Signal processing device, detection device, sensor, electronic apparatus and moving object

• US 10,215,785 B2
• Filed: 12/11/2014
• Issued: 02/26/2019
• Est. Priority Date: 12/12/2013
• Status: Active Grant

• Current Assignee: Seiko Epson Corporation (Seiko Group)
• Original Assignee: Seiko Epson Corporation (Seiko Group)
• Inventors: Haneda, Hideo
• Primary Examiner(s): Fan, Bo
• Application Number:
  US14/567,404
• Publication Number:
  US 20150168467A1
• Time in Patent Office: 1,538 Days
• Field of Search: 702191, 702 96, 324617, 382261, 342109, 342 70, 600324
• CPC Class Codes: G01C 21/1656 with passive imaging devices, e.g. cameras
• CPC Class Codes: G01C 21/188 for accumulated errors, e.g. by coupling inertial systems with absolute positioning systems
• CPC Class Codes: G01R 23/165 using filters
• CPC Class Codes: G01R 29/26 Measuring noise figure; Measuring signal-to-noise ratio

Abstract

A signal processing device includes a Kalman filter that extracts a DC component of an input signal by performing Kalman filter processing on the basis of observation noise and system noise which are estimated from the input signal.

Claims

1. A signal processing device for processing an input signal generated by a sensor and detected by a circuit, the device comprising:

a noise estimation circuit configured to (i) receive the input signal generated by the sensor, (ii) estimate an observation noise signal by extracting the observation noise signal from the received input signal, (iii) estimate a system noise signal by performing gain processing on the observation noise signal using a gain processing circuit, (iv) generate a dispersion of observation noise based on the observation noise signal, and (v) generate a dispersion of system noise based on the system noise signal; and

an arithmetic circuit configured to

receive the input signal generated by the sensor,

receive, from the noise estimation circuit, the dispersion of observation noise and the dispersion of system noise,

provide the input signal, the dispersion of observation noise, and the dispersion of system noise as inputs to a Kalman filter,

perform Kalman filter processing on the input signal to extract a DC component of the input signal as an estimation value of the Kalman filter processing using the dispersion of observation noise and the dispersion of system noise as estimated from the input signal by the noise estimation circuit, and

generate an output signal to be applied to the input signal, wherein the input signal corresponds to a physical measurement performed by the sensor and the output signal adjusts the input signal based on a difference between the input signal and the extracted DC component of the input signal,

wherein the Kalman filter processing includes performing a time update process of an error covariance using the dispersion of system noise received from the noise estimation circuit and performing an observation update process of a Kalman gain using the dispersion of observation noise received from the noise estimation circuit.

2. The signal processing device according to claim 1, wherein the noise estimation circuit mean-squares the input signal, and estimates the observation noise signal by performing limiter processing on the mean-squared signal.

3. The signal processing device according to claim 1, wherein the noise estimation circuit performs high-pass filter processing on the input signal, and increases the observation noise based on the high-pass filter processed signal.

4. The signal processing device according to claim 1, wherein the noise estimation circuit performs high-pass filter processing on the input signal, mean-squares the high-pass filter processed signal and estimates the observation noise signal by performing limiter processing on the mean-squared signal.

5. The signal processing device according to claim 3, wherein the noise estimation circuit performs peak hold processing on the high-pass filter processed signal, and increases the observation noise based on the peak hold processed signal.

6. The signal processing device according to claim 4, wherein the noise estimation circuit performs peak hold processing on the mean-squared signal, and estimates the observation noise based on the limiter processed signal and peak hold processed signal.

7. The signal processing device according to claim 1, further comprising a monitor that monitors the input signal,

wherein the monitor determines whether a signal level of the input signal exceeds a predetermined range, and

the Kalman filter stops observation updating an error covariance when it is determined by the monitoring portion that the signal level of the input signal exceeds the predetermined range.

8. The signal processing device according to claim 7,

wherein the noise estimation circuit stops updating the estimation of the observation noise when it is determined by the monitor that the signal level of the input signal exceeds the predetermined range.

9. The signal processing device according to claim 8, wherein the monitor compares an absolute value of the input signal with a threshold corresponding to the predetermined range, and outputs a stop signal when the absolute value is larger than the threshold,

the Kalman filter stops observation updating the error covariance when the stop signal is received, and

the noise estimation circuit stops updating the estimation of the observation noise when the stop signal is received.

10. The signal processing device according to claim 7,

wherein the noise estimation circuit increases the observation noise when the signal level of the input signal exceeds a second predetermined range.

11. The signal processing device according to claim 10, wherein the noise estimation circuit includes:

a limiter that limits the signal level of the input signal to a predetermined lower limit when the signal level of the input signal falls within the second predetermined range; and

a multiplier that multiplies the observation noise by the signal level of the input signal which is limited by the limiter.

12. The signal processing device according to claim 1, wherein the noise estimation circuit outputs the system noise based on an output signal of a temperature sensor to the Kalman filter.

13. The signal processing device according to claim 12, wherein the noise estimation circuit increases the system noise in response to a change in the output signal of the temperature sensor.

14. The signal processing device according to claim 1, further comprising a subtractor that subtracts a DC designated value corresponding to temperature characteristics of a DC component of the input signal from the input signal,

wherein the Kalman filter performs Kalman filter processing on the input signal after the subtraction.

17. An electronic apparatus comprising the signal processing device according to claim 1.

18. A moving object comprising the signal processing device according to claim 1.

19. The signal processing device according to claim 3, wherein the noise estimation circuit adds the high-pass filter processed signal to the observation noise signal.

20. The signal processing device according to claim 12, wherein the noise estimation circuit adds a signal based on the output signal of the temperature sensor to the system noise signal.

15. A detection device comprising:

a detection circuit that receives a detection signal from a physical quantity transducer, detects a physical quantity signal based on a physical quantity, and generates an input signal based on the physical quantity; and

a noise estimation circuit configured to (i) receive the input signal generated by the detection circuit, (ii) estimate an observation noise signal by extracting the observation noise signal from the received input signal, and (iii) estimate a system noise signal by performing gain processing on the observation noise signal using a gain processing circuit, (iv) generate a dispersion of observation noise based on the observation noise signal, and (v) generate a dispersion of system noise based on the system noise signal; and

an arithmetic circuit configured to

receive the input signal generated by the detection circuit,

receive, from the noise estimation circuit, the dispersion of observation noise and the dispersion of system noise,

provide the input signal, the dispersion of observation noise, and the dispersion of system noise as inputs to a Kalman filter,

perform Kalman filter processing on the input signal to extract a DC component of the input signal as an estimation value of the Kalman filter processing using the dispersion of observation noise and the dispersion of system noise as estimated from the input signal by the noise estimation circuit, and

generate an output signal to be applied to the input signal, wherein the input signal corresponds to a physical measurement performed by the physical quantity transducer and the output signal adjusts the input signal based on a difference between the input signal and the extracted DC component of the input signal,

wherein the Kalman filter processing includes performing a time update process of an error covariance using the dispersion of system noise received from the noise estimation circuit and performing an observation update process of a Kalman gain using the dispersion of observation noise received from the noise estimation circuit.

16. A sensor comprising:

a detection device according to claim 15; and

the physical quantity transducer.