Channel coding method with corresponding code and demodulation and decoding method and means
Saved in:
| Title: | Channel coding method with corresponding code and demodulation and decoding method and means |
|---|---|
| Patent Number: | 7,411,524 |
| Publication Date: | August 12, 2008 |
| Appl. No: | 11/573321 |
| Application Filed: | August 08, 2005 |
| Abstract: | A coding/decoding system and method are disclosed. Coding, as used herein, refers assigning values to cells. Information to be coded is processed stepwise in information pieces, with bitwise processing (single information bits as information pieces) as a special case. According to an illustrative embodiment, bitwise coding/decoding is disclosed. A predefined structure called a configuration is known to both the coding and decoding systems. The configuration includes a step configuration for each coding step. A step configuration is: 1) a distinction of two subsets of cells, one called inversion cells, the other called non inversion cells; 2) an assignment of at least one cell tuple consisting of at least one inversion cell and one non inversion cell each, where, for these tuples, for any preceding step configuration the tuple cells are either all inversion cells or all non inversion cells. For coding, in each step, its inversion cells are XORed with the information bit. For decoding, the information bits are recovered in reverse order. In each step, its step configuration is used; the information bit is recovered by tuplewise comparing inversion cells with non inversion cells. If, for a majority of cell tuples, the cells are different, the information bit is set, otherwise it is not set. If it is set, the inversion cells are inverted for further decoding steps, if any. The method provides for easy application specific integration of demodulation into decoding and for introduction of soft decision methods. Robust decoding is achieved by means of majority decisions. |
| Inventors: | Tropf, Ing. Hermann (68789 St. Leon-Rot, DE) |
| Claim: | 1. A method for coding information presentable by a plurality of pieces of partial information, comprising the assignment of cell values to cells, comprising: a) initialising each of the cells in that each cell is assigned a predetermined cell value; b) providing a predetermined configuration comprising a sequence of step configurations, wherein each step configuration comprises: (b1) dividing the cells into at least one inversion cell (i) and at least one non-inversion cell (n), wherein the division is specific for a partial information, and (b2) determining at least one cell tuple from the cells of the division from step (b1), wherein each cell tuple comprises at least one inversion cell (i) and at least one non-inversion cell (n), wherein the cell tuple(s) is/are specific for the partial information, and wherein for each of these tuples, the cells in any of the preceding step configurations are either all inversion cells (i) or all non-inversion cells (n); and (c) coding, further comprising (c1) using a step configuration, (c2) XOR combining the partial information, which is specific for the coding step, with the cell value of each of the inversion cells (i) of the step configuration and assigning the result as new cell value for the respective inversion cell (i), wherein the cell value of the non-inversion cells (n) remains unchanged, and (c3) repeating the steps (a) and (b) for the further step configurations of the configuration. |
| Claim: | 2. The method for coding according to claim 1 , wherein the cell tuples are cell pairs with one inversion cell and one non-inversion cell. |
| Claim: | 3. The method for coding according to claim 1 , wherein the cell values are assigned signal values in the form of pixel values of a picture. |
| Claim: | 4. The method for coding according to claim 1 , wherein a plurality of pixels are assigned to one cell, wherein the pixels are the picture elements of at least one of a picture and a picture sequence and the signal values of any other signal, characterised by the assignment of pixel patterns to the cell values. |
| Claim: | 5. The method for coding according to claim 1 , characterised by an XOR combination of the result with a predetermined pattern, wherein the predetermined pattern is a two-dimensional chequerboard pattern for a picture and a binary type 101010 . . . pattern for a one-dimensional signal. |
| Claim: | 6. The method for coding according to claim 1 , characterised by: for at least one coding step a placing of the cells, wherein at least three sets of cell tuples are formed which are separated from each other by at least one cell. |
