| 摘要 |
If a signal is repeatedly encoded and decoded, coding noise generally accumulates as it passes through the various coding systems. The present system provides means for preventing this, by providing means for identifying, at the second coding system, which coding system was previously used. The second coding can then use the same system as the previous coding. This use of matching coding systems results in very little extra coding noise being introduced by the second coding. The input signal is coded by two different encoders (11 and 12), their outputs are decoded and compared with the input signal at (20 and 21), the differences accumulated at (22 and 23). The smaller of the differences is used by selector (13) to select the encoder which matches the previous coding of the signal. The principles can be extended to MPEG Layer II Audio Coding and similar block coding systems. |
