HEROS-GAN: Honed-Energy Regularized and Optimal Supervised GAN for Enhancing Accuracy and Range of Low-Cost Accelerometers

TL;DR

This paper addresses the challenge of extending the accuracy and dynamic range of low-cost accelerometers by learning to map their signals to high-cost equivalents without frame-level paired data. The authors introduce HEROS-GAN, which combines Optimal Transport Supervision (OTS) to exploit similarities in unpaired high- and low-cost signals and Modulated Laplace Energy (MLE) to enrich local signal details. A new dataset, LASED, is released to benchmark accelerometer signal enhancement, and two metrics CSRE and ZVRE are proposed for evaluation. Experimental results show that OTS or MLE alone surpass prior methods by large margins, while the full HEROS-GAN achieves a doubling of range and two-orders-of-magnitude noise reduction, establishing a new benchmark for accelerometer signal processing.

Abstract

Low-cost accelerometers play a crucial role in modern society due to their advantages of small size, ease of integration, wearability, and mass production, making them widely applicable in automotive systems, aerospace, and wearable technology. However, this widely used sensor suffers from severe accuracy and range limitations. To this end, we propose a honed-energy regularized and optimal supervised GAN (HEROS-GAN), which transforms low-cost sensor signals into high-cost equivalents, thereby overcoming the precision and range limitations of low-cost accelerometers. Due to the lack of frame-level paired low-cost and high-cost signals for training, we propose an Optimal Transport Supervision (OTS), which leverages optimal transport theory to explore potential consistency between unpaired data, thereby maximizing supervisory information. Moreover, we propose a Modulated Laplace Energy (MLE), which injects appropriate energy into the generator to encourage it to break range limitations, enhance local changes, and enrich signal details. Given the absence of a dedicated dataset, we specifically establish a Low-cost Accelerometer Signal Enhancement Dataset (LASED) containing tens of thousands of samples, which is the first dataset serving to improve the accuracy and range of accelerometers and is released in Github. Experimental results demonstrate that a GAN combined with either OTS or MLE alone can surpass the previous signal enhancement SOTA methods by an order of magnitude. Integrating both OTS and MLE, the HEROS-GAN achieves remarkable results, which doubles the accelerometer range while reducing signal noise by two orders of magnitude, establishing a benchmark in the accelerometer signal processing.

Figures (5)

• Figure 1: Architecture of the HEROS-GAN. MLE (orange) and OTS (green) are applied to feature interaction on both sides.
• Figure 2: Illustration of the Optimal Transport Supervision.
• Figure 3: Diagram of the Modulated Laplace Energy. The MLE computes Laplace energy from feature volatility and applies modulation through a regularization term, thereby controlling the volatility by adjusting the energy of the feature layers.
• Figure 4: Visualization of CSRE for different methods as the clipping threshold $\tau$ decreases from $15g$ to $6g$.
• Figure 5: Visualization of the ZVRE for different methods and the raw low-cost signal across the x, y, and z axes.