Scaling device as for quantized B-Y signal
First Claim
1. Signal processing apparatus comprising:
- a source of substantially uniformly quantized signals having N possible levels, wherein the number of levels of said quantized signals utilized is KN, where N is a positive integer and K is a number between zero and unity;
first processing means, coupled to said source, for processing said quantized signals, said processing including at least one operation producing modified quantized signals having more than N possible levels;
scaling means, coupled to said processing means, for scaling said modified quantized signals by a predetermined factor having a value greater than unity but not exceeding 1/K; and
second processing means, coupled to said scaling means, for developing output signals having not more than N possible levels in response to said scaled modified quantized signals,whereby said output signals utilize a number of levels exceeding said KN number of levels.
2 Assignments
0 Petitions
Accused Products
Abstract
A signal processing apparatus includes a scaling device for increasing the magnitude of a signal from a source to better use the dynamic range of processing apparatus, and to reduce the effects of noise and error sources. This is particularly useful in a television receiver where the maximum magnitude of the (B-Y) chrominance signal component of a composite video signal is smaller than that of the (R-Y) chrominance signal component. To provide improved resolution of the (B-Y) signal components, a gain scaling arrangement increases the magnitude of the (B-Y) signal component to more closely approximate that of the (R-Y) signal component. In a TV receiver having digital signal processing apparatus, a digital multiplier scales the digitized (B-Y) signal component thereby to reduce the errors inherent in processing digital data having a limited number of bits, especially where such processing includes rounding or truncating operations which can introduce additional errors or decrease resolution.
29 Citations
20 Claims
-
1. Signal processing apparatus comprising:
-
a source of substantially uniformly quantized signals having N possible levels, wherein the number of levels of said quantized signals utilized is KN, where N is a positive integer and K is a number between zero and unity; first processing means, coupled to said source, for processing said quantized signals, said processing including at least one operation producing modified quantized signals having more than N possible levels; scaling means, coupled to said processing means, for scaling said modified quantized signals by a predetermined factor having a value greater than unity but not exceeding 1/K; and second processing means, coupled to said scaling means, for developing output signals having not more than N possible levels in response to said scaled modified quantized signals, whereby said output signals utilize a number of levels exceeding said KN number of levels.
-
-
2. Signal processing apparatus comprising:
-
a source of digital signals having N bits, wherein a number, K(2N), of uniformly spaced digital levels are utilized by said digital signals, N being a positive integer and K being a number between zero and unity; first digital processing means, coupled to said source, for digitally processing said digital signals, said digital processing including at least one operation producing modified digital signals having at least N+1 bits; digital multiplying means, coupled to said digital processing means, for multiplying said modified digital signals by a predetermined factor having a value greater than unity but not exceeding 1/K; means coupled to said multiplying means for reducing the number of bits in the signals provided by said multiplying means to N, wherein the number of digital levels utilized by said bit reduced signals exceeds K(2N); and second processing means, coupled to said bit reducing means, for developing output digital signals in response to the signals provided thereby. - View Dependent Claims (3, 4, 5, 6)
-
-
7. In a digital signal processing system including a source of first and second uniformly quantized digital signals utilizing respective first and second ranges of quantization levels, wherein the number of levels in said first and second ranges may be expressed as KN and LN respectively, N being an integer, K and L being numbers between zero and unity, and K being less than L, and means for combining said first and second quantized digital signals wherein said combined signals include distortion components related to the difference in the number of quantization levels utilized by said first and second quantized signals, apparatus comprising:
-
means coupled between said source and said combining means for reducing the magnitude of said quantization related distortion components including first scaling means for increasing the magnitude of said first quantized signals and the number of levels in said first range by a predetermined factor having a value greater than unity and not greater than L/K, the value of said factor being substantially independent of the instantaneous magnitude of said first digital signals; wherein said combining means includes second scaling means for scaling the relative contributions of at least one of said scaled first digital signals and said second digital signals to approximately counteract any disproportionality caused by scaling said first signals by said predetermined factor. - View Dependent Claims (8, 9, 10, 11, 12, 13, 14)
-
-
15. A method for processing digital signals to increase the number of the possible digital levels of an N-bit digital signals which are utilized comprising the steps of:
-
(a) receiving said digital signal utilizing K2N of its 2N possible uniformly spaced digital levels, where N is a positive integer and K is a number between zero and unity; (b) digitally processing said digital signals to develop modified digital signals having at least N+1 bits; (c) multiplying said modified digital signals by a factor greater than unity and not exceeding 1/K; and (d) developing output digital signals having not more than N bits in response to said multiplied modified digital signals, whereby said output digital signals utilize a number of digital levels exceeding said K2N number of digital levels. - View Dependent Claims (16, 17, 18)
-
-
19. In a signal processing system including a source of first and second substantially uniformly quantized signals utilizing respective first and second ranges of quantization levels, wherein the number of quantization levels in said first range is smaller than the number of quantization levels in said second range, and means for combining said first and second quantized signals wherein said combined signals include distortion components related to the difference in the number of quantization levels utilized by said first and second quantized signals, apparatus comprising:
-
means coupled between said source and said combining means for reducing the magnitude of said quantization related distortion components including first scaling means for increasing the magnitude of said first quantized signals and the number of levels in said first range by a predetermined factor having a value greater than unity and not exceeding the ratio of the numbers of quantization levels in said respective second and first ranges, said value being substantially independent of the instantaneous magnitudes of said first and second quantized signals; and second scaling means coupled to said combining means for scaling the magnitude of at least one of said first and second quantized signals to approximately counteract any disproportionality caused by the increasing of the magnitude of said first quantized signals by said first scaling means.
-
-
20. In a digital video signal processing system including a source of digital composite video signals having first and second chrominance signal components modulating a suppressed carrier signal in quadrature, the maximum magnitude of said first chrominance signal component being K times the maximum magnitude of said second chrominance signal component, K being a number between zero and unity, apparatus comprising:
-
chrominance signal demodulating means coupled to said source for providing first and second digital signals corresponding to said first and second chrominance signal components; digital multiplying means coupled to said demodulating means for multiplying said first digital signals by a predetermined factor greater than unity and not greater than 1/K; and combining means, coupled to said demodulating means and to said digital multiplying means, for developing at least one output signal in response to said multiplied first digital signals and to said second digital signals, said combining means including; scaling means for scaling the relative contributions to said output signals of at least one of said multiplied first digital signals and said second digital signals to approximately counteract any disproportionality caused by multiplying said first signals by said predetermined factor.
-
Specification