Weather Radar Time Series Simulation: Rapidly Looping through Signal Parameters

Abstract Time series simulations that loop through weather signal parameters are often used to assess the performance of signal processing techniques. They can be used to test new radar-variable estimators or ground clutter filters and to produce lookup tables. If the number of parameters that need to be looped through is large, the simulations can take a significant amount of time. Speeding up these looping simulations can increase productivity by reducing the simulation runtime and also by reducing the time needed to find simulation errors. We suggest using a time series simulator based on the autocovariance matrix combined with reusing the white noise matrices that are needed for simulating time series data to provide this speedup. The autocovariance-matrix-based simulator works directly in the time domain and has several characteristics that make it especially suitable for reusing white noise matrices. Using this approach makes looping simulations up to about 16 times faster than conventional simulations and 60 times faster when also utilizing a graphics processing unit. Significance Statement Time series simulations that loop through weather signal parameters are frequently used for signal processing research. The novel techniques in this paper can speed up these simulations significantly.