floating point arithmetic

hi mates. may peace and blessing be upon all of you.
experts, i am trying to implement a soft decision turbo decoder using
verilog. however i am stuck. i dont know how to deal with the floating
point numbers. i was thinking on implementing it using fixed point
logic. but the results are not right on the paper, forget about
modelsim . any ideas or refrences are welcome. i will be grateful
if you help me in this mini project.
umar

 

[glen herrmannsfeldt]

m.umar.kk@gmail.com wrote:

experts, i am trying to implement a soft decision turbo decoder using
verilog. however i am stuck. i dont know how to deal with the floating
point numbers. i was thinking on implementing it using fixed point
logic.

Can you do it in scaled fixed point? (That is, with the binary
point not directly to the right of the least significant bit.)

-- glen