A physically compact, wideband signal activity identification, demodulation and characterization, and direction finding device incorporates multiple and cascadable digital signal processing modules operating in asynchronous real time. The digital signal processing modules include a device for data buffering, a device for digital signal processing (DSP), and a device for high-speed data routing. The module device for data buffering is composed of, among other memory devices, a First In Tap Out (FITO) data buffer that may be accessed at any point for delay or faster than real time resynchronization by the DSP module device. The module device for digital signal processing function incorporates a general purpose digital processor for respective module calculation of overlapped hyperchannelization Fast Fourier Transforms (FFT). Hyperchannels may be combine in a flexible manner to tailor channel bandwidth for optimum signal spectral detection of signal activity, synthesis filter and tuning, demodulation and recognition, and direction finding. The module device for high-speed data routing is composed of one-to-one, one-to-many, or many-to-one digital data routing functions, and allows flexible ordering of digital signal processing modules.

Wideband communication intercept and direction finding device using hyperchannelization

A Fast Fourier Transformation (FFT) processing is disclosed that measures FFT output phase standard deviation over a number of consecutive FFT runs. The system corrects FFT output phase in an organized fashion for all potential signal filter offset positions while measuring changes in phase standard deviation, and selects the filter offset where the minimum standard deviation occurs. The system utilizes pseudo coherent integration to enhance traditional integration, where the pseudo coherent integration locates the mean phase shift within the number of FFTs integrated, and corrects all FFT runs by this mean shift value. The integration multiplies the magnitude of each FFT filter output by the cosine and sums all FFTs in the integration period for the respective filter. Detections are then declared based on outputs from a combination of a single filter detector, a sum coherent and traditional detector, a filtered coherent detector, and a filtered traditional detector, thereby utilizing various blended techniques across the filter bandwidth. Pulse Doppler radar architecture is predicated on more efficient time utilization during target dwells and is achieved by using managed PRF scheduling, frequency domain Doppler ambiguity resolution, determining signal position and phase within a Doppler filter extremely accurately, performing pseudo coherent integration and use of multi-strategy detection concepts. The PRF scheduling used in this radar architecture coupled with a frequency domain approach to perform Doppler corrected pulse compression and resolve Doppler ambiguity on the same scan provides the architecture to maintain same PRF during dwell and capitalize on integration.

Signal processing architecture which improves sonar and pulse doppler radar performance and tracking capability

A vehicle navigation method and system employing echo or doppler analysis to provide autonomous or enhanced navigational capabilities by correlating stored scene information with echo analysis information derived from an Active Traveling-Wave Device (ATWD) output representing information concerning the vehicle's state and velocity vectors with respect to a mapped scene.

High resolution autonomous precision positioning system

The present invention discloses a system for extracting targets from radar signatures. Disclosed is a unique combination of Wavelet technology and neural networks using fractal geometry techniques that estimates the Fractal Dimension of a target even in the presence of high background noise.

System for extracting targets from radar signatures

An aircraft including an approach and landing system, including a navigation unit for providing navigation information, a weather radar unit for providing radar information, a processor which receives navigation information from the navigation unit and information from the weather radar unit, the processor unit providing an output representing information concerning the aircraft in accordance with the provided navigation information and radar information, a memory for storing information representing a scene, the processor unit correlating the stored scene information with the output representing information concerning the aircraft to provide a mapped scene, a display unit for displaying the output of said processor and the mapped scene, and a steppable frequency oscillator for providing a signal which is stepped in frequency to the weather radar unit, thereby providing an increased range resolution.

High accuracy, high integrity scene mapped navigation

A coherent radar is used for achieving high resolution radar imaging of a moving object in sea clutter and noise. A stored replica of the transmitted waveform is combined with the returned signal in a synchronous detector to produce in-phase and quadrature (I and Q or complex) samples. High resolution is achieved by transmitting a series of pulses, each at a different frequency, and then processing these complex samples to produce a synthetic down-range profile. To enhance the radar target or object from system noise and sea clutter, the complex I and Q (in-phase and quadrature component) samples from the radar are coherently averaged in a special two-dimensional slice of the trispectrum (e.g. quadruple product or fourth-order moment). This formulation of the fourth moment retains all important target information and suppresses Gaussian noise. Once averaged, a new set of I and Q samples are reconstructed that produce the same trispectral slice as this average. These reconstructed samples are then transformed using conventional Fourier transform methods to produce the average, or enhanced, down-range profile.

High resolution radar profiling using higher-order statistics

A signal detection system divides a data sampling run into blocks and perms a fast Fourier transform on each block, sorting results by frequency. Combinations of results of the fast Fourier transform corresponding to each frequency are processed to derive a test statistic which is unbiased by Gaussian noise while including such combinations of results of the fast Fourier transform which would be redundant over other combinations. Information concerning the frequency behavior of the signal derived in the course of detection, is accomplished with increased sensitivity.

Coherent signal power detector using higher-order statistics

A signal detection system, preferably for use with a coherent radar systemelects certain combination of input signal samples in a block of signal samples to derive a test statistic which is unbiased by Gaussian noise. Products of pairs of sample values are stored in a data sample look-up table. An index look-up table is created by scanning through possible combinations of addresses and excluding those combinations of sample values which would be redundant over other combination and would result in a contribution to biasing by noise. Pairs of addresses from the index look-up table are then used in sequence to access first and second products from the data sample look-up table. The first product is multiplied by the complex conjugate of the second product and the (quadruple product) result is averaged to form a test statistic which is insensitive to position and constant velocity. Detection of the signal is determined by testing the real part of the averaged test statistic against a threshold. To allow such detection for an accelerating target, a lag index is derived for each combination of addresses in the index look-up table and the accumulation and averaging is done while sorting in accordance with the lag index. Dynamic adjustment of the test statistic compensates for time-varying clutter such as ocean waves, and allow the signal detector to track a substantially constant false alarm rate.

Detection of radar targets using higher-order statistics

A complex signal correlator such as can be implemented in a correlation detector system affords a unique algorithm which estimates the codifference correlation between two complex signals based on the sum and difference of codifference estimates, each codifference estimate having equivalently associated therewith a dispersion estimate. Typical embodiments provide a receiving antenna and a receiver inclusive of the codifference correlator, wherein radio frequency waves are down converted and sampled, the sampled signals are correlated with a reference signal contained in a memory, and the resultant correlation signal is detected and transduced. The inventive correlator is based on an alpha-stable distribution and, in comparison with conventional alpha-stable distribution-based correlators, can more effectively operate in a realm wherein alpha is less than one.

Codifference correlator for impulsive signals and noise

While the computation of high resolution radar images based on higher ordertatistics is position insensitive, velocity estimation may be based on the ratios of values of a trispectral slice and a cross-trispectral slice computed as quadruple products of complex valued signals developed by coherent radar. Signal-to-noise ratio is improved by either averaging over a plurality of bursts during computation of both the trispectral slice and the cross-trispectral slice or averaging of values of ratios of trispectral slice and cross-trispectral slice values at particular frequencies or wavenumbers, or both.

Robert D. Pierce

Papers

High resolution radar profiling using higher-order statistics

Coherent signal power detector using higher-order statistics

Detection of radar targets using higher-order statistics

Codifference correlator for impulsive signals and noise

Moving target indicator using higher order statistics