Method and apparatus for improving S/N ratio in digital-to-analog conversion of pulse density modulated (PDM) signal

• US 20020018013A1
• Filed: 06/13/2001
• Published: 02/14/2002
• Est. Priority Date: 06/14/2000
• Status: Active Grant

• Current Assignee: Texas Instruments, Inc.
• Original Assignee: Texas Instruments, Inc.
• Inventors: Hamasaki, Toshihiko, Nakao, Shigetoshi
• Granted Patent: US 6,489,909 B2
• Time in Patent Office: Days
• US Class Current: 341/144
• CPC Class Codes: H03M 3/368 of noise other than the quantisation noise already being shaped inherently by delta-sigma modulators
• CPC Class Codes: H03M 3/50 Digital/analogue converters using delta-sigma modulation as an intermediate step digital delta-sigma modulators per se H03M7/3004

Abstract

A method and apparatus for improving an S/N ratio in a digital-to-analog conversion of a PDM signal are provided. A digital-to-analog conversion system comprises an S/N ratio improving section. The S/N ratio improving section has a signal component extractor which extracts a signal component included in the PDM signal, and outputs a digitally filtered output signal. The digitally filtered output signal has a second full scale smaller than a first full scale of the PDM signal. The S/N ratio improving section also comprises a full-scale matching unit which matches the second full scale of the digitally filtered output signal with a third full scale of digital-to-analog conversion,

Claims

1. A method of improving an S/N ratio in digital-to-analog conversion of a pulse density modulated (PDM) digital signal represented by a waveform having a first predetermined full scale value, said PDM digital signal including a signal component having a second predetermined full scale value, said digital-to-analog conversion producing an analog output signal having a third predetermined fall scale value, said method comprising the steps of:

and

extracting said signal component from said PDM digital signal to generate an extracted digital signal having said second fall scale value;

matching said second full scale value of said extracted signal with said third full scale value of said analog output signal; and

performing a digital-to-analog conversion of said extracted signal having said matched second full scale value to produce said analog output signal having said third predetermined full scale value.

2. A method of improving an SIN ratio in digital-to-analog conversion of a pulse density modulated (PDM) digital signal represented by a waveform having a first predetermined full scale value, said PDM digital signal including a signal component having a second predetermined full scale value, said digital-to-analog conversion producing an analog output signal having a third predetermined full scale value, said method comprising the steps of:

extracting said signal component from said PDM digital signal to generate an extracted digital signal having said second full scale value;

increasing the amplitude of said extracted signal according to an amplification factor so as to match said second full scale value of said extracted signal with said third full scale value of said analog output signal; and

performing a digital-to-analog conversion of said extracted signal having said matched second full scale value to produce said analog output signal having said third predetermined full scale value.

3. The method of claim 2 wherein the first predetermined full scale value is determined by a reference voltage.

4. A method of improving an S/N ratio in digital-to-analog conversion of a pulse density modulated (PDM) digital signal represented by a waveform having a first predetermined full scale value, said PDM digital signal including a signal component having a second predetermined full scale value, said digital-to-analog conversion producing an analog output signal having a third predetermined full scale value, said method comprising the steps of:

extracting said signal component from said PDM digital signal to generate an extracted digital signal having said second full scale value by digital-low-pass filtering said PDM digital signal;

matching said second full scale value of said extracted signal with said third full scale value of said analog output signal; and

performing a digital-to-analog conversion of said extracted signal having said matched second full scale value to produce said analog output signal having said third predetermined full scale value.

5. A method of improving an S/N ratio in digital-to-analog conversion of a pulse density modulated (PDM) digital signal represented by a waveform having a first predetermined full scale value, said PDM digital signal including a signal component having a second predetermined full scale value, said digital-to-analog conversion producing an analog output signal having a third predetermined full scale value, said method comprising the steps of:

extracting said signal component from said PDM digital signal to generate an extracted digital signal having said second full scale value by decimation filtering said PDM digital signal;

matching said second full scale value of said extracted signal with said third full scale value of said analog output signal; and

performing a digital-to-analog conversion of said extracted signal having said matched second full scale value to produce said analog output signal having said third predetermined full scale value.

6. A method of improving an S/N ratio in digital-to-analog conversion of a pulse density modulated (PDM) digital signal represented by a waveform having a first predetermined full scale value, said PDM digital signal including a signal component having a second predetermined full scale value, said digital-to-analog conversion producing an analog output signal having a third predetermined full scale value, said method comprising the steps of:

extracting said signal component from said PDM digital signal to generate an extracted digital signal having said second full scale value;

increasing the amplitude of said extracted signal according to an amplification factor so as to match said second full scale value of said extracted signal with said third full scale value of said analog output signal by digitally multiplying said extracted signal by an amplification factor having a value which causes the analog output signal to have less than a predetermined distortion number; and

performing a digital-to-analog conversion of said extracted signal having said matched second full scale value to produce said analog output signal having said third predetermined full scale value.

