MATLAB has a function fixdt()
that creates objects describing fixed-point or floating-point data type. So if we write
a = fixdt(0,16,26)
MATLAB will return
NumericType with properties:
DataTypeMode: 'Fixed-point: binary point scaling'
Signedness: 'Unsigned'
WordLength: 16
FractionLength: 26
IsAlias: 0
DataScope: 'Auto'
HeaderFile: ''
Description: ''
My question is, is it even possible to have a fixed point data type, with fraction length exceeding the word length. Doesn't the fraction length represents the location of the binary point from LSB? I'm trying to fully understand what exactly a data type of word length 16 with fraction length 26 exactly represent. Many thanks for your inputs!