| 摘要 |
<p>The invention concerns a method for converting a physical quantity (Qi) measured by a measuring apparatus, into pulses representing each an elementary quantity P. The method comprises the following steps: measuring the physical quantity (Qi) at a sampling frequency (fe); calculating and coding over a number (r) of bits, an integer (ni) corresponding to the physical quantity(Qi); adding (mi) times the integer (ni) to itself at an adding frequency (f), using adding means whereof the storage capacity is limited to t = 2r, (mi) corresponding to the number of additions to be performed to exceed the storage capacity (t) of the adding means for a given value of (ni); when the storage capacity of the adding means is exceeded, generating an electric pulse and a fraction error (xi) when the ratio Qi/P is not equal to an integer, such that xi = (mi x ni)/t; transferring the fragmentation error (xi) to an input of the adding means to add it to the integer (ni) if no new physical quantity has been measured, or to a new ni+1 representing a new measured physical quantity Qi+1 if a new physical quantity has been measured; and repeating the step generating an electric pulse and a fraction error in the adding means output, and the step of transferring the fraction error to the input of the adding means.</p> |
