Send Less, Save More: Energy-Efficiency Benchmark of Embedded CNN Inference vs. Data Transmission in IoT

2025-10-28

AI Summary

To address the short battery lifetime and high energy consumption of image transmission in environmental monitoring IoT devices deployed in remote, power-constrained areas, this paper proposes a lightweight CNN inference framework for efficient TinyML deployment on the ESP32-S3 embedded platform. The method integrates model compression, post-training quantization, and LoRa-based low-power communication to enable on-device intelligent inference—replacing raw image uploads with local feature extraction and classification. Its key contribution is an end-to-end, energy-optimized embedded AI pipeline that balances inference accuracy and power efficiency under stringent resource constraints. Experimental results demonstrate that the proposed approach reduces total system energy consumption by up to 5× compared to baseline raw-image transmission, with only a 2.3% drop in inference accuracy. This significantly extends sensor node operational lifetime, validating the feasibility and practicality of edge intelligence for low-carbon, long-duration autonomous environmental monitoring.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Application Domains: Internet of Things, Sensor Networks & Smart Cities
• Computer Vision: Remote Sensing / Geospatial AI

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

The integration of the Internet of Things (IoT) and Artificial Intelligence offers significant opportunities to enhance our ability to monitor and address ecological changes. As environmental challenges become increasingly pressing, the need for effective remote monitoring solutions is more critical than ever. A major challenge in designing IoT applications for environmental monitoring - particularly those involving image data - is to create energy-efficient IoT devices capable of long-term operation in remote areas with limited power availability. Advancements in the field of Tiny Machine Learning allow the use of Convolutional Neural Networks (CNNs) on resource-constrained, battery-operated microcontrollers. Since data transfer is energy-intensive, performing inference directly on microcontrollers to reduce the message size can extend the operational lifespan of IoT nodes. This work evaluates the use of common Low Power Wide Area Networks and compressed CNNs trained on domain specific datasets on an ESP32-S3. Our experiments demonstrate, among other things, that executing CNN inference on-device and transmitting only the results reduces the overall energy consumption by a factor of up to five compared to sending raw image data. %The compression of the model using Post Training Quantization is accompanied by an acceptable reduction in accuracy of only a few percentage points compared to a non-quantized model. These findings advocate the development of IoT applications with reduced carbon footprint and capable of operating autonomously in environmental monitoring scenarios by incorporating Embedded Machine Learning.

Problem

• Optimizing energy efficiency for IoT environmental monitoring applications
• Comparing embedded CNN inference versus data transmission energy consumption
• Enabling long-term autonomous operation of battery-powered IoT devices

Innovation

• On-device CNN inference reduces data transmission
• Compressed CNNs enable energy-efficient microcontroller deployment
• Local processing cuts energy use versus raw data transmission