| 摘要 |
This invention is a method and apparatus for interpolation which enables simpler and cost efficient implementation in hardware or software. A function table stores values of the function at addresses corresponding to the argument points where the function is known. The input value enables identification of the function values for arguments immediately below and above the input value. Respective bits of the absolute value of the difference between these two function values enables corresponding gradient value tables. A set of gradient values are stored in these gradient value tables. The least significant bits of the input value, those bits less significant than the arguments of the stored function values, address entries in the enabled gradient value tables. The desired interpolation value is the sum of the first function value and the gradient value recalled from the gradient tables.
|
