Embedded Wearable Integrating Real-Time Processing of Electromyography Signals

Gentile, Paolo; Pessione, Marco; Suppa, Antonio; Zampogna, Alessandro; Irrera, Fernanda

doi:10.3390/proceedings1040600

Cited by 7 publications

(8 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first method is the classical detection method which compares the rectified and filtered sEMG signal with a predefined threshold, where this method was used in most of the practical implemented hand robotic devices [16][17][18][19][20][21][22][23][24][25]. The second method is the Teager Kaiser Energy Operator (TKE) [7][8][9][10]26] and the third method is the Integrated Profile (IP) [13,27] of the sEMG signal. These three methods were chosen because they can be implemented in real time and have low computational efforts.…”

Section: Performance Comparison Between the Fla-mse Algorithm And Thrmentioning

confidence: 99%

Amplitude independent versus amplitude dependent muscle activity detection algorithms: a comparative study

Hameed

Hasan

Shafie

et al. 2020

IJEECS

View full text Add to dashboard Cite

The amplitude dependent muscle activity detection algorithms of the surface electromyography (sEMG) signals are very sensitive to the changes in the background noise levels and the performance of these amplitude-based methods is highly deteriorated when the Signal to Noise ratio (SNR) of the sEMG signal is low. sEMG signals of deep and small muscles as well as sEMG signals recorded from patients that have neuromuscular diseases may not meet this SNR requirement which motivates the need for amplitude independent algorithms that can detect weak muscle activities. Moreover, the sEMG signal amplitude is not constant during the recording time due to the variation in the characteristics of the electrode-skin interface and due to the changes in the ground reference level. Therefore, the performance of the muscle activity detection algorithms should not be affected by the involuntary amplitude variations of the sEMG signal in order to achieve reliable control of robotic devices intended for disabled people. To accentuate the importance of the amplitude independent muscle activity detection methods over the amplitude dependent detection methods, a comparative study has been conducted in this paper between the performance of an amplitude independent muscle activity detection algorithm (FLA-MSE algorithm) and three amplitude dependent algorithms with respect to the detection capability of weak muscle activities and with respect to the immunity against false alarms. The results have showed that the performance of the amplitude independent algorithm outperformed the performance of the amplitude dependent algorithms for detecting weak muscle activities and for robustness against false alarms.

show abstract

Section: Performance Comparison Between the Fla-mse Algorithm And Thrmentioning

confidence: 99%

Amplitude independent versus amplitude dependent muscle activity detection algorithms: a comparative study

Hameed

Hasan

Shafie

et al. 2020

IJEECS

View full text Add to dashboard Cite

show abstract

“…It is widely used in EMG processing, for instance in [33] where a complete gesture recognition system on a ARM cortex-M4 microcontroller is implemented with a SVM used to recognize up to 7 hand gestures with a 8 channel EMG setup. Other works based on this architecture are presented in [36] and in [37] where an ANN with one hidden layer is implemented on a 4 EMG channel setup and it is capable to recognize up to 10 gestures with accuracy higher than 80%. Other works rely on more powerful platforms, to implement more complex models, like the one presented in [34], which leverages an ARM Cortex-A8 processor to execute the EMG gesture recognition using non linear modeling approach.…”

Section: Related Workmentioning

confidence: 99%

Online Learning and Classification of EMG-Based Gestures on a Parallel Ultra-Low Power Platform Using Hyperdimensional Computing

Benatti

Montagna

Kartsch

et al. 2019

IEEE Trans. Biomed. Circuits Syst.

View full text Add to dashboard Cite

This work presents a wearable EMG gesture recognition system based on the hyperdimensional (HD) computing paradigm, running on a programmable Parallel Ultra-Low-Power (PULP) platform. The processing chain includes efficient on-chip training, which leads to a fully embedded implementation with no need to perform any offline training on a personal computer. The proposed solution has been tested on 10 subjects in a typical gesture recognition scenario achieving 85% average accuracy on 11 gestures recognition, which is aligned with the State-Of-the-Art (SoA), with the unique capability of performing online learning. Furthermore, by virtue of the Hardware (HW) friendly algorithm and of the efficient PULP System-on-Chip (SoC) (Mr. Wolf) used for prototyping and evaluation, the energy budget required to run the learning part with 11 gestures is 10.04mJ, and 83.2µJ per classification. The system works with a average power consumption of 10.4mW in classification, ensuring around 29h of autonomy with a 100mAh battery. Finally, the scalability of the system is explored by increasing the number of channels (up-to 256 electrodes), demonstrating the suitability of our approach as universal, energy-efficient biopotential wearable recognition framework.

show abstract

“…These results are based on the values summarized in Table II, where we also show the current consumption of the system in streaming mode, with up to 9 h of autonomy, and sleep/standby (up to 1000 h). While it is difficult to compare wearable systems directly, it is still noticeable that SoA systems for EMG gesture recognition have a battery life ranging from 3 to 11h [21], [22], [13], independently from the algorithm that is used. As explained above, our architecture is capable of providing around 2x more autonomy with a tiny 60 mAh battery, offering superior performance and unintrusive form factor.…”

Section: Power Consumptionmentioning

confidence: 99%

“…More efficient solutions, such as [13] and [14] are based on dedicated industrial IoT microcontrollers (i.e. ARM CORTEX M4) and provide up to 10 hours with a 100 mAh LiPo battery.…”

Section: Introductionmentioning

confidence: 99%

An Energy-Efficient IoT node for HMI applications based on an ultra-low power Multicore Processor

Kartsch

Guermandi

Benatti

et al. 2019

2019 IEEE Sensors Applications Symposium (SAS)

View full text Add to dashboard Cite

Developing wearable sensing technologies and unobtrusive devices is paving the way to the design of compelling applications for the next generation of systems for a smart IoT node for Human Machine Interaction (HMI). In this paper we present a smart sensor node for IoT and HMI based on a programmable Parallel Ultra-Low-Power (PULP) platform. We tested the system on a hand gesture recognition application, which is a preferred way of interaction in HMI design. A wearable armband with 8 EMG sensors is controlled by our IoT node, running a machine learning algorithm in real-time, recognizing up to 11 gestures with a power envelope of 11.84 mW. As a result, the proposed approach is capable to 35 hours of continuous operation and 1000 hours in standby. The resulting platform minimizes effectively the power required to run the software application and thus, it allows more power budget for high-quality AFE.

show abstract

Embedded Wearable Integrating Real-Time Processing of Electromyography Signals

Cited by 7 publications

References 4 publications

Amplitude independent versus amplitude dependent muscle activity detection algorithms: a comparative study

Amplitude independent versus amplitude dependent muscle activity detection algorithms: a comparative study

Online Learning and Classification of EMG-Based Gestures on a Parallel Ultra-Low Power Platform Using Hyperdimensional Computing

An Energy-Efficient IoT node for HMI applications based on an ultra-low power Multicore Processor

Contact Info

Product

Resources

About