intelligence (AI) paradigms is being touted as the distinct innovation and disruption for enterprises to visualize and realize next-generation software services and applications. 

This talk is primarily intended to convey the latest trends and transitions happening in the edge/fog computing space, artificial intelligence(AI), the Internet of Things (IoT), block chain, digital twins, 5G, real-time and streaming data analytics, industry 4.5 vision, edge device monitoring, automated edge application deployment, governance, orchestration and management, etc.  There are AI chips exclusively to be used in edge devices to accelerate intelligent processing at edge. There are AI toolkits and platforms for facilitating edge intelligence. The proposed talk covers all these noteworthy developments in order to accentuate and articulate how the latest improvisations at chip, algorithm, and tool levels come handy in implementing and sustaining the relatively new field of edge AI. The talk also will cover the various industrial use cases such as smart factories, self-driving vehicles, robots-assisted surgeries, etc. out of the delectable technology clusters. 

Edge AI (alternatively termed as intelligent processing at edge, edge intelligence, intelligent edge or intelligence at edge) is being proclaimed as the next-generation technology paradigm to realize real-world intelligent applications and services for accelerating, augmenting and automating most of the industrial as well as people activities in real time. Capturing, cleansing and crunching edge device data in real-time can bring forth a number of real-time applications. Further on, with the increased usage of data analytics and AI toolkits in edge devices, business houses and commoners can gain hitherto unknown, sophisticated and situation-aware services. 

1. The prominent topics to be discussed include
2. Edge/Fog Computing
3. The formation of Edge Device Clusters/Clouds
4. Simplifying Edge Data Analytics
5. 5G for Edge Computing
6. The Emergence of AI-centric Chipsets
7. Running AI Algorithms on Edge Devices
8. Some Prominent Industry Use Cases using Edge AI

Go to Session

Attack IoT System Unknowns with Tracer Bullets

Presented by James Grenning

You have a product idea for using IoT technologies! You have an awesome greenfield development opportunity! Congratulation! You start staring at that clean sheet, that blank page. There are more unknowns than knowns! It's getting scary. What should you do?

This talk is the story of starting a greenfield IoT project. You see there are risks. You are attracted to the familiar tasks. If you defer risks, saving them until the end, it won't end well. In this story we'll look at identifying risks and attacking them directly with tracer bullets, building a walking skeleton of the product. The tracer bullets give you insight on a solution to your technical problem. The walking skeleton helps prove your product concept and demonstrate your ability to deliver it. You'll look at how to make progress with off-the-shelf hardware and software without becoming overly dependent on any vendor.

Go to Session

Using Visual Trace Diagnostics to Uncover Hidden Issues

Presented by Dr. Johan Kraft

Presented by Percepio

Embedded IoT software is often multi-threaded, event driven, and resource-constrained. This increases the difficulty of IoT software development and calls for better means for debugging, beyond the limited scope of traditional debugging tools.

Software event tracing can improve IoT software development by providing a timeline of the overall software activity. This is very useful information for debugging, profiling and for finding software design flaws. Tracing can be performed entirely in software and remain active over long durations. It can even be used in deployment.

However, tracing produces lots of repetitive data. When debugging, you need to find a small number of anomalies. This can be like searching for needles in a haystack.

The human brain is visual and excels at pattern recognition. Visual trace diagnostics makes it possible to leverage this capability to make the anomalies stand out. This builds on software event tracing, but refines the data into a rich, meaningful model and visual overviews that makes it easy to spot anomalies. The result is quick, intuitive comprehension, even for large traces. This approach is invaluable in system-level debugging and for finding software design flaws.

Join Dr. Johan Kraft as he explores visual trace diagnostics and presents real-world examples of its use in locating and solving hidden software issues.

Go to Session

Continuous Software Improvement with IoT Device Monitoring and Visual Trace Diagnostics

Presented by Dr. Johan Kraft

Presented by Percepio

Even the best developers cannot anticipate all the issues that may occur in deployed Internet of Things (IoT) devices: elusive software bugs, hardware glitches, algorithms that do not perform as well as intended, and user-interface issues. DevAlert enables continuous software improvement, so you can build flawless products through data-driven product improvements and the rapid resolution of remaining issues that would otherwise harm data integrity, device availability, battery life, and the overall user experience. Instead, IoT device monitoring and visual trace diagnostics adds an essential layer of defense to reduce support costs and increases sales revenue—and may even prevent accidents, product recalls and legal action.

Join Dr. Johan Kraft as he explores IoT device monitoring and remote diagnostics and demonstrates Percepio's DevAlert service.

Go to Session

Introduction to an Open Approach for Low-Power IoT Development

Presented by Reinhard Keil

Presented by Arm

Go to Session

Live Q&A - Using Visual Trace Diagnostics to Uncover Hidden Issues

Presented by Dr. Johan Kraft

Presented by Percepio

Go to Session

Live Q&A - Tips and Tricks for Developing Products using the ESP32

Presented by Jacob Beningo

Go to Session

Live Q&A - Easy CI/CD for Embedded IoT Devices

Presented by Dustin Reynolds

Go to Session

Live Q&A - Continuous Software Improvement with IoT Device Monitoring and Visual Trace Diagnostics

Presented by Dr. Johan Kraft

Presented by Percepio

Go to Session

Live Q&A - Envisaging Real-time Industrial and Intelligent Applications leveraging the Edge AI Paradigm

Presented by Pethuru Raj Chelliah

Go to Session

Live Q&A - Attack IoT System Unknowns with Tracer Bullets

Presented by James Grenning

Go to Session

Live Q&A - Best Practices for Designing IoT Edge Devices

Presented by Jacob Beningo

Go to Session

Live Q&A - Developing Rugged IoT Devices by Integrating Off-the-Shelf Hardware

Presented by Sai Yamanoor

Go to Session

Live Q&A - Hawkbit and SWUpdate for OTA Updates of Linux IoT Devices

Presented by Mohammed Billoo

Go to Session

Live Q&A - Batteryless Sensing Systems Powered by Light

Presented by Andres Gomez

Go to Session

Live Q&A - Introduction to an Open Approach for Low-Power IoT Development

Presented by Reinhard Keil

Presented by Arm

Go to Session