Toward Ultra-Low-Power Remote Health Monitoring: An Optimal and Adaptive Compressed Sensing Framework for Activity Recognition

Pagán, Josué; Fallahzadeh, Ramin; Pedram, Mahdi; Risco‐Martín, José L.; Moya, José Manuel Huidobro; Ayala, José L.; Ghasemzadeh, Hassan

doi:10.1109/tmc.2018.2843373

Cited by 42 publications

(21 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Yan et al [25] specifically selected sampling frequencies based on a formalized trade-off between activity classification and accuracy. Pagan et al [15] and Fallahzadeh et al [26] incorporate insights about activity-specific sensing granularity as well as compressive sensing to enhance this trade-off.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Easing Power Consumption of Wearable Activity Monitoring with Change Point Detection

Culman

Aminikhanghahi

Cook

2020

Sensors

View full text Add to dashboard Cite

Continuous monitoring of complex activities is valuable for understanding human behavior and providing activity-aware services. At the same time, recognizing these activities requires both movement and location information that can quickly drain batteries on wearable devices. In this paper, we introduce Change Point-based Activity Monitoring (CPAM), an energy-efficient strategy for recognizing and monitoring a range of simple and complex activities in real time. CPAM employs unsupervised change point detection to detect likely activity transition times. By adapting the sampling rate at each change point, CPAM reduces energy consumption by 74.64% while retaining the activity recognition performance of continuous sampling. We validate our approach using smartwatch data collected and labeled by 66 subjects. Results indicate that change point detection techniques can be effective for reducing the energy footprint of sensor-based mobile applications and that automated activity labels can be used to estimate sensor values between sampling periods.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Energy consumption is a known obstacle to wearable computing in general and to activity monitoring in particular [11][12][13][14][15]. For complex activities, however, recognition and monitoring may require an even greater energy footprint.…”

Section: Introductionmentioning

confidence: 99%

Easing Power Consumption of Wearable Activity Monitoring with Change Point Detection

Culman

Aminikhanghahi

Cook

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Because of the energy efficiency issue in WBANs, power-aware techniques, such as power-efficient communication technology solutions, as well as other methods that can improve energy efficiency, are highly desirable in WBANs for sustainable operations, including meeting the crucial requirements of WBAN systems in HCM. For instance, the authors of [31] proposed an ultra-low power activity recognition system for remote health monitoring, using a metaheuristic optimization scheme that is based on a Grammatical Evolution technique to save energy during data transmission.…”

Section: Overview and Concept Of Wban Solutions In Hcmmentioning

confidence: 99%

A Survey on LPWAN Technologies in WBAN for Remote Health-Care Monitoring

Olatinwo

Abu‐Mahfouz

Hancke

2019

Sensors

View full text Add to dashboard Cite

In ubiquitous health-care monitoring (HCM), wireless body area networks (WBANs) are envisioned as appealing solutions that may offer reliable methods for real-time monitoring of patients’ health conditions by employing the emerging communication technologies. This paper therefore focuses more on the state-of-the-art wireless communication systems that can be explored in the next-generation WBAN solutions for HCM. Also, this study addressed the critical issues confronted by the existing WBANs that are employed in HCM. Examples of such issues include wide-range health data communication constraint, health data delivery reliability concern, and energy efficiency, which are attributed to the limitations of the legacy short range, medium range, and the cellular technologies that are typically employed in WBAN systems. Since the WBAN sensor devices are usually configured with a finite battery power, they often get drained during prolonged operations. This phenomenon is technically exacerbated by the fact that the legacy communication systems, such as ZigBee, Bluetooth, 6LoWPAN, and so on, consume more energy during data communications. This unfortunate situation offers a scope for employing suitable communication systems identified in this study to improve the productivity of WBANs in HCM. For this to be achieved, the emerging communication systems such as the low-power wide-area networks (LPWANs) are investigated in this study based on their power transmission, data transmission rate, data reliability in the context of efficient data delivery, communication coverage, and latency, including their advantages, as well as disadvantages. As a consequence, the LPWAN solutions are presented for WBAN systems in remote HCM. Furthermore, this research work also points out future directions for the realization of the next-generation of WBANs, as well as how to improve the identified communication systems, to further enhance their productivity in WBAN solutions for HCM.

show abstract

“…As the orientation of the sensor is generally unknown, this contribution needs some specific signal processing techniques to be removed [8]. In this context, the delay of the data transfer and the availability of connections make cloud unusable when real-time operation is required [9], let alone when privacy is at stake. Instead, processing "at the edge", namely locally on the device and close to the sensors is highly desirable [10].…”

Section: Introductionmentioning

confidence: 99%

Low Power Tiny Binary Neural Network with improved accuracy in Human Recognition Systems

Vita

Pau

Benedetto

et al. 2020

2020 23rd Euromicro Conference on Digital System Design (DSD)

View full text Add to dashboard Cite

Human Activity Recognition requires very high accuracy to be effectively employed into practical applications, ranging from elderly care to microsurgical devices. The highest accuracies are achieved by Deep Learning models, but these are not easily deployable in handheld or wearable devices with very constrained resources. We therefore present a new HAR system suitable for a compact FPGA implementation. A new Binarized Neural Network (BNN) architecture achieves the classification based on data from a single tri-axial accelerometer. From our experiments, the effect of gravity and the unknown orientation of the sensor cause a degradation of the accuracy. In order to compensate for these issues, we propose a HW-friendly algorithm to pre-process the raw acceleration signal. Moreover, the very low power and hardware friendly BNN has been trained and validated on the PAMAP2 dataset, for which the pre-processing operations increase the accuracy from 51% to 99% in the best case. Aiming for a low-power design, we designed both a custom circuit to perform the pre-processing operations and a hardware accelerator for the BNN. The design on FPGA features a power dissipation of 72 mW and occupies 6788 LUTs.

show abstract

Toward Ultra-Low-Power Remote Health Monitoring: An Optimal and Adaptive Compressed Sensing Framework for Activity Recognition

Cited by 42 publications

References 42 publications

Easing Power Consumption of Wearable Activity Monitoring with Change Point Detection

Easing Power Consumption of Wearable Activity Monitoring with Change Point Detection

A Survey on LPWAN Technologies in WBAN for Remote Health-Care Monitoring

Low Power Tiny Binary Neural Network with improved accuracy in Human Recognition Systems

Contact Info

Product

Resources

About