| Claim: | 7. A coding system used for coding information presentable by a plurality of pieces of partial information, comprising the assignment of cell values to cells, comprising: (a) means for initialising each of the cells, wherein each of the cells is assigned a predetermined cell value, and (b) a predetermined configuration comprising a sequence of step configurations, and wherein each step configuration comprises: (b1) means for dividing the cells into at least one inversion cell (i) and at least one non-inversion cell (n), wherein the division is specific for a partial information; and (b2) means for determining at least one cell tuple from the cells of the division from step (b1), wherein each cell tuple comprises at least one inversion cell (i) and at least one non-inversion cell (n), wherein the cell tuple(s) is/are specific for the partial information, and wherein for each of these tuples, the cells in any of the preceding step configurations are either all inversion cells (i) or all non-inversion cells (n); (c) means for using a step configuration; (d) means for XOR combining the partial information, which is specific for the coding step, with the cell value of each of the inversion cells (i) of the step configuration and assigning the result as new cell value for the respective inversion cell (i), wherein the cell value of the non-inversion cells (n) remains unchanged, wherein coding is carried out for all step configurations of the configuration. |
| Claim: | 8. The coding system according to claim 7 including a plurality of registers comprising the cells, wherein the registers are arranged in a pipeline. |
| Claim: | 9. A method for decoding information presentable by a plurality of pieces of partial information, comprising: (A) providing a predetermined configuration comprising a sequence of step configurations, and wherein each step configuration comprises: (A1) dividing the cells into at least one inversion cell (i) and at least one non-inversion cell (n), wherein the division is specific for partial information, and (A2) determining at least one cell tuple from the cells of the division from step (A1), wherein each cell tuple comprises at least one inversion cell (i) and at least one non-inversion cell (n), wherein the cell tuple(s) is/are specific for the partial information, and wherein for each of these tuples, the cells in any of the preceding step configurations are either all inversion cells (i) or all non-inversion cells (n), and (B) wherein the decoding steps are in reverse order of the step configurations, comprising: (B1) using a step configuration, (B2) comparing the inversion cell(s) of the step configuration with the non-inversion cell(s) of the at least one cell tuple, and comparison-dependent determining of the respective partial information, (B3) as long as at least one decoding step follows, depending on the partial information thus determined, inverting each of the inversion cells (i), wherein the non-inversion cell(s) remain(s) unchanged, and (B4) repeating the steps (B1) to (B3) for the further step configurations of the configuration. |
| Claim: | 10. The decoding method according to claim 9 , characterised in that the comparison of inversion cells with non-inversion cells is effected by comparison of cell values, which are deduced from the cells. |
| Claim: | 11. The decoding method according to claim 9 , characterised in that the comparison of inversion cells with non-inversion cells is effected by comparison of the contents of the respective cells. |
| Claim: | 12. The decoding method according to claim 9 , characterised in that a comparison result can be obtained by evaluating a number of comparisons of cell values according to a quality function, by at least one of a) the median of individual comparison measures, and b) the arithmetic mean of individual comparison measures. |
| Claim: | 13. The decoding method according to claim 9 , characterised in that a comparison result can be determined by evaluation of cell values according to a quality function, by the comparison of mean values among inversion cells on the one hand and mean values among non-inversion cells on the other hand. |
| Claim: | 14. The decoding method according to claim 9 , characterised in that a comparison result is determined by analysing at least one of: a) the median of comparison functions with respect to several pixels of the corresponding cells, and b) the arithmetic mean of comparison functions with respect to several pixels of the corresponding cells. |
| Claim: | 15. The decoding method according to claim 9 , further wherein (5) the cells are interpreted as an error picture or an error signal for a quality assessment. |
| Claim: | 16. The decoding method according to claim 9 , including a multi-valued logic with the coded cell values and a corresponding multi-valued phase modulation at signal level, further characterised by a multi-valued fuzzy logic corresponding to the cell values to be decoded. |
