Pulse or digital communications – Spread spectrum – Direct sequence
Patent
1992-08-24
1994-11-01
Chin, Stephen
Pulse or digital communications
Spread spectrum
Direct sequence
381 29, 381 36, 381 37, H04B 166, G10L 302, G10L 500
Patent
active
053612787
DESCRIPTION:
BRIEF SUMMARY
BACKGROUND OF THE INVENTION
This invention is directed to a method of transmitting a signal using digital compression techniques.
The transmission of an audio signal, for example, radio broadcast transmission, cable transmission, satellite transmission and with recording devices entails converting the analog signal into a digital signal with a certain resolution. The digital signal is transmitted and reconverted into an analog signal upon reproduction. The signal-to-noise ratio is enhanced, particularly upon reproduction, by using digital transmission.
The band width required for the transmission of such a signal is essentially determined by the number of scanning values per time unit which are to be transmitted. The resolution is also a function of the number of scanning values transmitted. In practice, it is preferable to keep the transmission band width as small as possible in order to be able to transmit as many audio signals as possible simultaneously via a wide band channel would appear that the necessary band width can be reduced by decreasing the number of scanning values or the number of hits per scanning value. However, as a rule this measure results in a deterioration in the quality of the reproduction.
A method described in DE-OS 35 06 912, improves the reproduction quality by separating the digital audio signal into successive temporal segments and transforming the audio signal into a short-time spectrum which represents the spectral components of the signal for the respective time segments. Generally, in the short-time spectrum, for reasons of psycho-acoustic laws, components which are not perceived by the listener, i.e. are irrelevant from a communications technology viewpoint, can be discovered more readily than in the time domain. Upon transmission, these components are given less weight, or are left out entirely. This allows a considerable part of the otherwise necessary data to be left out and the average bit rate can be considerably reduced.
However, the formation of time segments impairs the frequency resolution because the spectral components brought about by the signal rise and fall at the start and finish of the window are also fed to the spectrum of the original signal. An improvement in the frequency resolution can be attained by having the edge gradient of the window function less steep, also by extending the edge region within the window. These measures require overlapping of adjacent time segments. If the edge region is expanded so far that the window functions no longer have a constant value in any region, then adjacent temporal segments must overlap each other by 50 per cent. This means that the number of scanning values and, accordingly the quantity of data, is doubled.
The publications of J. P. Princen and A. B. Bradley "Analysis Synthesis Filter Bank Design Based On Time Domain Aliasing Cancellation", IEEE Transactions, ASSP-34, No. 5, Oct. 1986, pp 1153 through 1161, and of J. P. Princen, A. W. Johnson and A. B. Bradley "Subband/Transform Coding Using Filter Bank Design Based On Time Domain Aliasing Cancellation", IEEE Int. Conference on Acoustics, Speech and Signal Processing 1987, pp 2161 through 2164, teach that for a 50 per cent overlap of successive temporal segments the quantity of data is reduced to the original value by encoding only every second scanning value. The aliasing components resulting from the subsampling cancel each other out by using the method described in the above citation after inverse transformation upon assembling the time segments.
It has become apparent that with amplitude fluctuations within a time segment, in particular with signals first appearing from a silence during the course of a block, these signals are superimposed with perceivable disturbances after transmission. The cause of the perceivability lies in the fact that the disturbances also appear before the signals start to appear and, therefore, are insufficiently masked. These disturbances can, for example, ensue through quantization noise which superimposes the short-time spectrum
REFERENCES:
patent: 4942607 (1990-07-01), Schroder et al.
patent: 5109417 (1992-04-01), Fielder et al.
IEEE Transactions on Acoustics, Speech, and Signal Processing, vol. ASSP-34, No. 5, Oct. 1986; "Analysis/Synthesis Filter Bank Design Based on Time Domain Aliasing Cancellation", John P. Princen and Alan Bernard Bradley, pp. 1153-1161.
IEEE Transactions, 1987; "Subband/Transform Coding Using Filter Bank Designs Based on Time Domain Aliasing Cancellation", J. P. Princen, A. W. Johnson and A. B. Bradley, pp. 2161-2164.
Rundfunktechnische Mitteilungen, vol. 33, No. 4, Jul. 1989, Norderstedt DE; pp. 149-154; T. Vaupel: "Transformation-Scodierung Fuer Qualitativ Hoshwertige Audiosignale mit Signal vor-und-Nachverarbeitung im Zeitbereich" see p. 151, left hand column, line 25-31; see p. 152, left hand column, last paragraph-p. 153, left hand column, line 8.
Krahe Detlef
Vaupel Thomas
Chin Stephen
Herrmann Eric P.
Kim Kevin
Kurdyla Ronald H.
Telefunken Fernseh und Rundfunk GmbH
LandOfFree
Process for transmitting a signal does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Process for transmitting a signal, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process for transmitting a signal will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1806968