Method of error correction

Claim 3

A method for use in detecting and correcting errors according to claim 1, including the steps of selecting ones of said signals representative of error syndrome words and ones of said signals representative of error constant words to have a zero signal level, determining whether said zero signal levels are present, and declaring that no error words are detected.

Claim 4

A method for use in detecting and correcting errors according to claim 1, including the steps of selecting ones of said signals representative of error syndrome words to have a nonzero signal level, selecting ones of said signals representative of error constant words to have a zero signal level, determining whether said respective zero and nonzero signal levels are present, and declaring the presence of one word error for correction in accordance with said error pattern signal.

Claim 5

A method for use in detecting and correcting errors according to claim 1, including the steps of selecting ones of said signals representative of error constant words to have nonzero signal levels, determining whether said nonzero signal levels are present, declaring the presence of two words error, generating a second plurality of signals representative of error constant words, and solving an error location equation using said second plurality of signals representative of error constant words to produce said error pattern signals for correcting said two words error.

Claim 6

A method for use in detecting and correcting errors according to claim 4, in which the step of generating a plurality of signals representative of error syndrome words includes the step of k syndrome words, S 0 to S k-1 , by multiplying one block V T consisting of n received words by a parity check matrix H ##EQU20## where the parity check matrix H has n columns and k rows, and in which each element of one predetermined row is selected from α 0 (=1) to α 2m-2 , α being a root to satisfy F(x)=0 when F(x) is an irreducible polynomial on Galois field GF(2), so that the same value does not appear twice in said predetermined row, and wherein the elements in the remaining rows are selected to be a given power, for all the elements in each respective row, of the corresponding elements in said predetermined row.

Claim 7

A method for use in detecting and correcting errors according to claim 5, in which the step of generating a plurality of signals representative of error constant words includes the further steps of obtaining error constant words A, B, and C utilizing said syndromes, so that: ##EQU21##

Claim 8

A method for use in detecting and correcting errors according to claim 6, in which said step of determining whether preselected equalities or inequalities are satisfied includes the steps of determining: (a) when S 0 =S 3 =S 4 = . . . =S k-1 =0, A 1 =A 2 = . . . =A k-3 =0, B 1 =B 2 = . . . =B k-3 =0, and C k-3 =0 are satisfied and producing a signal indicating that there is no error word, (b) when S 0 ≠0, S 3 ≠0, S 4 ≠0, . . . , S k-1 ≠0, A k' =0, B k' =0, where k'=1 to k-3 and C k-3 =0 are satisfied, and producing a signal indicating that there is one word error, and (c) when A k' ≠0, B k' ≠0 and C k-3 ≠0 are satisfied and ##EQU22## and producing a signal indicating that there are two words in error.

Claim 9

A method for use in detecting and correcting errors according to claim 7, in which the step of correcting located errors includes the further steps of generating error location equation α 2i +Dα i +E=0 and solving same to detect error locations i and j so that two word errors are corrected.

Claim 12

A method for use in detecting and correcting errors according to claim 9, including the steps of establishing ratios between selected ones of said generated signals representative of error syndrome words, selecting a pattern of elements in the parity check matrix, memorizing the selected pattern of elements of the parity check matrix in a read only memory, and comparing the established ratios of signals representative of error syndrome words with the memorized pattern to determine the location of one word error, and generating the error pattern signals to be equal to the first one of said generated signals representative of error syndrome words.

Claim 13

A method for use in detecting and correcting errors according to claim 9, in which the step of correcting the one or two words error includes the steps of determining that two words are in error, selecting relationships between the signals representative of error syndrome words and an error pattern signal specific to each of the two words error, determining the error pattern signals specific to each error word based upon a selected relationship between first and second ones of said generated signals representative of error syndrome words and a pattern of elements forming the parity check matrix.

Claim 14

A method for use in detecting and correcting errors according to claim 9, in which the step of generating a number of signals representative of error syndrome words includes the steps of obtaining k syndrome words, S 0 to S k-1 , by multiplying one block V T of n received digital data words and said parity check matrix H, so that ##EQU23## where the parity check matrix H has n columns and k rows, and in which each element of one predetermined row is selected from α 0 (-1) to α 2m-2 , the element α being a root to satisfy F(x)=0 when F(x) is an irreducible polynomial on a Galois field GF(2), so that the same value does not appear twice in said predetermined row, and wherein the elements in the remaining rows are selected to be a given power, for all the elements in each respective row, of the corresponding elements in said predetermined row.

Claim 15

A method for use in detecting and correcting errors according to claim 12, in which the step of generating a plurality of signals representative of error constant words includes the further steps of obtaining error constant words A, B, and C based upon said syndromes, so that: ##EQU24##

Claim 16

A method for use in detecting and correcting errors according to claim 13, in which the step of determining whether the generated signals representative of error words all have zero signal level includes the steps determining when S o =S 3 =S 4 . . . =S k-1 =0, A 1 =A 2 = . . . =A k-3 =0, B 1 =B 2 = . . . =B k-3 =0 and C k-3 =0 are satisfied and producing a signal indicating that there is no error word.

Claim 17

A method for use in detecting and correcting errors according to claim 13, in which the step of determining whether all generated signals representative of error syndrome words have a nonzero signal level includes the steps of determining when S 0 ≠0, S 3 ≠0, S 4 ≠0, . . . , S k-1 ≠0, A k' =0, B k' =0 are satisfied, where k'=1 to k-3 and C k-3 =0, and producing a signal indicating there is one word error.

