A prototype multi-channel laser Doppler anemometry (LDA) processor was assembled using a wideband transient recorder and a microcomputer with an array processor for fast Fourier transform (FFT) computations. The prototype instrument was used to acquire, process, and record signals from a three-component wind tunnel LDA system subject to various conditions of noise and flow turbulence. The recorded data was used to evaluate the effectiveness of burst acceptance criteria, processing algorithms, and selection of processing parameters such as record length. The recorded signals were also used to obtain comparative estimates of signal-to-noise ratio between time-domain and frequency-domain signal detection schemes. These comparisons show that the FFT processing scheme allows accurate processing of signals for which the signal-to-noise ratio is 10 to 15 dB less than is practical using counter processors.
