![]() | Only 14 pages are availabe for public view |
Abstract Complex systems today are becoming more and more mixed-signal, with the analog part being the design and verification bottle neck. Traditionally, the virtual verification for the analog part of the design was only available through transistor-level SPICE simulation. Lately, behavioral modeling has been proved to be the right methodology to cope with today’s design and verification demands. Some behavioral models of complex systems consume a lot of time during simulation due to reasons such as high switching rates or many operating frequencies. An example of such circuits is the family of switched-capacitor DC-DC converters, also known as charge pumps. Our target is to study the different behavioral modeling techniques which, when deployed, enable the speed up of the simulation time of such circuits, while preserving accuracy. In this thesis we will also present an average modeling technique that concentrates on the information bearing signal of low frequency, instead of the detailed transient higher frequency signal, thereby greatly relaxing the simulation time-step, and gaining the same information in a shorter time. A real-life design is used to verify the presented methodology. |