Multirate digital voice communication processor

• US 4,330,689 A
• Filed: 01/28/1980
• Issued: 05/18/1982
• Est. Priority Date: 01/28/1980
• Status: Expired due to Term

• Current Assignee: the united states of america as represented by the secretary of the navy
• Original Assignee: the united states of america as represented by the secretary of the navy
• Inventors: Kang, George S., Fransen, Lawrence J., Kline, Evans L.
• Primary Examiner(s): Krass, Errol A.
• Assistant Examiner(s): KEMENY, EMANUEL
• Application Number:
  US06/115,860
• Time in Patent Office: 841 Days
• Field of Search: 179/15.55 R, 179/1.5 A, 179/1.5 R, 370/84, 370/106, 370/112, 370/43, 375/112
• US Class Current: 704/219
• CPC Class Codes: G10L 19/06 Determination or coding of the spectral characteristics, e.g. of the short-term prediction coefficients
• CPC Class Codes: H04B 1/66 for reducing bandwidth of signals; for improving efficiency of transmission H04B1/68 takes precedence

Abstract

To provide alternative operation at 2.4 or 9.6 Kb/s bit rates, a multirateigital processor generates a data frame of 216 bits representing sync and voice encoded at both rates, with the 2.4 Kb/s embedded as a subset within the frame. Rate conversion is made by inserting or deleting 162 bits per frame. Speech signal is transformed by linear prediction into coefficients and residual signals. The coefficients are transmitted similarly at both 2.4 and 9.6 Kb/s rates. However, the residual is processed into a simple pitch/hiss excitation signal for 2.4 Kb/s, but is processed into a Fourier transform of its baseband (250-1500 Hz) for 9.6 Kb/s excitation signal transmission.

Claims

1. A multirate digital voice communication processor for generating and utilizing a high data-rate digital signal with a low data-rate subset therein to permit the direct processing of this signal at both wideband and narrowband processors without analog tandeming comprising:

a speech input;

a linear predictive coding processor for decomposing an input speech signal into filter coefficient components and a prediction residual component;

a narrowband excitation parameter extractor for representing said prediction residual component by a low data-rate digital signal;

a wideband excitation signal encoder for representing said prediction residual component by a high data-rate digital signal;

means for multiplexing into a data stream and transmitting said filter coefficient components, said narrowband low data-rate digital signal, and said wideband high data-rate digital signal;

means for receiving a transmitted encoded signal and demultiplexing said filter coefficient components and said low and high data-rate digital signals therefrom; and

means for decoding said demultiplexed encoded signals and reconstructing an approximation of the original speech prior to encoding.

2. A multirate digital voice communication processor as defined in claim 1, wherein wideband excitation encoder includes:

means for determining the amplitude and phase spectral components of said prediction residual component; and

,

means for representing these spectral components by said high data-rate digital signal.

3. A multirate digital voice communication processor as defined in claim 2, wherein said decoding means comprises:

means connected to said receiving and demultiplexing means for processing narrowband low data-rate digital signals,

means connected to said receiving and demultiplexing means for reconstructing an approximation of said prediction residual component from the amplitude and phase spectral components represented by said high data-rate digital signal; and

synthesizing means for utilizing said filter coefficient components and either the processed narrowband low data-rate digital signal from said narrowband processing means or the reconstructed approximation of said prediction residual component from said reconstructing means to produce an output speech signal approximately corresponding to an input speech signal.

4. A multirate digital voice processor as defined in claim 3, wherein said wideband excitation signal encoder includes means for representing by said high data rate digital signal only the lower range or baseband of the total spectral bandwidth of the amplitude and phase spectral components.

5. A multirate digital voice processor as defined in claim 4, wherein said prediction residual reconstructing means includes means for replicating in the lower and the various upper spectral bandwidths the amplitude and phase spectral components from the encoded lower range of the spectral bandwidth in order to reconstruct said prediction residual component.

6. A multirate digital voice processor as recited in claim 5, wherein the high data-rate for said high data-rate digital signal is 9.6 Kbps and comprises a set of 216 bits per frame of time, and the low data-rate for said low data-rate digital signal is 2.4 kbps and comprises 13 bits for the narrowband excitation parameter plus 41 bits for said filter coefficient components as a subset of said high data-rate, said high data-rate being converted to said low data-rate by deleting 162 bits per frame of time from the high data-rate, and the low data-rate being converted to the high data-rate by inserting 162 bits per frame of time.

7. A multirate digital voice processor as recited in claim 5, wherein said encoding means comprises:

first gating means for forming a number of samples of said prediction residual per frame of time;

first data buffering means for receiving said samples from the first gating means;

second data buffering means for storing said samples;

second gating means for transferring said samples from the first buffering means to the second buffering means;

Fourier transform means for converting said samples of the prediction residual to said amplitude and phase spectral components;

third gating means for transferring said samples of the prediction residual from said second data buffering means to said Fourier transform means;

multiplier means for scaling said samples prior to input to the Fourier transform means to facilitate waveform smoothing;

amplitude encoding means for encoding said amplitude spectral components;

phase encoding means for encoding said phase spectral components;

fourth gating means for transferring only a selected number (less than half) of contiguous amplitude and phase components from the Fourier transform means to said amplitude and phase encoding means respectively;

third data buffering means for storing the encoded amplitude and phase components for the transmission thereof; and

fifth gating means for transferring said encoded baseband components from said amplitude and phase encoding means to said third data buffering means.

8. A multirate digital voice processor as recited in claim 5, wherein said decoding means comprises:

fourth data buffering means for storing the transmitted encoded lower spectral or baseband amplitude and phase components;

amplitude decoding means for decoding said encoded baseband amplitude components;

phase decoding means for decoding said encoded baseband phase components;

sixth gating means for transferring said encoded baseband amplitude and phase components from said fourth data buffering means to said amplitude and phase decoding means respectively;

inverse Fourier transform means for receiving said decoded baseband amplitude and phase components from the amplitude and phase decoding means, respectively;

seventh gating means for transferring said decoded baseband amplitude and phase components from the amplitude and phase decoding means, respectively, to fill their respective baseband spectral inputs and to also to fill the upperband spectral inputs of said inverse Fourier transform means in order to permit the conversion of said amplitude and phase spectral components to an approximation of the prediction residual component;

fifth data buffering means for storing the output of the inverse Fourier transform means;

eight gating means for transferring the output of the inverse Fourier transform means to said fifth data buffering means;

means for smoothing said output of the inverse Fourier transform means for use as an excitation signal.

9. A multirate digital voice processor as recited in claim 7, wherein said first data buffering means is a FIFO data buffer.

10. A multirate digital voice processor as recited in claim 7, wherein said second, third, fourth and fifth gating means are parallel transfer gates.

11. A multirate digital voice processor as recited in claim 7, wherein said third data buffering means comprises a parallel output buffer for storing said encoded baseband amplitude components, and a parallel output buffer for storing said encoded baseband phase components.

12. A multirate digital voice processor as recited in claim 7, wherein said Fourier transform means is a fast Fourier transform means.

13. A multirate digital voice processor as recited in claim 8, wherein said sixth, seventh and eighth gating means are parallel transfer gates.

14. A multirate digital voice processor as recited in claim 8, wherein said fifth data buffering means is a FIFO data buffer.

15. A multirate digital voice processor as recited in claim 8, wherein said inverse Fourier transform means is an inverse fast Fourier transform means, and said decoding means comprises means for truncating the output of said inverse fast Fourier transform means.