On Quasi-Isometry of Threshold-Based Sampling

Moser, Bernhard; Lunglmayr, Michael

doi:10.1109/tsp.2019.2919415

Cited by 11 publications

(8 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It augments the computational complexity while comparing with the suggested solution. The adaptive rate signal collection, in contrast to the conventional approach, implicitly gives information about the signal frequency contents in time-domain [23,28]. It enables adequate feature mining to be realized only through time-domain analysis.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning with Adaptive Rate Processing for Power Quality Disturbances Identification

et al. 2021

View full text Add to dashboard Cite

Power quality disturbances (PQDs) commonly occur in large-scale power systems and networks that rise critical issues. Therefore, an automated recognition and mitigation of PQDs is necessary. In this scenario, an efficient adaptive rate technique is proposed for viable features mining and recognition of PQDs. Event-driven A/D converters (EDADCs) are used to acquire PQD signals. To accurately segment the sampled signal, an appealing approach is used. A time-domain evaluation is carried out in the next phase to investigate the characteristics of these fragments. The mature machine learning algorithms are employed to carry out the classification. Compared to conventional counterparts, the findings indicate a decrease of 13.26 times in collected information. An average maximum identification precision of 99.33% is achieved by the proposed method. Compared to predecessors, this confirms the considerable performance of the processing and power usage of the engineered solution while achieving high recognition accuracy. Keywords Adaptive rate sampling • Time-domain features extraction • Power quality disturbances • Compression ratio • ClassificationThis article is part of the topical collection "Intelligent Systems" guest edited by

show abstract

Section: Resultsmentioning

confidence: 99%

“…The ASA is a novel approach which permits to exploit the important information of the signal without any computationally complex transformation. Therefore, in the studied case, the classifiable features are mined directly in time-domain without involving any complex transformation scheme [23].…”

Section: Features Extractionmentioning

confidence: 99%

Machine Learning with Adaptive Rate Processing for Power Quality Disturbances Identification

et al. 2021

View full text Add to dashboard Cite

show abstract

“…1. This can be explained by a limited use of the analytical information contained in (5). An early observation is that little is done in these methods to make the bandlimited estimates consistent with this information.…”

Section: Reconstruction By Pocsmentioning

confidence: 99%

“…1). Time encoding is attracting more and more interest in data acquisition [1], [2], [3], [4], [5], [6], [7], [8], [7], [9], [10] as the downscaling of semiconductor integration is increasing time precision while resulting in less amplitude accuracy [11], [12], [13]. This is also part of the trend on event-based sampling [14] with more general objectives such as having acquisition activity dependent on input activity for power efficiency.…”

Section: Introductionmentioning

confidence: 99%

Bandlimited signal reconstruction from leaky integrate-and-fire encoding using POCS

Thảo¹,

Rzepka²,

Miśkowicz³

2022

Preprint

View full text Add to dashboard Cite

Leaky integrate-and-fire (LIF) encoding is a model of neuron transfer function in biology that has recently attracted the attention of the signal processing and neuromorphic computing communities as a technique of event-based sampling for data acquisition. While LIF enables the implementation of analog-circuit signal samplers of lower complexity and higher accuracy simultaneously, the core difficulty of this technique is the retrieval of an input from its LIF-encoded output. This theoretically requires to perform the pseudo-inversion of a linear but time-varying operator of virtually infinite size. In the context of bandlimited inputs to allow finite-rate sampling, we show two fundamental contributions of the method of projection onto convex sets (POCS) to this problem: (i) single iterations of the POCS method can be used to deterministically improve input estimates from any other reconstruction method; (ii) the iteration limit of the POCS method is the pseudo-inverse of the above mentioned operator in all conditions, whether reconstruction is unique or not and whether the encoding is corrupted by noise or not. The algorithms available until now converge only under particular situations of unique of reconstruction.

show abstract

“…It captures and processes redundant samples that increase the system's overall computational load, power consumption, processing, and transmission activities [27,29]. These shortfalls can be compensated for to a certain extent by using level-crossing analog-to-digital converters (LCADCs) [16,24,[30][31][32][33][34][35]. These converters adapt the system acquisition and processing rates according to the incoming signal's temporal variations, a process that renders a significant computational efficiency of the suggested approach compared to traditional ones.…”

Section: Introductionmentioning

confidence: 99%

Arrhythmia Diagnosis by Using Level-Crossing ECG Sampling and Sub-Bands Features Extraction for Mobile Healthcare

Qaisar

Hussain

2020

Sensors

View full text Add to dashboard Cite

Mobile healthcare is an emerging technique for clinical applications. It is usually based on cloud-connected biomedical implants. In this context, a novel solution is presented for the detection of arrhythmia by using electrocardiogram (ECG) signals. The aim is to achieve an effective solution by using real-time compression, efficient signal processing, and data transmission. The system utilizes level-crossing-based ECG signal sampling, adaptive-rate denoising, and wavelet-based sub-band decomposition. Statistical features are extracted from the sub-bands and used for automated arrhythmia classification. The performance of the system was studied by using five classes of arrhythmia, obtained from the MIT-BIH dataset. Experimental results showed a three-fold decrease in the number of collected samples compared to conventional counterparts. This resulted in a significant reduction of the computational cost of the post denoising, features extraction, and classification. Moreover, a seven-fold reduction was achieved in the amount of data that needed to be transmitted to the cloud. This resulted in a notable reduction in the transmitter power consumption, bandwidth usage, and cloud application processing load. Finally, the performance of the system was also assessed in terms of the arrhythmia classification, achieving an accuracy of 97%.

show abstract

On Quasi-Isometry of Threshold-Based Sampling

Cited by 11 publications

References 43 publications

Machine Learning with Adaptive Rate Processing for Power Quality Disturbances Identification

Machine Learning with Adaptive Rate Processing for Power Quality Disturbances Identification

Bandlimited signal reconstruction from leaky integrate-and-fire encoding using POCS

Arrhythmia Diagnosis by Using Level-Crossing ECG Sampling and Sub-Bands Features Extraction for Mobile Healthcare

Contact Info

Product

Resources

About