A correlator which is capable of correlating two or more signals of unknown frequency and bandwidth receives inputs on each of two or more channels. For the case of a two-channel, or pairwise, correlator the two channels are designated a first channel and a second channel. For the pairwise correlator, a first adaptive linear predictive (ALP) filter filters the input signal from the first channel and a second ALP filter filters the input signal from the second channel, the two output signals of the filters being x(k) and y(k). The adaptive linear predictive filter is an adaptive time-domain digital filter which adapts its impulse response according to a certain algorithm. The filter output represents an adaptive linear predictive estimate of the current input value, based on many past inputs. The input signals to the two channels of the pairwise correlator may correspond to sensor outputs from two or more widely separated sensors. The two output signals x(k) and y(k), are correlated in a pairwise correlator to result in a signal which corresponds to an estimate of the magnitude-squared coherence between the signal x(k) and a time-delayed, frequency-shifted, version of the signal y(k).
Improved wavelength division multiplexed optical links and signal processing methods that employ cross tap equalization to reduce crosstalk arising from inadequate optical filtering. After optical filtering and photodetection, the wavelength division multiplexed signals adjacent either side of a signal whose crosstalk is to be reduced are sampled, weighted and are then subtracted from the signal whose crosstalk is to be reduced. This is done using adaptive correlation, performed simultaneously on all signals, each with respect to its respective adjacent channels. The present invention enables more closely spaced optical carriers to be used in the wavelength division multiplexed optical link.
A surgical lamp includes a thermoplastic reflector which reflects visible light but passes infrared radiation. The reflector is fabricated from molded polyetherimide plastic resin. A dichroic coating is vacuum deposit directly upon the front surface of the reflector.
An echo canceller in a system, having a long impulse response such as in an acoustic system, which employs a fast Fourier transform to transform input data represented in the time domain into signals represented in the frequency domain to reduce calculations. To solve the problem of a long delay, the impulse response length is divided into a plurality of blocks. Each block then has a decreased number of samples within each block. Thus, a fast Fourier transform and finite impulse response type digital filtering are effected, so that the processing delay is decreased while the amount of calculations is kept small.
An apparatus (10) for determining while drilling in the earth with a drill bit (12) the positions of seismic wave reflecting geologic formations (15, 17) has seismic wave sensors (A, B) positioned in the earth near its surface (42). The sensors (A, B) are disposed at known positions with respect to the borehole. The drill bit (12) generates the seismic waves. The sensors (A, B) receive the seismic waves both directly and by reflection from the geologic formations (15, 17). Adaptive filters (20, 26) are connected to each of the sensors (A, B). A summing circuit (34) receives outputs from the adaptive filters (20, 26). The converged impulse responses of the adaptive filters provide information about the amplitude of the reflected waves and their arrival times at the surface, thus allowing detection of the reflecting horizons and their location and physical characteristics.
A method is provided for detecting a target signal of a specific known form in the presence of clutter. The method includes dividing a set of initial training data, derived from returns from a burst of identical pulses, into a set of censored data and a set of uncensored data. A covariance matrix estimate, based on the uncensored data, is used to compute adaptive coherence estimate values, and an average adaptive coherence estimate threshold level is computed for each Doppler band to obtain a corresponding threshold. The censored data and the covariance matrix estimate are used to compute adaptive coherence estimate values for the uncensored data for each Doppler band, and these values are compared with the respective thresholds to determine the presence or absence of the target signal.
