EL SEGUNDO, USA: AWR Corp., the innovation leader in high-frequency EDA software, has been issued a US Patent by the United States Patent and Trademark Office for a “block-specific harmonic balance analysis system.”
Originally filed on December 3, 2008, US Patent No. 8,131,521 addresses circuit simulation using multi-rate harmonic balancing. Specifically, the invention -- known as MRHB™ and first available within AWR’s 2009 product release -- speeds the design of complex circuits by enabling the effective reduction of analysis dimensions (e.g., frequency or time).
"MRHB satisfies the needs of our customers to perform steady-state analysis of distributive or dispersive systems with more than two or three signal sources," said Taisto Tinttunen, chief director of engineering of AWR's APLAC division, who explained the limitation of previous tools and the significance of MRHB.
"Harmonic balance engines traditionally do not scale well as the number of tones increases. As a result, simulating a complete receiver or high-speed digital circuit was extremely difficult or impossible because of the high computational cost of nonlinear model evaluations and the extensive memory utilization. MRHB eliminates this limitation by defining harmonic balance analysis on a block-by-block basis. Reducing filtered sources and their harmonics by defining new hybrid tones based on linear combinations of the source tones, as MRHB does now, makes it possible to simulate designs that were previously beyond the reach of the harmonic balance technique."
AWR was also recently awarded US Patent No. 7,346,480, entitled "Impedance Mismatch Modeling in a Data Flow or Discrete Time Based System Simulation.” This invention pertains to directional connectivity described by the interconnections of blocks in a schematic (or netlist) and can be used to propagate arbitrary data from one block to another. For instance, the propagation of impedance data for discrete time-based simulation programs allows for simulation under less than ideal termination conditions between blocks, reducing simulation time and complexity and producing better results.