7. A method of improving an S/N ratio in digital-to-analog conversion of a pulse density modulated (PDM) digital signal represented by a waveform having a first predetermined full scale value, said PDM digital signal including a signal component having a second predetermined full scale value, said digital-to-analog conversion producing an analog output signal having a third predetermined full scale value, said method comprising the steps of:

extracting said signal component from said PDM digital signal to generate an extracted digital signal having said second full scale value;

increasing the amplitude of said extracted signal according to an amplification factor so as to match said second fall scale value of said extracted signal with said third full scale value of said analog output signal by digitally multiplying said extracted signal by an amplification factor having a value which causes the analog output signal to have at least a predetermined signal-to-noise ratio; and

performing a digital-to-analog conversion of said extracted signal having said matched second full scale value to produce said analog output signal having said third predetermined full scale value.

8. A method of digital-to-analog conversion of a signal from either a CD disc or a SACD disc, comprising:

(a) producing a PDM signal which is either

i. a 1 Fs 24 bit PCM signal from digital information stored in a CD disc, or

ii. a 64 Fs 1 bit DSD signal from digital information stored in a SACD disc, wherein Fs is a frequency of sampling of the PDM signal;

(b) if the PDM signal is the 1 Fs 24 bit PCM signal, filtering the 1 Fs 24 bit PCM signal by means of an 8x interpolation digital filter to produce an 8 Fs 24 bit signal and interpolating the 8 Fs 24 bit signal to produce a first 64 Fs 24 bit signal 18A,

and if the PDM signal is the 64 Fs 1 bit DSD signal, filtering the 64 Fs 1 bit DSD signal by means of a digital filter and digitally multiplying an output produced by the digital filter by an amplification factor to produce a second 64 Fs 24 bit signal;

(c) selecting the produced 64 Fs 24 bit signal for the 64 Fs 24 bit signal by means of a switch circuit; and

(d) converting the selected one of the first and second 64 Fs 24 bit signals to an analog signal by means of a digital-to-analog converter,

the amplification factor having a value that causes a signal-to-noise ratio of the second 64 Fs 24 bit signal to be approximately equal to a signal-to-noise ratio of the first 64 Fs 24 bit signal.

9. The method of claim 8 wherein the converting of step (d) includes delta-sigma modulating the selected one of the first and second 64 Fs 24 bit signals, performing a digital-to-analog conversion of a signal resulting from the delta-sigma modulating by means of digital-to-analog conversion elements, and low pass filtering a signal resulting from the analog conversion of the signal resulting from the delta-sigma modulating.

10. A method of digital-to-analog conversion of a signal from either a CD disc or a SACD disc, comprising:

(a) producing a PDM signal which is either

i. a P1×

Fs M-bit PCM signal from digital information stored in a CD disc, wherein P1 is a first oversampling ratio and M is an integer, or

ii. a P2×

Fs 1 bit DSD signal from digital information stored in a SACD disc, wherein P2 is a second oversampling ratio which is greater P1, and wherein Fs is a frequency of sampling of the PDM signal;

(b) if the PDM signal is the P1×

Fs M-bit PCM signal, filtering the P1×

Fs M-bit PCM signal by means of an interpolation digital filter to produce a P3×

Fs M-bit signal and interpolating the P3×

Fs M-bit signal to produce a first P1×

Fs M-bit signal, wherein P3 is a third oversampling ratio,

and if the PDM signal is the P2×

Fs 1 bit DSD signal, filtering the P2×

Fs 1 bit DSD signal by means of a digital filter and digitally multiplying an output produced by the digital filter by an amplification factor to produce a second P2×

Fs M-bit signal;

(c) selecting the produced P2×

Fs M-bit signal for the P2×

Fs M-bit signal by means of a switch circuit; and

(d) converting the selected one of the first and second P2×

Fs M-bit signals to an analog signal by means of a digital-to-analog converter,

the amplification factor having a value that causes a signal-to-noise ratio of the second P2×

Fs M-bit signal to be approximately equal to a signal-to-noise ratio of the first P2×

Fs M-bit signal.

11. The method of claim 10 wherein the converting of step (d) includes delta-sigma modulating the selected one of the first and second P2×

Fs M-bit signals, performing a digital-to-analog conversion of a signal resulting from the delta-sigma modulating by means of digital-to-analog conversion elements, and low pass filtering a signal resulting from the analog conversion of the signal resulting from the delta-sigma modulating.

12. The method of claim 11 including amplifying the low pass filtered signal to provide a drive signal to a speaker.

14. The method of claim 13 wherein the converting of step (d) includes delta-sigma modulating the selected one of the first and second 8 Fs 24 bit signals, performing a digital-to-analog conversion of a signal resulting from the delta-sigma modulating by means of digital-to-analog conversion elements, and low pass filtering a signal resulting from the analog conversion of the signal resulting from the delta-sigma modulating.

