ECG data compression by multiscale peak analysis

Nakashizuka, Makoto; Kikuchi, Hisakazu; Makino, Hideo; Ishii, Ikuo

doi:10.1109/icassp.1995.480428

“…As the maximum filter gain is unity, potential high-frequency interference does not lead to clipping of the signal at the supply rails. Using (8) and (9) we can rewrite (6) as follows:…”

Section: B Circuit Implementationmentioning

confidence: 99%

An Adaptive Sampling System for Sensor Nodes in Body Area Networks

Rieger

¹

,

Taylor

²

2009

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

The importance of body sensor networks to monitor patients over a prolonged period of time has increased with an advance in home healthcare applications. Sensor nodes need to operate with very low-power consumption and under the constraint of limited memory capacity. Therefore, it is wasteful to digitize the sensor signal at a constant sample rate, given that the frequency contents of the signals vary with time. Adaptive sampling is established as a practical method to reduce the sample data volume. In this paper a low-power analog system is proposed, which adjusts the converter clock rate to perform a peak-picking algorithm on the second derivative of the input signal. The presented implementation does not require an analog-to-digital converter or a digital processor in the sample selection process. The criteria for selecting a suitable detection threshold are discussed, so that the maximum sampling error can be limited. A circuit level implementation is presented. Measured results exhibit a significant reduction in the average sample frequency and data rate of over 50% and 38%, respectively.Index Terms-Adaptive sampling, analog electronics, analog signal processing, bio-signal recording, sensor networks.

show abstract

“…In general, the derivative of a sinusoidal signal is expressed in the frequency domain as multiplication by angular frequency and the operator (8) which is a function providing 180 phase shift and a gain proportional to . It can thus be approximated by a second-order filter with a suitably high cutoff frequency .…”

Section: B Circuit Implementationmentioning

confidence: 99%

“…As the maximum filter gain is unity, potential high-frequency interference does not lead to clipping of the signal at the supply rails. Using (8) and (9) we can rewrite (6) as follows: (10) And so the criterion for switching to the higher clock speed is given by (11) (11) It is customary to oversample medical data such as ECG by a factor typically around 4-5 [24]. The oversampling factor (OSF) determines the required fast sample rate.…”

Section: B Circuit Implementationmentioning

confidence: 99%

An Adaptive Sampling System for Sensor Nodes in Body Area Networks

Rieger

¹

,

Taylor

²

2009

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

The importance of body sensor networks to monitor patients over a prolonged period of time has increased with an advance in home healthcare applications. Sensor nodes need to operate with very low-power consumption and under the constraint of limited memory capacity. Therefore, it is wasteful to digitize the sensor signal at a constant sample rate, given that the frequency contents of the signals vary with time. Adaptive sampling is established as a practical method to reduce the sample data volume. In this paper a low-power analog system is proposed, which adjusts the converter clock rate to perform a peak-picking algorithm on the second derivative of the input signal. The presented implementation does not require an analog-to-digital converter or a digital processor in the sample selection process. The criteria for selecting a suitable detection threshold are discussed, so that the maximum sampling error can be limited. A circuit level implementation is presented. Measured results exhibit a significant reduction in the average sample frequency and data rate of over 50% and 38%, respectively.

show abstract

“…Transform-based techniques seek for a sparse representation in a particular basis and compression is achieved by discarding coefficients with magnitude less than a certain threshold. Typical examples include the Fourier transform, the Karuhnen-Loève transform, or the wavelet transform [5].…”

Section: Prior Work On Ecg Compression and Modelingmentioning

confidence: 99%

Finite rate of innovation based modeling and compression of ECG signals

Baechler

¹

,

Freris

²

,

Quick

³

et al. 2013

2013 IEEE International Conference on Acoustics, Speech and Signal Processing

View full text Add to dashboard Cite

Mobile health is gaining increasing importance for society and the quest for new power efficient devices sampling biosignals is becoming critical. We discuss a new scheme called Variable Pulse Width Finite Rate of Innovation (VPW-FRI) to model and compress ECG signals. This technique generalizes classical FRI estimation to enable the use of a sum of asymmetric Cauchy-based pulses for modeling electrocardiogram (ECG) signals. We experimentally show that VPW-FRI indeed models ECG signals with increased accuracy compared to current standards. In addition, we study the compression efficiency of the method: compared with various widely used compression schemes, we showcase improvements in terms of compression efficiency as well as sampling rate.

show abstract

ECG data compression by multiscale peak analysis

Cited by 5 publications

References 5 publications

An Adaptive Sampling System for Sensor Nodes in Body Area Networks

An Adaptive Sampling System for Sensor Nodes in Body Area Networks

An Adaptive Sampling System for Sensor Nodes in Body Area Networks

Finite rate of innovation based modeling and compression of ECG signals

Contact Info

Product

Resources

About