Data processing: speech signal processing – linguistics – language – Audio signal bandwidth compression or expansion
Reexamination Certificate
1999-01-28
2002-03-19
{haeck over (S)}mits, T{overscore (a)}livaldis Ivars (Department: 2641)
Data processing: speech signal processing, linguistics, language
Audio signal bandwidth compression or expansion
C704S503000, C708S551000
Reexamination Certificate
active
06360204
ABSTRACT:
The invention relates to an apparatus and a concomitant method for decoding an audio signal. More particularly, the invention relates to a method for increasing the precision of arithmetic operations that are performed in the course of decoding an audio signal.
BACKGROUND OF THE DISCLOSURE
As the demand for digital information proliferates, digital audio signal has become an indispensable part in digital video applications. Such demand for digital audio signal is evidenced by the incorporation of digital audio requirements in various well-known standards, e.g., the AC-3 standard for DVD and ATSC (Advanced Television Systems Committee) digital video applications (including HDTV) and MPEG-1 and MPEG-2 (Moving Picture Experts Group) audio standards for DVB (European digital video standard) applications.
However, audio standards such as the AC-3 standard, generally specify a compression algorithm to produce a digital representation of an audio signal which, when decoded and reproduced, will sound the same or very similar to the original audio signal (e.g., Pulse Code Modulation (PCM) representation), while minimizing the number of bits used to represent the compressed audio representation. As such, the audio decoder plays an important role in the decoding and reproduction of the compressed audio representation.
In order to assure the quality of digital audio, both Dolby AC-3 and MPEG-1/-2 audio standards have specifications on the precision of the audio decoder. For example, Dolby classifies the AC-3 decoder into three groups, Group A, Group B and Group C, corresponding to approximately 20-bit, 18-bit and 16-bit precision, respectively. Similarly, MPEG audio decoders can be classified into “ISO/IEC 13818-3 audio decoder” and “limited accuracy ISO/IEC 13818-3 audio decoder” corresponding to different RMS and maximum absolute value of decoding errors. The ISO/IEC 13818-3 audio decoder should have approximately 16 bit precision.
Generally, if an audio decoder is implemented using a digital signal processor (DSP) having a word-width that is much larger than the required precision, then the decoding accuracy should be easily achieved. For example, using a 32-bit DSP to implement a Group C Dolby AC-3 decoder (20-bit precision) or an ISO/IEC 13818-3 audio decoder (16 bit precision) should be attainable. However, using a more powerful DSP increases cost and complexity. Namely, as the processing word-width (i.e., the number of data bits processed at a given time by the decoder hardware) of the DSP increases, the data bus will be wider accordingly. This, in turn, increases the processing delay, and increases hardware requirements such as read only memory (ROM), random access memory (RAM), arithmetic logic unit (ALU), multiplier/accumulator widths and sizes, thereby increasing the cost and power requirement of the audio decoder.
Therefore, a need exists in the art for an apparatus and concomitant method to achieve a high level of precision without the need to use a high word-width DSP.
SUMMARY OF THE INVENTION
In the present invention, the precision of an audio decoder is increased by proper rounding during arithmetic and shifting operations. Since such operations often generate results that exceed the data word-width of a processor, e.g., a DSP, bits from the results are often discarded at the expense of loss of precision (called “truncation precision loss”) in reproducing the audio signal by the audio decoder. In the present invention, the most significant bit of the discarded bits is evaluated for a rounding up operation. However, prior to the application of rounding, the truncated result must first be evaluated to determine if a saturation condition exists. If a saturation condition is present, no rounding is applied.
For example, using the present invention, an audio decoder is able to achieve precision defined for Dolby AC-3 group B and group C decoder and ISO/IEC 13818-3 audio decoder by simply using a DSP processor with only 20-bit word-width. The increased precision is achieved by proper rounding during the arithmetic and shifting operations.
REFERENCES:
patent: 6029017 (2001-03-01), Lim et al.
Ted Painter and Andreas Spanias, “A Review of Algorithms for Perceptual Coding of Digital Audio Signals,” Proc. 1997 13th International Conference on Digital Signal Processing, vol. 1, p. 179-208, Jul. 1997.*
Il-Taek Lim et al: Fixed-Point Analysis and Simulations of AC-3 Algorithm 1997 IEEE ASSP Workshop on Applications of Signal Processing to Audio and Acoustics, New Paltz, NY Oct. 19-22, 1997 p. 4 pp XP002111193 1997, NY, NY IEEE, USA9SBN abstract paragraph 0003 figures 1,3.
Patent Abstracts of Japan vol. 016, No. 463 (P-1428), Sep. 25, 1992 & JP 04 165422 A (Fujitsu Ltd; Others: 01) Jun. 11, 1992.
DE 198 39 627 A (LG Electronics Inc). Mar. 4, 1999 abstract claims 1-8 figures 3-5-8-.
EP 0 469 841 A (Matsushita Electric Ind Co Ltd) Feb. 5, 1992 abstract figure 1.
Davidson G et al: “A Lowcost-Adaptive Transform Decoder Implementation for High Quality Audio” Speech Processing 2, Audio, Neural Networks, Underwater Acoustics, San Francisco, Mar. 23-26, 1992, vol. 2, No. Conf. 17, Mar. 23, 1992 pp. 193-196, XP000356970 IEEE ISBN; 0-7803-0532-9.
Patent Abstracts of Japan vol. 015 No. 295 (E-1094), Jul. 26, 1991 & JP 03 104310 A (Matsushita Electric Ind Co Ltd), May 1, 1992 Abstract.
PCT Search Report for our reference SAR 12949 Application No. PCT/US99/08508 filing date Apr. 23, 1999.
Branco Richard Gerald
Li Shipeng
Burke William J.
Sarnoff Corporation
{haeck over (S)}mits T{overscore (a)}livaldis Ivars
LandOfFree
Method and apparatus for implementing rounding in decoding... 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 and apparatus for implementing rounding in decoding..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for implementing rounding in decoding... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2850040