15. The method of claim 14 including amplifying the low pass filtered signal to provide a drive signal to a speaker.

17. The method of claim 16 wherein the converting of step (d) includes delta-sigma modulating the selected one of the first and second P3×

Fs M-bit signals, performing a digital-to-analog conversion of a signal resulting from the delta-sigma modulating by means of digital-to-analog conversion elements, and low pass filtering a signal resulting from the analog conversion of the signal resulting from the delta-sigma modulating.

18. The method of claim 17 including amplifying the low pass filtered signal to provide a drive signal to a speaker.

20. An apparatus for improving an S/N ratio according to claim 19 wherein said digital filtering means includes a digital low pass filter.

21. An apparatus for improving an S/N ratio according to claim 19, wherein said digital filtering means includes a decimation filter.

22. An apparatus for improving an S/N ratio according to claim 19, wherein said full-scale matching means includes multiplying means for digitally multiplying said filtered output signal to generate a digitally multiplied output signal in a digital form.

13. A method of digital-to-analog conversion of a signal from either a CD disc or a SACD disc, comprising:

(a) producing a PDM signal which is either

i. a 1 Fs 24 bit PCM signal from digital information stored in a CD disc, or

ii. a 64 Fs 1 bit DSD signal from digital information stored in a SACD disc, wherein Fs is a frequency of sampling of the PDM signal;

(b) if the PDM signal is the 1 Fs 24 bit PCM signal, filtering the 1 Fs 24 bit PCM signal by means of an 8x interpolation digital filter to produce an 8 Fs 24 bit signal,

and if the PDM signal is the 64 Fs 1 bit DSD signal, filtering the 64 Fs I bit DSD signal by means of a decimation filter and digitally multiplying an output produced by the decimation filter by an amplification factor to produce a second 8 Fs 24 bit signal;

(c) selecting the produced 8 Fs 24 bit signal for the 8 Fs 24 bit signal by means of a switch circuit; and

(d) converting the selected one of the first and second 8 Fs 24 bit signals to an analog signal by means of a digital-to-analog converter,

the amplification factor having a value that causes a signal-to-noise ratio of the second 8Fs 24 bit signal to be approximately equal to a signal-to-noise ratio of the first 8 Fs 24 bit signal.

16. A method of digital-to-analog conversion of a signal from either a CD disc or a SACD disc, comprising:

(a) producing a PDM signal which is either

i. a P1×

Fs M-bit PCM signal from digital information stored in a CD disc, wherein P1 is a first oversampling ratio and M is an integer, or

ii. a P2×

Fs 1 bit DSD signal from digital information stored in a SACD disc, wherein P2 is a second oversampling ratio which is greater than PI, and wherein Fs is a frequency of sampling of the PDM signal;

(b) if the PDM signal is the P 1×

Fs M-bit PCM signal, filtering the P1×

Fs M-bit PCM signal by means of an interpolation digital filter to produce an P3×

Fs M-bit signal, wherein P3 is a third oversampling ratio,

and if the PDM signal is the P2×

Fs 1 bit DSD signal, filtering the P2×

Fs 1 bit DSD signal by means of a decimation filter and digitally multiplying an output produced by the decimation filter by an amplification factor to produce a second P3×

Fs M-bit signal;

(c) selecting the produced P3×

Fs M-bit signal for the P3×

Fs M-bit signal by means of a switch circuit; and

(d) converting the selected one of the first and second P3×

Fs M-bit signals to an analog signal by means of a digital-to-analog converter, the amplification factor having a value that causes a signal-to-noise ratio of the second P3×

Fs M-bit signal to be approximately equal to a signal-to-noise ratio of the first P3×

Fs M-bit signal.

19. An apparatus for improving an S/N ratio in digital-to-analog conversion of a pulse density modulated (PDM) digital signal represented by a waveform having a first predetermined full scale value, said PDM digital signal including a signal component having a second predetermined full scale value, said digital-to-analog conversion producing an analog output signal having a third predetermined full scale value, said apparatus comprising:

digital filtering means for digitally filtering said PDM signal to extract said signal component, and generating a digitally filtered output signal in a digital form, said digitally filtered output signal having said second full scale value;

full scale matching means for matching said second full scale value of said digitally filtered output signal with said third full scale value of said processing; and

digital-to-analog conversion means for performing a digital-to-analog conversion of said extracted signal having said matched second full scale value to produce said analog output signal having said third predetermined full scale value.

23. An apparatus for improving an S/N ratio in digital-to-analog conversion of a pulse density modulated (PDM) digital signal represented by a waveform having a first predetermined full scale value, said PDM digital signal including a signal component having a second predetermined full scale value, said digital-to-analog conversion producing an analog output signal having a third predetermined full scale value, said apparatus comprising:

a digital low pass filter for digitally filtering said PDM signal to extract said signal component, and generating a digitally filtered output signal in a digital form, said digitally filtered output signal having said second full scale value;

gain control circuitry for increasing the amplitude of said extracted signal according to in amplification factor so as to match said second full scale of said digitally filtered output signal with said third full scale of said processing; and

a digital-to-analog converter for performing a digital-to-analog conversion of said extracted signal having said matched second full scale value to produce said analog output signal having said third predetermined full scale value.