Temporal Localization And Systolization Of Signal Flow Graph (SFG) Computing Networks
Sun‐Yuan Kung, S. C. Lo, Jurgen Annevelink · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 1984
This paper addresses the theoretical and algorithmic issues related to optimal temporal localization (and systolization) of Signal Flow Graph (SFG) computing networks. Based on a cut-set localization procedure we propose an algorithm that computes the optimal localization of an SFG. The basic algorithm is then extended so that it can be used with hierarchically specified SFG's, thus significantly improving the computational efficiency. The algorithms can be easily coded and incorporated into a computer-aided-design (CAD) system.