SYSTEM AND METHOD FOR REMOVING NARROWBAND NOISE

• US 20060142821A1
• Filed: 03/08/2006
• Published: 06/29/2006
• Est. Priority Date: 01/04/2001
• Status: Active Grant

• Current Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Original Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Inventors: Bange, Joseph E., Schleifer, Fred, Schmitt, Steven
• Granted Patent: US 7,551,965 B2
• Time in Patent Office: Days
• US Class Current: 607/60
• CPC Class Codes: A61N 1/3718 Monitoring of or protection against external electromagnetic fields or currents
• CPC Class Codes: A61N 1/37211 Means for communicating with stimulators
• CPC Class Codes: H04B 1/1036 with automatic suppression of narrow band noise or interference, e.g. by using tuneable notch filters H04B1/123 takes precedence; filter circuits H03H
• CPC Class Codes: Y10S 128/901 Suppression of noise in electric signal

Abstract

A system and method for removing narrowband noise from an input signal in which notch filters having notch frequencies corresponding to the noise are dynamically adjusted in accordance with a detected noise spectrum. The method may be applied to telemetry systems for implantable medical devices such as cardiac pacemakers to result in improved noise immunity.

Claims

1. A system for receiving data from an implantable medical device, comprising:

an antenna for receiving a signal transmitted from the implantable device;

circuitry for computing a frequency spectrum of the received signal;

circuitry for detecting a noise spectrum of the received signal by subtracting a template spectrum from the frequency spectrum of the received signal, the template spectrum representing a noise-free signal;

one or more notch filters for removing noise from the received signal, the notch filters having notch frequencies corresponding to frequency components of the detected noise spectrum;

a filter generator for updating filter coefficients of the notch filters so that the notch frequencies correspond to the detected noise spectrum.

2. The system of claim 1 wherein the filter generator is configured to set the notch frequencies of the notch filters to match those spectral components of the noise spectrum that exceed a specified spectral threshold value.

3. The system of claim 2 wherein the specified spectral threshold value is calculated as a specified number of standard deviations from the mean of the noise spectrum.

4. The system of claim 1 wherein the filter coefficients of the notch filters are periodically updated by the filter generator.

5. The system of claim 4 wherein the filter coefficients of the notch filters are updated based upon samples of the received signal that are continuously input into a buffer.

6. The system of claim 1 wherein the notch filters are digital infinite impulse response filters synthesized with notch frequencies corresponding to the detected noise spectrum.

7. The system of claim 1 wherein the circuitry for computing a spectrum of the received signal is configured to do so by sampling the received signal and discrete Fourier transforming the samples to produce a power spectrum of the received signal.

8. The system of claim 7 wherein the spectrum of the received signal is produced by summing the power spectra of a series of transformed sample sets of the received signal.

9. The system of claim 1 wherein the circuitry for detecting the noise spectrum scales the template spectrum by a factor that reduces the total power in the detected noise spectrum to a minimal value.

10. The system of claim 9 wherein the circuitry for detecting the noise spectrum is configured to compute the noise spectrum as:

P _n =P _i −

R*P _e

where P_i is the power spectrum of the received signal, P_e is the template spectrum corresponding to the noise-free signal, and R is a scaling factor chosen to minimize P_n.

11. A method for receiving data from an implantable medical device, comprising:

receiving a signal transmitted from the implantable device;

computing a frequency spectrum of the received signal;

detecting a noise spectrum of the received signal by subtracting a template spectrum from the frequency spectrum of the received signal, the template spectrum representing a noise-free signal;

removing noise from the received signal with one or more notch filters, the notch filters having notch frequencies corresponding to frequency components of the detected noise spectrum;

updating filter coefficients of the notch filters so that the notch frequencies correspond to the detected noise spectrum.

12. The method of claim 11 further comprising setting the notch frequencies of the notch filters to match those spectral components of the noise spectrum that exceed a specified spectral threshold value.

13. The method of claim 12 wherein the specified spectral threshold value is calculated as a specified number of standard deviations from the mean of the noise spectrum.

14. The method of claim 11 further comprising periodically updating the filter coefficients of the notch filters.

15. The method of claim 14 further comprising periodically updating the filter coefficients of the notch filters based upon samples of the received signal that are continuously input into a buffer.

16. The method of claim 11 wherein the notch filters are digital infinite impulse response filters synthesized with notch frequencies corresponding to the detected noise spectrum.

17. The method of claim 11 further comprising sampling the received signal and discrete Fourier transforming the samples to produce a power spectrum of the received signal.

18. The method of claim 17 further comprising summing the power spectra of a series of transformed sample sets of the received signal.

19. The method of claim 11 further comprising scaling the template spectrum by a factor that reduces the total power in the detected noise spectrum to a minimal value.

20. The method of claim 19 further comprising computing the noise spectrum as:

P _n =P _i −

R*P _e

where P_i is the power spectrum of the received signal, P_e is the template spectrum corresponding to the noise-free signal, and R is a scaling factor chosen to minimize P_n.