Coded data generation or conversion – Digital code to digital code converters
Reexamination Certificate
2002-09-25
2004-01-13
Jeanglaude, Jean Bruner (Department: 2819)
Coded data generation or conversion
Digital code to digital code converters
C341S051000
Reexamination Certificate
active
06677865
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to a method and a system for decoding information that is output by an information source, coded and transmitted via a transmission channel affected by interference.
Source signals such as speech or video signals generally contain statistical redundancy. This redundancy can be reduced by source coding so as to permit efficient transmission and storage of the source signal. It is known to add redundancy specifically at the transmitting end by channel coding in order to cancel channel interference at the receiving end.
Because of incomplete knowledge of the source signals or restrictions associated with the complexity of the coding method, it is usually possible to implement source coding only suboptimally; that is to say, there is still a certain redundancy present in the compressed data. This residual redundancy can be utilized as a priori/a posteriori information in the case of so-called source-controlled or joint source and channel decoding in order to correct further bit errors. In this case, the decoding operation of the channel decoder is controlled both by the transmitted information and by a priori/a posteriori information on the likely value of a few important source bits. A method based on a Kalman filter can be used, for example, to estimate the a priori/a posteriori information.
However, it has emerged that the use of this residual redundancy, or the a priori/a posteriori information, can lead to decoding errors during channel decoding in the course of source-controlled channel decoding. Thus, the application of the a priori/a posteriori information leads to impairment of the decoding when the value of a bit position that is to be transmitted deviates from the statistical a priori/a posteriori information on the value of this bit position.
It is, therefore, on object of the present invention to render decoding of information possible by applying statistical information via the information source with satisfactory decoding results.
SUMMARY OF THE INVENTION
Thus, the present invention is based on the principle, for source-controlled channel decoding, of modifying the influence of the a priori or a posteriori information on the source statistics as a function of the reliability of the decoding.
In order to decide on a value assigned to a bit position, alongside the transmitted information, statistical information is applied via the information source, the influence of the statistical information via the information source on the decision being modified as a function of the reliability of the decoding.
For example, a value can be determined that characterizes the reliability of the decoding and with which the statistical information via the information source is weighted; in particular, multiplied.
It is thereby possible, depending on the current reliability of the decoding, to accord the statistical information via the information source a greater or lesser influence on the decision on the value to be assigned to a bit position.
One embodiment of the present invention provides that the decoding, particularly the channel decoding, has a soft output that constitutes a measure of the reliability of the decoding. The value of a soft output can directly or indirectly modify the influence of the statistical information via the information source on the decoding. Soft-output decoding methods, such as soft-output Viterbi algorithms or soft-output maximum likelihood decoding methods, are known as such to a person skilled in the art.
A number of consecutive decoding steps also can be carried out for decoding purposes. Another embodiment of the present invention provides that a first decoding step is carried out without applying statistical information via the information source, a second decoding step is carried out by applying statistical information via the information source, and the influence of the statistical information via the information source on the decoding is modified as a function of the reliability of the decoding in the first decoding step.
It is possible, in this case, for the first decoding step to relate to the same bit position as the second decoding step, or the first decoding step to relate to a first bit position and the second decoding step to relate to a second bit position; for example, a subsequent frame. Particularly owing to introduction into an iterative decoding method such as turbo decoding, for example, this leads, as shown by complex simulations, to good decoding results without substantially increasing the computational outlay for the decoding.
Finally, another embodiment of the present invention provides that the influence of the statistical information via the information source on the decoding is relatively strong in conjunction with relatively low reliability of the decoding, whereas it is relatively small in conjunction with relatively high reliability of the decoding.
Additional features and advantages of the present invention are described in, and will be apparent from, the following detailed description of the invention and Figures.
REFERENCES:
patent: 4670890 (1987-06-01), Titchener
patent: 5218622 (1993-06-01), Fazel et al.
patent: 6292922 (2001-09-01), Ruscitto et al.
Bell Boyd & Lloyd LLC
Jeanglaude Jean Bruner
Lauture Joseph
LandOfFree
Method and configuration for decoding information 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 configuration for decoding information, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and configuration for decoding information will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3215990