Stream synthesis for efficient power simulation based on spectral transforms

One way of minimizing the time required to perform simulation-based power e stimation is that of reducing the length of the input trace to be fed to th e simulator. Obviously, the use of a reduced stream may introduce some erro rs in the estimation results. The generation (or synthesis) of the short in put sequence to he used for power simulation should then he carried out in such a way that the resulting error is minimized. Existing techniques explo it the knowledge of some statistical and correlation characteristics concer ning the original input trace to generate a reduced stream that closely mat ches such characteristics. In this paper, we introduce a new stream synthesis method. Its distinguishi ng feature is the use of spectral analysis based on the discrete Fourier tr ansform to determine a reduced sequence of vectors that enables us to short en the overall power simulation time at a very limited penalty in accuracy. The effectiveness and the robustness, in terms of estimation accuracy, of t he proposed synthesis procedure are demonstrated by the experimental result s we have obtained on standard combinational benchmarks for a variety of in put streams with different statistical and correlation properties, Data for sequential circuits are also reported.