Embedded systems are continuing their technical evolution at an ever-faster pace; devices in our homes (Figure 1), vehicles, and workplaces are advancing in their capabilities by leaps and bounds. A key driver of this advancement is the ability for even the smallest electronic devices to connect to our modern network infrastructure.
Figure 1. Wirelessly connected embedded sensor systems are proliferating in our homes. Image used courtesy of Adobe Stock
Wi-Fi, Bluetooth, and other connectivity options enable in-field updates and easier maintenance while adding the benefits of AI and Machine Learning algorithms. This increased connectivity effectively makes these devices IoT edge nodes—but that comes at the cost of increased processing requirements and larger memory subsystems.
IoT Edge Sensor System Challenges
Most embedded systems are also “connected” to their immediate environment. That is, they offer facilities for some type of environmental sensing, mechanical actuation, or human interface.
For example, smart thermostats are connected to local networks of temperature and humidity sensors in addition to housing an array of buttons or capacitive sensors for human input. And, of course, a connected cooking appliance’s primary objective is to understand your desires for food temperature and translate them into precisely applied amounts of heat.
These primarily “analog” systems are making their way into the increasingly fast-paced world of cloud communication, which sets up a unique problem: do you tailor your system for the slow-moving input of the analog world, or do you compromise analog fidelity for the sake of speed and increased overall functionality?
To explore the problem in depth, we’ll look at a ubiquitous and simple example of this type of application—the IoT edge sensor node.
Sleepy, Low-Power Analog Subsystems
IoT edge sensor nodes need some analog subsystem to measure and monitor environmental conditions like temperature, humidity, or motion. Analog subsystems include a microcontroller (MCU) that will read the sensor data, process it, and communicate it over a network.
Typically, environmental data is slowly changing, so most edge nodes don’t need to process a continuous, uninterrupted stream of data (Figure 2). As an edge node often operates on the same small form-factor battery for several years, it spends most of the time in a low-power “sleep” mode and only wakes up periodically to detect changes in the environment.
Figure 2. Most environmental sensors monitor slowly varying signals like temperature. Image used courtesy of Adobe Stock
During the waking period, the node gathers data and transmits it across a network. Then, it goes back to sleep until it needs to take the next measurement.
As the number of edge nodes and data collected increases in our hyper-connected world, power efficiency and low power operation are vital design considerations to extend battery life in analog subsystems.
Segmenting Embedded Systems for Improved Efficiency
For embedded systems, it is best to segment the system into different speed domains utilizing a bridge to connect the fast main processor to analog sub-systems. Partitioning allows the analog subsystem to focus on handling slow-changing tasks while fast, compute-intensive processing tasks are managed with a fast main processor, thereby maximizing the functional strength of each processor type.
With the growing trend of more connected devices, I3C® is the next-generation serial communication interface to support high-speed chip-to-chip communication (Figure 3). As a successor to I2C, it is better suited for future applications with a faster, smarter interface and sophisticated control capabilities.
Figure 3. An example I3C connection between a controller and two target sensor nodes. Image used courtesy of Microchip
I3C maintains backward compatibility with I2C devices, which is essential to easing the adoption of I3C into existing hardware platforms. Also, I2C devices can coexist with I3C controllers operating at 12.5 MHz, enabling the migration of existing I2C-bus designs to the I3C specification.
For instance, a microcontroller that supports both I3C and a legacy communication interface (such as I2C, SPI, or UART) can serve as a bridge device. This bridge connects a fast processor to a sensor via the microcontroller. The microcontroller measures the sensor input, calculates results, and efficiently transfers the data.
This setup maintains the integrity and speed of the I3C bus while enabling communication between the I3C controller and I2C/SPI devices through the microcontroller. By partitioning embedded systems and leveraging I3C, it becomes possible to implement system designs successfully and robustly.
PIC18-Q20 MCU
Microchip has developed the PIC18-Q20 product family, as shown in Figure 4, specifically for modern distributed processor embedded systems. These MCUs offer advanced serial communication interfaces, including up to two I3C peripherals, for high-speed connectivity to multiple buses, enhancing flexibility.
Figure 4. Microchip PIC18F16Q20 microcontroller. Image used courtesy of Microchip
Additionally, they come equipped with built-in legacy communication protocols such as UART, SPI, I2C, and SMBus, enabling seamless integration as a bridge device and isolation of I2C/SPI client devices from a pure I3C bus. This setup maintains the speed of the I3C bus while allowing an I3C controller to communicate with I2C/SPI devices through the microcontroller.
Moreover, the PIC18-Q20 supports multiple voltage domains, so it can readily connect to various components with different operating voltage levels. As illustrated in Figure 5, this eliminates the need for level shifters, reducing the Bill of Material cost and simplifying the system design.
Figure 5. Microcontrollers that support multi-voltage I/O can eliminate the need for external level shifters. Image used courtesy of Microchip
PIC18-Q20 MCUs also include on-chip Core Independent Peripherals (CIPs) that can operate without constant interaction from the CPU and communicate directly with other peripherals (Figure 6). These hardware-based peripherals consume minimal power and require little to no code and less RAM and Flash memory to implement the same functions in software. Additionally, many simultaneous functions can be enabled in a single MCU.
Figure 6. Core Independent Peripherals (CIPs) and integrated analog features reduce CPU processing load and power. Image used courtesy of Microchip
Designers can easily customize combinations of CIPs, including the I3C peripheral using the MPLAB® Code Configurator (MCC), a simple Graphical User Interface (GUI) environment, to generate application code without reading through datasheets. With (CIPs), engineers can partition each system task for easier function management, reducing component count, code size, development time, and power consumption.
To Learn More
In our fast-changing world, technological innovations and advancements demand faster processing speeds, faster connectivity, and miniaturization. While modern electronics are increasingly connected to our outside world, small-scale and energy-efficient analog subsystems are needed to sense and measure the ‘real world’ in connected systems. As environmental data changes usually occur gradually, design goals are in opposing directions.
Efficient embedded systems are achieved by partitioning the system into different speed domains, using a bridge to connect the fast processor to the surrounding parts of the system where analog sub-systems exist. With I3C becoming the de facto interface for high-speed chip-to-chip communication, it’s important for engineers to select advanced MCUs that are equipped to fully support the increasing requirements for high performance in the digital realm while maintaining high analog precision for next-generation designs.
To learn more about the PIC18-Q20 family of microcontrollers, visit https://www.microchip.com/Q20.
Feature image background used courtesy of Adobe Stock.
Industry Articles are a form of content that allows industry partners to share useful news, messages, and technology with All About Circuits readers in a way editorial content is not well suited to. All Industry Articles are subject to strict editorial guidelines with the intention of offering readers useful news, technical expertise, or stories. The viewpoints and opinions expressed in Industry Articles are those of the partner and not necessarily those of All About Circuits or its writers.
