How to implement floating point using 16 bit words

Question

How would I implement floating point operations if I only have memory locations that are 16 bits wide?

I am trying to implement IEEE-754 32-bit single precision floating point.

Thank you

Answer 1

I ended up figuring out that there is no easy way to to this. I had to create my own software library for all of the functions in assembly.
If anybody has questions regarding how to do this, I implemented add,subtract,multiply,square root, divide, cosine, and sine.

Answer 2

There is a 16bit version of IEEE754