Smart Devices are Different

TLDR

The widespread use of motion sensors on personal mobile devices has spurred research in human activity recognition, but large‑scale deployment often yields lower performance due to hardware and operating‑system heterogeneity. The study aims to systematically evaluate how sensor, device, and workload heterogeneities affect human activity recognition and to propose a clustering‑based mitigation technique for large‑scale deployment. The authors used 36 smartphones and smartwatches from 13 models across four manufacturers, performed experiments with nine users, and examined feature representations and classification methods to assess heterogeneity effects and develop a clustering‑based mitigation strategy. The study found that sensor and handling heterogeneities significantly degrade HAR performance, with variations across devices and recognition techniques, and that a clustering‑based mitigation approach can alleviate these effects.

Abstract

The widespread presence of motion sensors on users' personal mobile devices has spawned a growing research interest in human activity recognition (HAR). However, when deployed at a large-scale, e.g., on multiple devices, the performance of a HAR system is often significantly lower than in reported research results. This is due to variations in training and test device hardware and their operating system characteristics among others. In this paper, we systematically investigate sensor-, device- and workload-specific heterogeneities using 36 smartphones and smartwatches, consisting of 13 different device models from four manufacturers. Furthermore, we conduct experiments with nine users and investigate popular feature representation and classification techniques in HAR research. Our results indicate that on-device sensor and sensor handling heterogeneities impair HAR performances significantly. Moreover, the impairments vary significantly across devices and depends on the type of recognition technique used. We systematically evaluate the effect of mobile sensing heterogeneities on HAR and propose a novel clustering-based mitigation technique suitable for large-scale deployment of HAR, where heterogeneity of devices and their usage scenarios are intrinsic.