Keep sign when using bitsliceget()

3 views (last 30 days)
Lennart
Lennart on 1 Jul 2015
Edited: Krishnan Gopal on 24 Jul 2019
Hi there,
I want to extract 8 bits from a 16bit fixed point number using bitsliceget(). Unfortunately the result is always an unsigned variable, which is not what I want.
Here is a little example:
x = fi(-0.12835,true, 16,10);
x.bin
ans = 1111111101111101
y = bitsliceget(x,9,2)
ans = 190
DataTypeMode: Fixed-point: binary point scaling
Signedness: Unsigned
WordLength: 8
FractionLength: 0
y.bin
ans = 10111110
bitsliceget(x,9,2) actually does exactly what I want it to but I just can't get y to become a signed variable without recasting and therefore changing the binary representation.
Any help would be highly appreciated.
Thanks it advance for taking the time!
Regards,
Len

Answers (1)

Krishnan Gopal
Krishnan Gopal on 24 Jul 2019
Edited: Krishnan Gopal on 24 Jul 2019
Hi
The result of bitsliceget is always interpreted as an unsigned type, so you can use the 'reinterpretcast' function to re-evaluate it as a numeric signed type. In your case, the you should cast it to a signed numerictype with 8 bits as shown:
y = reinterpretcast(bitsliceget(x,9,2), numerictype(1,8,0));
regards
Krishnan

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!