Back to results
US 4,654,863 AGranted
Wideband adaptive prediction
AssigneeAT&T Bell Laboratories
InventorsWilliam R. Belfield, Raymond S. Medaugh
Life of this patent
Estimating the term…
Abstract
Superior performance, better tracking, and stability of an adaptive predictor are realized by employing a so-called high order (eighth) prediction filter and by updating the filter coefficients in accordance with an algorithm based on the sign of a reconstructed signal sample (s.sub.q (k)) multiplied by a quantized version (e.sub.q (k)) of a differential error signal normalized by an expected magnitude of the reconstructed signal sample.
Claims16
An adaptive predictor for use in an adaptive differential PCM system including
- means for algebraically combining a supplied sample and a predicted signal sample to generate a differential error signal sample, and
- means for generating a quantized version of the differential error signal sample, the predictor comprising,
- a first n-tap adaptive transversal filter,
- a first adaptation control circuit for providing tap coefficients for said n-tap adaptive transversal filter,
- a second m-tap adaptive transversal filter,
- a second filter adaptation control circuit for providing tap coefficients to said second m-tap adaptive transversal filter,
- means for algebraically combining outputs of said first and second transversal filters to form said predicted signal sample,
- means for algebraically combining said quantized version of said differential error signal sample and said predicted signal sample to form a reconstructed signal sample,
- wherein said second filter adaptation control circuit generates said second adaptive transversal filter tap coefficients in accordance with a predetermined relationship based on the sign of said reconstructed signal sample multiplied by said quantized version of said differential error signal sample which is normalized by a value representative of an expected magnitude of said reconstructed signal sample.
6 more elements
An adaptive predictor for use in an adaptive differential PCM system including
- means for algebraically combining a supplied sample and a predicted signal sample to generate a differential error signal sample, and
- means for generating a quantized version of the differential error signal sample, the predictor comprising,
- a first n-tap adaptive transversal filter,
- a first adaptation control circuit for providing tap coefficients for said n-tap adaptive transversal filter,
- a second m-tap adaptive transversal filter,
- a second filter adaptation control circuit for providing tap coefficients to said second m-tap adaptive transversal filter,
- means for algebraically combining outputs of said first and second transversal filters to form said predicted signal sample,
- means for algebraically combining said quantized version of said differential error signal sample and said predicted signal sample to form a reconstructed signal sample,
- wherein said second filter adaptation control circuit generates said second adaptive transversal filter tap coefficients in accordance with the equation
- a.sub.i (k)=.lambda..sub.i a.sub.i (k-1)+g.sub.a sign (s.sub.q (k-i))F(e.sub.q (k)r.sub.q (k))
- where a.sub.i (k) are said second adaptive transversal filter tap coefficients, i=1, 2, 3 . . . m, .lambda..sub.i is a so-called predetermined leak factor, g.sub.a is a predetermined constant, s.sub.q (k) is the reconstructed signal sample, e.sub.q (k) is the quantized version of the differential error signal and F(e.sub.q (k)r.sub.q (k)) is a prescribed nonlinear relationship.
8 more elements
An adaptive predictor for use in an adaptive differential PCM system including
- means for generating a quantized version of a differential error signal sample from a received digital signal sample, the predictor comprising,
- a first n-tap adaptive transversal filter,
- a first adaptation control circuit for providing tap coefficients for said n-tap adaptive transversal filter,
- a second m-tap adaptive transversal filter,
- a second filter adaptation control circuit for providing tap coefficients to said second m-tap adaptive transversal filter,
- means for algebraically combining outputs of said first and second transversal filters to form a predicted signal sample,
- means for algebraically combining said quantized version of said differential error signal sample and said predicted signal sample to form a reconstructed signal sample,
- wherein said second filter adaptation control circuit generates said second adaptive transversal filter tap coefficients in accordance with the equation
- a.sub.i (k)=.lambda..sub.i a.sub.i (k-1)+g.sub.a sign (s.sub.q (k-i))F(e.sub.q (k)r.sub.q (k))
- where a.sub.i (k) are said second adaptive transversal filter tap coefficients, i=1, 2, 3 . . . m, .lambda..sub.i is a so-called predetermined leak factor, g.sub.a is a predetermined constant, s.sub.q (k) is the reconstructed signal sample, e.sub.q (k) is the quantized version of the differential error signal and F(e.sub.q (k)r.sub.q (k)) is a prescribed nonlinear relationship.