| Claim: | 17. A decoding system used for decoding information presentable by a plurality of pieces of partial information, comprising: (A) means for providing a predetermined configuration comprising a sequence of step configurations, comprising: (A1) means for dividing the cells into at least one inversion cell (i) and at least one non-inversion cell (n), wherein the division is specific for partial information, and (A2) means for determining at least one cell tuple from the cells of the division from step (A1), wherein each cell tuple comprises at least one inversion cell (i) and at least one non-inversion cell (n), wherein the cell tuple(s) is/are specific for the partial information, and wherein for each of these tuples, the cells in any of the preceding step configurations are either all inversion cells (i) or all non-inversion cells (n), and (B1) means for using a step configuration, (B2) means for comparing the inversion cell(s) of the step configuration with the non-inversion cell(s) of the at least one cell tuple, and comparison-dependent determination of the respective partial information, (B3) means for inverting each of the inversion cells (i) depending on the partial information thus determined as long as at least one decoding step follows, wherein the non-inversion cell(s) remain(s) unchanged, wherein decoding for all step configurations of the configuration is carried out in reverse order of the step configurations. |
| Claim: | 18. The decoding system according to claim 17 , further comprising a plurality of registers each comprising one or more register cell of which at least one register is used to receive the information to be decoded, and wherein at least one register cell is invertible by decoding steps depending on the decoded partial information by the XOR gate, wherein at least an invertible register cell is interpretable as an inversion cell from one or more cell tuples including inversion cells and non-inversion cells. |
| Claim: | 19. The decoding system according to claim 17 , characterised by summation and threshold value elements with which a decision can be taken about the partial information to be decoded. |
| Claim: | 20. The decoding system according to claim 17 , wherein at least one comparison is realised by a fuzzy XOR gate. |
| Claim: | 21. The method according to claim 9 , wherein the information has been coded according to a method with a predetermined configuration according to claim 1 . |
| Claim: | 22. The system according to claim 17 , wherein the information has been coded according to a method with a predetermined configuration according to claim 1 . |
| Current U.S. Class: | 341/51 |
| Patent References Cited: | 5324923 June 1994 Cymbalski et al. 5446747 August 1995 Berrou 5553084 September 1996 Ackley et al. 5559506 September 1996 Leitch 5591956 January 1997 Longacre et al. 5861922 January 1999 Murashita et al. 6371373 April 2002 Ma et al. 7085987 August 2006 Hewitt et al. 2003/0016770 January 2003 Trans et al. 2004/0021588 February 2004 Luby 2005/0138533 June 2005 Le Bars et al. 2006/0212781 September 2006 Hewitt et al. 690 24 045 December 1995 695 26 915 February 1996 691 18 891 April 1996 199 26 197 August 2000 199 26 194 December 2000 103 07 775 September 2003 1 048 113 December 1998 |
| Other References: | Anonymous: “XOR-Verknupfung” Internet Article, [Online] Jul. 18, 2004, XP002348227 Retrieved from the Internet: URL:http://de.wikipedia.org/w/index.php?title=. cited by other Petrowitsch H T et al: “Die Uebertragung Von Signalen Mit Relativen Uebertragungsverfahren (Differenzmodulation)” Nachrichten Technik, Veb Verlag Technik. cited by other Schouhamer Immink K A: “Coding Techniques for Digital Recorders” 1991. Prentice Hall, XP002348229 p. 183-184. cited by other The Art of Error Correcting Coding Robert H. Morelos-Zaragoza, Sony Computer Science Laboratories, Inc. Japan John Wiley & Sons, Ltd. cited by other |
| Primary Examiner: | Nguyen, Linh V |
| Attorney, Agent or Firm: | Greener, William Bond, Schoeneck & King PLLC |
| Accession Number: | edspgr.07411524 |
| Database: | USPTO Patent Grants |
| Abstract: | A coding/decoding system and method are disclosed. Coding, as used herein, refers assigning values to cells. Information to be coded is processed stepwise in information pieces, with bitwise processing (single information bits as information pieces) as a special case. According to an illustrative embodiment, bitwise coding/decoding is disclosed. A predefined structure called a configuration is known to both the coding and decoding systems. The configuration includes a step configuration for each coding step. A step configuration is: 1) a distinction of two subsets of cells, one called inversion cells, the other called non inversion cells; 2) an assignment of at least one cell tuple consisting of at least one inversion cell and one non inversion cell each, where, for these tuples, for any preceding step configuration the tuple cells are either all inversion cells or all non inversion cells. For coding, in each step, its inversion cells are XORed with the information bit. For decoding, the information bits are recovered in reverse order. In each step, its step configuration is used; the information bit is recovered by tuplewise comparing inversion cells with non inversion cells. If, for a majority of cell tuples, the cells are different, the information bit is set, otherwise it is not set. If it is set, the inversion cells are inverted for further decoding steps, if any. The method provides for easy application specific integration of demodulation into decoding and for introduction of soft decision methods. Robust decoding is achieved by means of majority decisions. |
|---|