An adaptive digital filter

In recent years, catastrophic damage and loss of off-shore installations due to hurricane generated waves in the Gulf of Mexico have repeatedly pointed out the need for better design criteria for offshore structures. Information concerning energy concentrations in hurricane waves and resonant modes of Gulf platforms is of primary importance in the design of offshore structures. The purpose of the research reported here was to develop an adaptive digital filter which would allow analysis of ocean wave and platform response records in terms of resonant components. Digital signal processing techniques have become widely used in many different areas including seismic work, speech processing, medical research and ocean wave analysis. Common techniques to process these data include the fast Fourier transform (FFT) and the autocorrelation and crosscorrelation functions. More recently, recursive digital filtering has become important in many of these areas. Recursive filters may be designed from present analog filter theory by using the s-domain to z-domain transformation and then converting the filter function in z to a difference equation on a digital computer. The recursive filter difference equation is a function of a limited number of present and past inputs as well as previous outputs. The characteristics of a recursive filter allow efficient programming on a general purpose digital computer or realization of the filter through special purpose digital hardware. Adaptive systems have also become increasingly important in recent years. Much of the work done in this area has been of a developmental nature. The results of some of this work become directly applicable to the design of an adaptive digital filter. The digital filter characteristics may be easily changed to allow the filter to adapt to resonant components in a signal waveform. A method of changing these constants to implement the adaptive process has been developed. A Fortran computer program has been developed to implement the adaptive digital filter on the Sigma 7 computer. The computer program results show that an adaptive filter configuration is feasible and that the technique provides a useful signal processing algorithm.