| 摘要 |
<p>PROBLEM TO BE SOLVED: To compress an analog signal with high precision regardless of a simple circuit configuration without the need for huge arithmetic processing. SOLUTION: A function close to a 1st pattern 11 is generated by using samples at prescribed sampling points included in the 1st pattern, a 2nd pattern consisting of samples at each sampling point is generated by using the function, and new samples are calculated by using the 1st and 2nd patterns so as to generate a 3rd pattern 12 where new samples at least at optional sampling points are zero. The 3rd pattern 12 is compared with each reference pattern 14 and a reference pattern 14 closest to the 3rd pattern is selected and a pattern code 16 provided to the reference pattern 14 is outputted as compression data of the 1st pattern 11 for each sampling period.</p> |
