Time-alignment measurement for hybrid HD radio™ technology

Abstract

A method for processing a digital audio broadcast signal in a radio receiver, includes: receiving a hybrid broadcast signal; demodulating the hybrid broadcast signal to produce an analog audio stream and a digital audio stream; and using a normalized cross-correlation of envelopes of the analog audio stream and the digital audio stream to measure a time offset between the analog audio stream and the digital audio stream. The time offset can be used to align the analog audio stream and the digital audio stream for subsequent blending of an output of the radio receiver from the analog audio stream to the digital audio stream or from the digital audio stream to the analog audio stream.

What is claimed is: 
     
       1. A method for processing a digital of operating an audio signal receiver to process an audio broadcast signal in a radio receiver that includes multiple audio streams, the method comprising:
 receiving a hybrid broadcast signal; 
 demodulating the hybrid broadcast audio signal to produce an analog a first audio sample stream of the audio signal and a digital second audio sample stream of the audio signal; 
 usingcomputing a normalized cross-correlation of envelopes of the analogfirst audio sample stream and the digitalsecond audio sample stream to measure a time offset between the analogfirst audio sample stream and the digitalsecond audio sample stream using audio processing circuitry of the audio signal receiver; 
 computing a coarse envelope cross-correlation over a first range of lag values to locate a vicinity of the time offset; 
 subsequently computing a fine envelope cross-correlation over a second range of lag values, wherein the second range of lag values is narrower than the first range of lag values; 
 producing an audio blend by blending an output of the radio audio signal receiver from the analog first audio sample stream to the digital second audio sample stream or from the digital second audio sample stream to the analog first audio sample stream; 
 phase-adjusting the digital second audio sample stream; and 
 using the phase-adjusted digital second audio sample stream to temporarily replace input digital audio frames during blend ramps used in the blending audio blend of the output of the radio audio signal receiver. 
 
     
     
       2. The method of  claim 1 , further comprising:
 using the time offset to align the analog first audio sample stream and the digital second audio sample stream. 
 
     
     
       3. The method of  claim 1 , wherein the first audio sample stream is an analog audio sample stream, and the method further comprising:
 using the time offset to scale an analog signal blend metric, to control blend thresholds in the blending of the output of the radio audio signal receiver, and to inhibit blending when misalignment is detected. 
 
     
     
       4. The method of  claim 1 , further comprising:
 computing cross-correlation of bass signals to detect potential inversion, to validate time offset measurements, or to improve blend quality. 
 
     
     
       5. The method of  claim 1 , further comprising:
 computing cross-correlation of the analog first audio sample stream and the digital second audio sample stream to predict sound quality of a potential audio blend. 
 
     
     
       6. The method of  claim 1 , wherein:
 the normalized cross-correlation of envelopes is computed using a vector of bandpass samples of the analog first audio sample stream and a vector of bandpass samples of the digital second audio sample stream. 
 
     
     
       7. The method of  claim 1 , wherein the step of using a normalized cross-correlation of envelopes of the analog audio stream and the digital audio stream to measure a time offset between the analog audio stream and the digital audio stream comprises:
 using a coarse envelope cross-correlation computed over a first range of lag values to locate a vicinity of the time offset; and   subsequently using a fine envelope cross-correlation computed over a second range of lag values, wherein the second range of lag values is narrower than the first range of lag values.   
     
     
       8. The method of claim  7  1, further comprising: using quadratic interpolation of peak indices to improve resolution of a computed peak lag. 
     
     
       9. The method of  claim 1 , wherein
 the normalized cross-correlation of envelopes is computed using a vector of samples of the analog first audio sample stream and a vector of samples of the digital second audio sample stream; and 
 the normalized cross-correlation of envelopes of the vector of samples of the analog first audio sample stream and the vector of samples of the digital second audio sample stream produces bifurcated and composite correlation peaks that are compared for correlation validation via temporal consistency. 
 
     
     
       10. The method of  claim 1 , wherein:
 the normalized cross-correlation of envelopes is computed using a first audio vector of samples of the analog first audio sample stream and a second audio vector of samples of the digital second audio sample stream; and 
 the normalized cross-correlation of envelopes of the analog first audio vector and the digital second audio vector produces current and previous peaks that are compared for correlation validation via temporal consistency. 
 
     
     
       11. The method of  claim 1 , further comprising:
 calculating phase-adjusted frequency-domain correlation coefficients to validate the time offset. 
 
     
     
       12. A radio An audio signal receiver comprising:
 processing circuitry configured:
 to receive a hybrid broadcast an audio signal that includes multiple audio streams; 
 to demodulate the hybrid broadcast audio signal to produce an analog a first audio sample stream and a digital second audio sample stream; 
 to use a normalized cross-correlation of envelopes of the analog first audio sample stream and the digital second audio sample stream to measure a time offset between the analog first audio sample stream and the digital second audio sample stream; 
 to compute a coarse envelope cross-correlation over a first range of lag values to locate a vicinity of the time offset; and 
 to subsequently compute a fine envelope cross-correlation over a second range of lag values, wherein the second range of lag values is narrower than the first range of lag values. 
 
 
     
     
       13. The radio audio signal receiver of  claim 12 , wherein the processing circuitry is further configured to use quadratic interpolation of peak indices to improve resolution of a computed peak lag. 
     
     
       14. The radio audio signal receiver of  claim 12 , wherein the processing circuitry is further configured:
 to compute the normalized cross-correlation of envelopes using a vector of samples of the analog first audio sample stream and a vector of samples of the digital second audio sample stream to produce bifurcated and composite correlation peaks; and 
 to compare the bifurcated and composite correlation peaks for correlation validation via temporal consistency. 
 
     
     
       15. The radio audio signal receiver of  claim 12 , wherein the processing circuitry is further configured:
 to compute the normalized cross-correlation of envelopes using a vector of samples of the analog first audio sample stream and a vector of samples of the digital second audio sample stream to produce current and previous peaks; and 
 to compare the current and previous peaks for correlation validation via temporal consistency. 
 
     
     
       16. The radio audio signal receiver of  claim 12 , wherein the processing circuitry is further configured to calculate phase-adjusted frequency-domain correlation coefficients to validate the time offset.