PROCEDE ET APPAREIL DE TRAITEMENT DE SIGNAUX NUMERIQUES

• Main Author: YOSHITAKA HASHIMOTO, NORIHISA SHIROTA ET KAICHI YAMAMOTO
• Format: Patent
• Language: French
• Published: 13.06.1986
• Edition: 3
• Subjects: INFORMATION STORAGE; INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORDCARRIER AND TRANSDUCER; PHYSICS

A method for processing a digital color video signal formed of 8-bit words, includes the recording steps of converting each 8-bit word into a 10-bit word in accordance with a predetermined mapping function, grouping the 10-bit words into 48 word blocks, adding to each block an identification signal, identification signal check words P1 and Q1, and data check words P2 and Q2, the latter being formed by the equations: where T-1, T-2, . . . T-(n-1), T-n, T1, T2, . . . Tn-1, Tn are distinct, non-zero elements of a Galois field (2m); and includes the reproducing steps of reconverting each of the 10-bit words to its corresponding 8-bit word in accordance with the predetermined mapping function with the reconverted 8-bit words forming a data unit for each 12 words thereof and a data block for each 48 words thereof, detecting whether any of the 8-bit words contain an error by failing to satisfy the mapping function reconversion, determining that each data unit that includes an 8-bit word having an error is erroneous, detecting and correcting any errors in the identification signals by means of the identification signal check words P1 and Q1, determining whether each erroneous data block and data unit is capable of being corrected and concealing each erroneous data unit that is determined to be uncorrectable and correcting errors in the 8-bit words in each erroneous data block and data unit that is determined to be correctable, by means of the data check words P2 and Q2 and parity words, respectively, thereby forming a continuous, composite digital color video signal, and correcting the phase of the color sub-carrier of each data unit of the composite digital color video signal by comparison of a reference signal with the identification signal of the block to which each respective data unit belongs.