ConvXformer: Differentially Private Hybrid ConvNeXt-Transformer for Inertial Navigation

TL;DR

ConvXformer is a novel hybrid neural network architecture that combines ConvNeXt and Transformer components, integrated with an advanced differential privacy mechanism, to enhance inertial navigation accuracy while safeguarding sensitive training data.

Contribution

This work introduces ConvXformer, a hybrid model with a new differential privacy method using adaptive gradient clipping and GANI, improving privacy-utility trade-off in inertial navigation.

Findings

Achieves over 40% improvement in positioning accuracy compared to state-of-the-art methods.

Demonstrates robustness in environments with severe sensor perturbations.

Ensures $(psilon,elta)$-differential privacy guarantees.

Abstract

Data-driven inertial sequence learning has revolutionized navigation in GPS-denied environments, offering superior odometric resolution compared to traditional Bayesian methods. However, deep learning-based inertial tracking systems remain vulnerable to privacy breaches that can expose sensitive training data. \hl{Existing differential privacy solutions often compromise model performance by introducing excessive noise, particularly in high-frequency inertial measurements.} In this article, we propose ConvXformer, a hybrid architecture that fuses ConvNeXt blocks with Transformer encoders in a hierarchical structure for robust inertial navigation. We propose an efficient differential privacy mechanism incorporating adaptive gradient clipping and gradient-aligned noise injection (GANI) to protect sensitive information while ensuring model performance. Our framework leverages truncated singular value decomposition for gradient processing, enabling precise control over the privacy-utility trade-off. Comprehensive performance evaluations on benchmark datasets (OxIOD, RIDI, RoNIN) demonstrate that ConvXformer surpasses state-of-the-art methods, achieving more than 40% improvement in positioning accuracy while ensuring $(\epsilon,\delta)$-differential privacy guarantees. To validate real-world performance, we introduce the Mech-IO dataset, collected from the mechanical engineering building at KAIST, where intense magnetic fields from industrial equipment induce significant sensor perturbations. This demonstrated robustness under severe environmental distortions makes our framework well-suited for secure and intelligent navigation in cyber-physical systems.