Low Power IoT Architectures: Principles and Practices

By Przemyslaw Bakowski

In this book, we present a structured approach to building low-power IoT architectures by organizing the essential concepts and available IoT components.
These architectures are built using well-known System-on-Chip (SoC) and System-on-Module (SoM) solutions, specifically the ESP32-C3 and ASR6501, which are integrated into corresponding IoT boards.
We introduce a simple classification of IoT terminals based on how they connect to the Internet infrastructure:
Direct Terminals, which operate directly using the IP protocol
Close Terminals, which communicate over the MAC/Wi-Fi layer
Remote Terminals, which rely on long-range radio links such as LoRa
To connect to the Internet infrastructure, both Close Terminals and Remote Terminals require the use of Gateway nodes.
The connection between terminals and target IoT servers or brokers is established through the IoT socket concept. An IoT socket is composed of three elements: an IP address, a port number, and a channel or topic identifier. The channel typically refers to a ThingSpeak server channel and its fields, while the topic designates an entry point to an MQTT broker.

Based on this architecture, we propose and implement low-power IoT protocols that enable adaptive behavior greatly reducing power consumption. These protocols leverage the energy-saving mechanisms provided by the selected SoC and SoM platforms, primarily focusing on deep sleep modes supported by low-power kernels and memory.
The deep sleep mode, along with associated low-power and non-volatile memory features, enables the implementation of two key strategies:
Keeping the terminal in a low-power (deep sleep) state for as long as possible
Transmitting data in a high-power state as infrequently as possible
Implementing these strategies requires the use of parameters stored in non-volatile memory, as well as meta-data preserved in low-power memory.
In the book, we progressively develop these mechanisms, starting with MicroPython (Thonny IDE) on the ESP32-C3, and then moving to the ASR6501 (CubeCell) platform using C/C++ with the Arduino IDE.
We also include an analysis of power consumption using the Power Profiler Kit II from Nordic Semiconductor.
To complement the text, we provide a complete code and design repository on GitHub (github.com). The repository includes all MicroPython and C/C++ source code and PCB Gerber files that may be useful to build the full hardware platforms.

• Format: Paperback
• Language: English
• ISBN: B0FLPSRRXT
• ISBN13: 9798296981684
• Release Date: August 2025
• Publisher: Independently Published
• Length: 150 Pages
• Weight: 0.62 lbs.
• Dimensions: 0.3" x 7.4" x 9.7"