Pulse or digital communications – Systems using alternating or pulsating current – Antinoise or distortion
Patent
1993-02-03
1996-07-09
Chin, Stephen
Pulse or digital communications
Systems using alternating or pulsating current
Antinoise or distortion
375296, 348613, H04N 521
Patent
active
055352464
DESCRIPTION:
BRIEF SUMMARY
BACKGROUND OF THE APPLICATION
1. Field of the Invention
The present invention relates to the field of signal processing and, in particular to the processing of video signals so as to achieve an improved noise reduction.
2. Related Art
Non-linear pre and de-emphasis has been used for noise reduction of video signals. Non-linear pre/de-emphasis of a type referred to as E7 has been used successfully for satellite transmission of the MAC packet family and is now specified in the EBU MAC packet specification (Tech Doc 3258). Due to the trend towards lower satellite power there is a desire to achieve further noise improvement. Attempts to achieve this by using more severe non-linear functions have shown two problems:
Firstly, it has been found that using a more severe non-linear function the peak to peak deviation of the FM signal increases and this can result in truncation noise caused by the restricted channel bandwidth.
Secondly, it has been found that although better noise performance can be achieved in plain areas of TV pictures, the noise in highly detailed areas is not improved and is sometimes worsened. E7 relies on the subjective masking effect of the picture on the noise. This principle breaks down if the non-linear parameters are modified in an attempt to achieve a significantly greater noise improvement.
FIG. 1 shows de-emphasis configurations which have been used for E7. The signal is split into two frequency bands using either a low pass filter (FIG. 1a) or a high pass filter (FIG. 1b) and a subtraction. The high frequency components are subjected to non-linear processing and then recombined within the unmodified low frequency components,. The non-linear law used in the de-emphasis is generally as shown in FIG. 2.
Further alternative structures which could be used to implement this type of non-linear pre-emphasis are shown in FIGS. 1 (c) and 1 (d). FIG. 1(c) shows a simple case where low pass and high pass filters are used to separate the low and high frequency bands so that the non-linearity may be applied to the high frequency band alone. FIG. 1 (d) shows a more sophisticated arrangement in which, instead of generating the low pass path using the high pass filter and subtractor of FIG. 1 (b), it has been generated by modifying the non-linearity to a form N' = N(V) - V, which produces a result which is mathematically equivalent. This structure is preferred because it uses fewer elements than the other arrangements. Also, when the non-linearity is implemented using a look-up table implemented as a PROM, fewer bits are needed to represent the output of the PROM when the modified non-linearity N' is used. Thus the capacity of the PROM may be reduced.
Non-linear pre-emphasis for PAL signals has also been proposed in our co-pending British patent application no. 9027598.3. FIG. 3 shows the pre-emphasis structure including pre-correction which has been used. The non-linearities N1, N2 are related to a non-linearity N of the same form as that in FIG. 2, by the following relations: by the relation V.sub.o =N(V.sub.1), and N.sup.-1 is the reciprocal function of N. N is a monotonic function.
SUMMARY OF THE INVENTION
In a first aspect, the present invention provides a method of pre-emphasising a signal comprising the steps of: parallel signal processing paths each comprising a non-linear phase filter and a non-linearity, and a further parallel path, the filters in the pair of processing paths having equal but opposite phase responses and substantially equal magnitude responses; and pre-emphasised signal, whereby the input signal is subjected to a non-linear dynamic range compression which is phase linear.
The first aspect of the present invention further provides pre-emphasis apparatus comprising: a pair of parallel processing paths each comprising a non-linear phase filter and a non-linearity, and a further parallel path, wherein the filters in the pair of processing paths have equal but opposite phase responses and substantially equal magnitude responses; and a raw pre-emphasised signal, whereby the si
REFERENCES:
patent: 3972010 (1976-07-01), Dolby
patent: 4109212 (1978-08-01), Donnell
patent: 4281295 (1981-07-01), Nishimura et al.
patent: 4462001 (1984-07-01), Girard
patent: 4803732 (1989-02-01), Dillon
patent: 4890300 (1989-12-01), Andrews
patent: 5079633 (1992-01-01), Hagino et al.
patent: 5105446 (1992-04-01), Ravoalavoson et al.
patent: 5107520 (1992-04-01), Karam et al.
patent: 5191597 (0293-02-01), Ekelund et al.
WO88/0780, PCT, Oct. 6, 1988.
European Conference on Speech Communication and Technology; Eurospeech 89; Sep. 1989; vol. 1 Tubach et al.: pp. 582-585.
Moorer et al; Linear-Phase Bandsplitting: Theory and Applications; AES Convention; 1984; pp. 1-15.
Etude et concpetion d'un DE-ESSER; Electronique Radio Plans; Mar. 1990; No. 508; pp.25-32.
Chin Stephen
May Timothy J.
National Transcommunications Limited
LandOfFree
Method of video noise reduction using non-linear pre/de-emphasis does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of video noise reduction using non-linear pre/de-emphasis, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of video noise reduction using non-linear pre/de-emphasis will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1874358