Claim 18

A method for use, in detecting and correcting errors according to claim 13, in which said step of determining that not more than two words error are present includes the steps of determining when A k' =0, B k' =0 and C k-3 =0 are satisfied, generating respective signals according to: ##EQU25## and solving an error location equation of α 2i +Dα i +E=0 to detect two word error locations i and j.

Claim 19

Apparatus for use in performing error detection and correction on a received coded digital data signal to which a selected number of error check words have been added comprising: memory means for memorizing said received digital data signal; error syndrome word generating means connected to receive said digital data signal for generating a plurality of error syndrome words therefrom; calculating circuit means having first inputs connected to receive said plurality of error syndrome words for generating a first plurality of error constant words therefrom and having second inputs connected to receive words for generating a plurality of error patterns therefrom; first buffer register means connected to receive said first plurality of error constant words from said calculating circuit means; read only memory means containing first and second roots of an irreducible polynominal on Galois field GF(2); error location decoder means connected to receive said first plurality of error constant words from said first buffer register means and said first and second roots from said read only memory means for producing a second plurality of error constant words and error location words; error judging means connected to receive said first plurality of error constant words from said first buffer register means and said plurality of error syndrome words for determining whether there is an error, whether the error is one word error, whether the error is two word errors, whether the error is more than two word errors and producing a control signal to which said calculating circuit means and said error location decoder means are responsive; means for feeding said second plurality of error constant words, said error location words, and at least one of said first plurality of error constant words to said second inputs of said calculating circuit means; second buffer register means receiving said error pattern signals from said calculating circuit means; and error correction means connected to the output of said memory means and to the outputs of said second buffer register means for correcting the output of said memory means in accordance with said error pattern signal from said second buffer register means.

Claim 20

Apparatus for use in performing error detection and correction on a received coded digital data signal to which a selected number of error check words have been added comprising: memory means for memorizing said received digital data signal; error syndrome word generating means connected to receive said digital data signal for generating a plurality of error syndrome words therefrom; calculating circuit means having first inputs connected to receive said plurality of error syndrome words for generating a first plurality of error constant words therefrom and having second inputs connected to receive words for generating a plurality of error patterns therefrom; first buffer register means connected to receive said first plurality of error constant words from said calculating circuit means; read only memory means containing first and second roots of an irreducible polynominal on Galois field GF(2); error location decoder means connected to receive said first plurality of error constant words from said first buffer register means and said first and second roots from said read only memory means for producing a second plurality of error constant words and error location words; error judging means connected to receive said first plurality of error constant words from said first buffer register means and said plurality of error syndrome words for determining whether there is an error, whether the error is one word error, whether the error is two word errors, whether the error is more than two word errors and producing a control signal to which said calculating circuit means and said error location decoder means are responsive; means for feeding said second plurality of error constant words, said error location words, and at least one of said first plurality of error constant words to said second inputs of said calculating circuit means; second buffer register means receiving said error pattern signals from said calculating circuit means; and error correction means connected to the output of said memory means and to the outputs of said second buffer register means for correcting the output of said memory means in accordance with said error pattern signal from said second buffer register means.

Claim 21

Apparatus for use in performing error detection and correction according to claim 17, further comprising controller means for supplying synchronizing clock pulses to said error syndrome word generating means, said error judging means, said calculating circuit means, said first and second buffer register means, said error location decoder means, and said read only memory means.

Claim 22

Apparatus for use in performing error detection and correction according to claim 17, in which said error syndrome word generating means includes syndrome calculation means for obtaining k syndrome words S 0 to S k-1s by calculating one block V T of received n words in said coded digital data signal by a parity check matrix H ##EQU26## where the parity check matrix H has n columns and k rows, in which each element of one predetermined row is selected from α 0 (=1) to α 2 .spsp.m -2 , α being a root to satisfy F(x)=0, where F(x) is an irreducible polynomial, on a Galois field GF(2), so that the same value does not appear twice in said predetermined row, and wherein the elements in the remaining rows are selected to be a given power, for all the elements in each respective row, of the corresponding elements in said predetermined row.

Claim 23

Apparatus for use in performing error correcting and detection according to claim 19, in which said calculating circuit means includes for obtaining the error constant words A, B and C so that ##EQU27##

Claim 24

Apparatus for use in performing error detection and correction according to claim 20, in which said error judging means includes syndrome equality determination means for determining when S 0 =S 3 =S 4 = . . . =S k-1 =0, A 1 =A 2 = . . . =A k-3 =0, B 1 =B 2 = . . . =B k-3 =0, and C k-3 =0 are satisfied, and producing a signal indicating that there is no error word.

Claim 25

Apparatus for use in performing error detection and correction according to claim 20, in which said error judging means includes syndrome equality determination means for determining when S 0 ≠0, S 3 ≠0, S 4 ≠0,= . . . , S k-1 ≠0, A k' =0, B k' =0, are satisfied and producing a signal indicating that there is one word error.

Claim 26

Apparatus for use in performing error detection and correction according to claim 20, in which said error judging means includes syndrome equality determination means for determining when A k , ≠0, B k' ≠0 and C k-3 =0 are satisfied and generating followings be assumed ##EQU28## means for generating error location equation of α 2i +Dα i +E=0, and means for solving said error location equation to detect two word error locations i and j.