Implementation Aspects of the Matrix Inverse Computation and Inverse Computations Completness Method

摘要

It becomes interesting to analyze and apply the features of reconfigurable computations because the concept allows algorithm creation. An important algorithm that resulted from this concept is the known matrix inverse computation. This algorithm is becoming common for matrix based computations because it is free from singularities in comparison with other "Gauss" methods. However, this extreme ease computational requirement has a limitation. The generated matrix inverse are all upper triangular or lower triangular. Since there is a need to extend computations to any matrix, we then present some implementation aspects of the reconfigurable matrix inverse and an extension of the process that handles full matrix inverse. This research uses the results of the reconfigurable matrix inverse computations completes them and makes the process capable of generating full matrix inverse.