Analytic Johns matrix and its applications in TLM diakoptics

The storage of a general TLM Johns matrix is large because the dimension of Johns matrix is dependent on the iteration number. The time-domain convolution with Johns matrix is computationally intensive as the whole time history of the incident impulses is involved. In this paper, analytic Johns matrix is derived. A recursive convolution formula is then developed, resulting in a considerable reduction of the computation count.<>