| 摘要 |
PROBLEM TO BE SOLVED: To encode a numerical data string composed of fixed length bits by the use of a variable length bit string, and decode it by parallel processing using a plurality of decoding units.SOLUTION: As shown in Figure (a), identification codes C1, C2, ... indicating bit length b (circled numbers) are added to heads of corresponding real data D1, D2, ... composed of variable length bits to create unit code data U1, U2, .... The use of codes shown in a coding table shown in Figure (b) as identification code C indicating bit length b produces encoded data shown in Figure (c). With bit length of identification code C itself being a, the table satisfies the condition of a+b=k×N, where k is an arbitrary integer, and N is an integer of two or greater and, specifically, N=4 in this example. Since boundaries of the unit code data U1, U2, ... shown in Figure (c) are necessarily integral multiples of N bits from the head, a plurality of divisions are made along such boundaries for decoding by parallel processing. |