7 more elements
An adaptive predictor for use in an adaptive differential PCM system including
- means for generating a quantized version of a differential error signal sample from a received digital signal sample, the predictor comprising,
- a first n-tap adaptive transversal filter,
- a first adaptation control circuit for providing tap coefficients for said n-tap adaptive transversal filter,
- a second m-tap adaptive transversal filter,
- a second filter adaptation control circuit for providing tap coefficients to said second m-tap adaptive transversal filter,
- means for algebraically combining outputs of said first and second transversal filters to form a predicted signal sample,
- means for algebraically combining said quantized version of said differential error signal sample and said predicted signal sample to form a reconstructed signal sample,
- wherein said second filter adaptation control circuit generates said second adaptive transversal filter tap coefficients in accordance with a predetermined relationship based on the sign of said reconstructed signal sample multiplied by said quantized version of said differential error signal sample which is normalized by a value representative of an expected magnitude of said reconstructed signal sample.
5 more elements
An adaptive predictor for use in an adaptive differential PCM system including
- means for algebraically combining a supplied sample and a predicted signal sample to generate a differential error signal sample, and
- means for generating a quantized version of the differential error signal sample, the predictor comprising,
- a first n-tap adaptive transversal filter,
- a first adaptation control circuit for providing tap coefficients for said n-tap adaptive transversal filter,
- a second m-tap adaptive transversal filter,
- a second filter adaptation control circuit for providing tap coefficients to said second m-tap adaptive transversal filter,
- means for algebraically combining outputs of said first and second transversal filters to form said predicted signal sample,
- means for algebraically combining said quantized version of said differential error signal sample and said predicted signal sample to form a reconstructed signal sample,
- wherein said second filter adaptation control circuit generates said second adaptive transversal filter tap coefficients in accordance with the equation ##EQU10## where a.sub.i (k) are said second adaptive transversal filter tap coefficients, i=1, 2, 3 . . . m, .lambda..sub.i is a so-called predetermined leak factor, g.sub.a is a predetermined constant, s.sub.q (k) is the reconstructed signal sample, e.sub.q (k) is the quantized version of the differential error signal and E{.vertline.s.sub.q (k).vertline.} is the expected magnitude of s.sub.q (k).
6 more elements
An adaptive predictor for use in an adaptive differential PCM system including
- means for generating a quantized version of a differential error signal sample from a received digital signal sample, the predictor comprising,
- a first n-tap adaptive transversal filter,
- a first adaptation control circuit for providing tap coefficients for said n-tap adaptive transversal filter,
- a second m-tap adaptive transversal filter,
- a second filter adaptation control circuit for providing tap coefficients to said second m-tap adaptive transversal filter,
- means for algebraically combining outputs of said first and second transversal filters to form a predicted signal sample,
- means for algebraically combining said quantized version of said differential error signal sample and said predicted signal sample to form a reconstructed signal sample,
- wherein said second filter adaptation control circuit generates said second adaptive transversal filter tap coefficients in accordance with the equation ##EQU11## where a.sub.i (k) are said second adaptive transversal filter tap coefficients, i=1, 2, 3 . . . m, .lambda..sub.i is a so-called predetermined leak factor, g.sub.a is a predetermined constant, s.sub.q (k) is the reconstructed signal sample, e.sub.q (k) is the quantized version of the differential error signal and E{.vertline.s.sub.q (k).vertline.} is the expected magnitude of s.sub.q (k).
5 more elements
Dependencies are inferred from the claim text.