With this AI-based approach, IoT devices can be 40% more energy efficient

With this AI-based approach, IoT devices can be 40% more energy efficient

How does it work: Backscattering allows devices to modulate incoming wireless signals and reflect them back to the transmitter, much like RFID chips and contactless payment cards work by harvesting energy from the reader. This enables IoT devices to achieve battery-free communication.

Now, a research team from Pusan ​​National University in South Korea has developed a new method around the same technology that is about 40 percent more energy efficient than existing techniques.

The concept of backscattering is nothing new. But the team’s innovation is using AI to optimize the system and make it significantly more efficient for low-power applications like IoT sensors. Their approach involves using machine learning to precisely model the optimal “reflection coefficients,” which determine how much of the wireless signal gets reflected.

Traditionally, calculating these coefficients relied on simulations that didn’t perfectly match real-world conditions, making it difficult to achieve low bit error rates and high data rates. However, the researchers overcame this by using a technique called “transfer learning,” where an AI model is first trained on a task and then improved using data from the real target task.

To make it work, they pre-trained an artificial neural network on simulated input voltages to understand the behavior of the modulation circuit under different voltage conditions. They then further trained the pre-trained model using real experimental data, allowing it to accurately predict reflection coefficients for their particular hardware.

With these fine-tuned AI models, the team was able to optimize 4-QAM and 16-QAM modulation schemes for maximum efficiency. QAM is short for Quadrature Amplitude Modulation, a scheme commonly used in Wi-Fi communication systems. Their resulting prototype system consumes less than 0.6 milliwatts during transmission—a fraction of the power required for traditional wireless radios.

The system also features a 2×2 MIMO antenna setup to improve signal reception. When tested in the 5.7-5.8 GHz range, a spectral efficiency of 2 bits/second/hertz was achieved using 4-QAM modulation.

“The combination of accurate circuit modeling, advanced modulation techniques, and polarization diversity, all tested in wireless environments, offers a holistic approach to address challenges in ISC and IoT,” said Professor Sangkil Kim, who led the study.

This lays the groundwork for reliable, ultra-low-power backscatter systems with potential applications in areas such as consumer electronics, health monitoring, smart urban infrastructure, environmental sensing and radar communications, according to the Pusan ​​researchers.

Their findings are as follows: published In an article published in the IEEE Internet of Things Magazine.