Optimal inverse projection of floating-point addition

An Improved Floating Point Addition Algorithm

Floating point addition/subtraction has been designed in literature. The methods involved two’s complements add/subtract logic and XOR for counting leading zeroes. This paper proposes algorithm to perform add/subtract operation using one’s complement and counting leading zeroes for normalization using NOR operation as reduction operator for two bits of mantissa at a time. The proposed model is ...

Optimal Compression of Floating-Point FITS Images

Lossless compression (e.g., with GZIP) of floating-point format astronomical FITS images is ineffective and typically only reduces the file size by 10% to 30%. We describe a much more effective compression method that is supported by the publicly available fpack and funpack FITS image compression utilities that can compress floating point images by a factor of 10 without loss of significant sci...

Improving the Floating Point Addition and Subtraction Constraints

Solving constraints over the oating point numbers is a key issue in the process of software validation and veri cation. Techniques to solve such constraints on the basis of projection functions have been successfully developed. However, though correct, this approach can lead to slow convergence phenomena for very common constraints like addition and subtraction constraints. In this paper, we in...

An Improved Algorithm for High-speed Floating-point Addition

This paper describes an improved, IEEE conforming oating-point addition algorithm. This algorithm has only one addition step involving the signi cand in the worst-case path, hence o ering a considerable speed advantage over the existing algorithms, which typically require two to three addition steps.

Exact Projection Functions for Floating Point Number Constraints