×

Adaptive prediction differential PCM-type transmission apparatus and process with shaping of the quantization noise

  • US 4,385,393 A
  • Filed: 04/10/1981
  • Issued: 05/24/1983
  • Est. Priority Date: 04/21/1980
  • Status: Expired due to Term
First Claim
Patent Images

1. A coding process of the adaptive prediction differential PCM type comprising the steps of:

  • forming an error signal et from the difference between a sample of signal yt to be coded and a prediction signal pt of said signal in which t is a sampling time;

    quantizing said error signal et ;

    coding the quantized signal;

    forming a restored error signal et from one of said quantized and coded signals;

    forming a restored signal yt by adding said restored error signal et to said prediction signal pt ; and

    forming said prediction signal pt from said restored signals yt and et by first and second linear filtering operations;

    said first operation being performed using N successive samples of yt, namely, yt, yt-1, . . . , yt-N+1, and comprising taking a sequence of N samples from a first sequence formed by said N samples of yt and from a second sequence obtained by orthogonalizing said N samples of yt, and multiplying said sequence of N samples respectively by coefficients A1t, A2t, . . . , ANt, then summing the products obtained to obtain a prediction signal pyt ;

    said second operation being performed using P successive samples of et, namely, et, et-1, . . . , et-p+1, and comprising multiplying said P successive samples respectively by coefficients B1t, B2t, . . . , BPt and summing the products obtained to obtain a prediction signal pet ;

    adjusting sequentially said coefficients A1t, A2t, . . . , ANt and B1t, B2t, . . . , BPt at each time t so that the mean power of said error signal et is minimized; and

    carrying out at least one of the two following operations (a) and (b);

    (a) forming a linear filtering of said signal yt by using N successive samples of yt, namely, yt, yt-1, . . . , yt-N+1, and comprising taking a sequence of N samples from a first sequence formed by said N samples of yt and from a second sequence formed by N derivative samples obtained by orthogonalizing said N samples of yt, and multiplying said sequence of samples by N coefficients equal to the said coefficients A1t, A2t, . . . , ANt, then adding the products obtained to supply a filtered signal pyt, then forming on the basis of said signal pyt and said signal pyt previously obtained by the filtering of yt, a signal pARt equal to γ

    AR pyt +(1-γ

    AR)pyt, in which γ

    AR is a regulatable coefficient between 0 and 1 (terminals included);

    (b) filtering the unquantized error signal et by multiplying P successive samples of said signal et, namely, et, et-1, . . . , et-P+1, by P coefficients equal respectively to said coefficients B1t, B2t, . . . , BPt and adding the products obtained, to supply a filtered signal pet, then forming on the basis of said signal pet and the signal pet obtained previously by filtering et a signal pMAt which is equal to γ

    MA pet +(1-γ

    MA)pet, in which γ

    MA is a regulatable coefficient between 0 and 1 (terminals included), the coefficients γ

    AR and γ

    MA not being simultaneously zero; and

    after said operations (a) and (b) adding the signals pARt and pMAt, then delaying one sampling time of the sum obtained, to supply said prediction signal pt.

View all claims
  • 1 Assignment
Timeline View
Assignment View
    ×
    ×