Smart devices are different: Assessing and mitigating mobile sensing heterogeneities for activity recognition

TY - GEN

T1 - Smart devices are different

T2 - ACM Conference on Embedded Networked Sensor Systems

AU - Stisen, Allan

AU - Blunck, Henrik

AU - Bhattacharya, Sourav

AU - Prentow, Thor Siiger

AU - Kjærgaard, Mikkel Baun

AU - Dey, Anind

AU - Sonne, Tobias

AU - Jensen, Mads Møller

N1 - Conference code: 13

PY - 2015

Y1 - 2015

N2 - 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.

AB - 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.

KW - Mobile Sensing

KW - Activity Recognition

M3 - Article in proceedings

BT - Proceedings of the 13th ACM Conference on Embedded Network Sensor Systems

Y2 - 1 November 2015 through 4 November 2015

ER -