Abstract: An apparatus and method that accurately and reliably detects ice on a target surface (20). Intrinsic birefringent properties of hexagonal ice crystals are used to detect ice by analyzing elliptically-polarized reflection of linearly-polarized radiation from ice. A linearly-polarized radiation beam (15) is focused on the target surface (20), and returned radiation (25) is filtered (30) as a function of its elliptical polarization. Filtered radiation (35) is directed to a sensor (40) and quantified according to the intensity of radiation received at various points (41.sub.1 -41.sub.n) on the filter (30), and an output signal for each point is produced (45.sub.1 -45.sub.n). The presence or formation of ice is determined by a signal processor (50) that detects the variance among the output signals (45.sub.1 -45.sub.n)--a variance (55) reaching a threshold value being indicative of ice.

Abstract: An aperture coupled microwave antenna (10) for processing circularly polarized signals. The antenna (10) comprises a first planar dielectric layer (22) upon which a conductive radiating patch (12) is mounted. Attached to the radiating patch (12) are tuning means (20, 72) for converting linearly polarized signals into circularly polarized signals. The tuning means preferably takes the form of conductive tuning stubs (20). Abutting an opposite face (23) of the first dielectric layer (22) is a conductive ground plane (24) having two orthogonal elongated apertures (26, 28). The radiating patch (12) is electromagnetically coupled, through the two elongated apertures (26, 28), to two input/output ports (48, 56) by two conductive feeding circuits (38, 40). Each of the feeding circuits (38, 40) interacts with only one of the elongated apertures (26, 28, respectively). The two feeding circuits (38, 40) and the elongated apertures (26, 28) are designed to operate in isolation.

Abstract: Apparatus and methods for reducing the quantization error in shift-matrix finite impulse response (FIR) filters (210) operate on a sampled analog signal (11) converted to a sequence of digital signal samples (13). A weight conversion device (23) converts a weight signal W comprising a sequence of tap weights W.sub.i into both a corresponding series of logarithmically-encoded weights W'.sub.i and a series of weight error values .DELTA.W.sub.i. A multiplication device (30) multiplies each of the digital signal samples (13) with the first series of logarithmically-encoded weights W'.sub.i and provides the results to an accumulator (26). A signal encoding device (40) converts the digital signal samples (13) into a sequence of logarithmically-encoded digital signal samples (42). A quantization error correction device (220) multiplies each of the weight error values .DELTA.W.sub.i by the logarithmically-encoded digital signal samples (42) and also provides the results simultaneously to the accumulator (26).

Abstract: An improved method for transitioning a spacecraft from a thruster (5) controlled, station keeping mode to an operational, on-orbit mode, in which attitude control is maintained using internal momentum wheels (3). The method first utilizes a thruster compensation technique to supplement the conventional transition mode control system. This compensation uses empirical thruster data to derate thruster (5) efficiency for extremely short pulse durations necessarily used in transition mode operation. Secondly, a three step deadbeat thruster (5) sequence is employed in which a first pulse linearizes spacecraft (1) motion prior to orienting and stopping nutation. This linearization provides improved accuracy in computing and shortens the waiting time for subsequent thruster (5) firings.

Abstract: A circuit bypass device (30) having terminals (70, 80) is connected to contacts (70, 80) of a cell (20) to provide a conductive path (40) in response to voltage excursion therebetween. Device (30) includes: a conductive shunt (40), coupled to contacts (70, 80), that is movable from a normally non-conductive position to a conductive position between terminals (70, 80); a detector (50), coupled to conductive shunt (40), for sensing voltage excursion beyond specified limits between contacts (70, 80); and an actuator (60) having a heat-to-recover shape memory metal alloy that is formed to contract when detector (50) senses the voltage excursion, for moving conductive shunt (40) from the non-conductive position to the conductive position. The formed alloy is preferably spring-shape which contracts translationally and torsionally.

Abstract: A conducting plane resonator (10) is used to stabilize an oscillating means (12) typically operating at microwave frequencies. The conducting plane resonator (10) is in proximity to the oscillating means (12) and is magnetically coupled thereto. In a first implementation, the conducting plane resonator (10) is magnetically coupled to an input conductor (16) of the oscillating means (12). In a second implementation, the conducting plane resonator (10) is magnetically coupled to an output conductor (18), which is coupled to the oscillating means (12). In a third implementation, the conducting plane resonator (10) is magnetically coupled to both the input conductor (16) and the output conductor (18). The conducting plane resonator (10) preferably comprises a thin, substantially planar conducting plane and is dimensioned such that it is resonant at the desired operating frequency. The conducting plane resonator (10) is preferably fabricated of a superconductor material to achieve high-Q performance.

Abstract: A digital computer architecture specifically tailored for implementing a neural network. Several simultaneously operable processors (10) each have their own local memory (17) for storing weight and connectivity information corresponding to nodes of the neural network whose output values will be calculated by said processor (10). A global memory (55,56) is coupled to each of the processors (10) via a common data bus (30). Output values corresponding to a first layer of the neural network are broadcast from the global memory (55,56) into each of the processors (10). The processors (10) calculate output values for a set of nodes of the next higher-ordered layer of the neural network. Said newly-calculated output values are broadcast from each processor (10) to the global memory (55,56) and to all the other processors (10), which use the output values as a head start in calculating a new set of output values corresponding to the next layer of the neural network.

Abstract: An electronic detection system for detecting vibration patterns from physical events, such as engine knocking or articulated speech, senses a vibration source with a wideband transducer to provide a sensed signal, then converts this signal to energy amplitude and spectral frequency form to extract detectable features of the vibration pattern. Events are detected when a controller determines that the sensed energy and spectral data exceed adaptively-predetermined energy and spectral thresholds, preferably for a duration exceeding predefined time windows.

Abstract: A waveform generator for a resolver transducer or the like is described. Such generator forms two 3-level waveforms which approximate sinusoidal waveforms for the excitation and/or output detection of such a resolver. Each of the sinusoidal waveforms is formed by combining a pair of digitally programmed bi-state signals which have their phases shifted relative to one another by an amount proportional to an angular input variable. The result is that the amplitude of these sinusoidal waveforms which are formed are representative of the sine and cosine of such variable.

Abstract: An equalizer (10) for removing parabolic phase distortion from an analog signal (3), utilizing a pair of series connected transversal filters. The parabolic phase distortion is cancelled by generating an inverse parabolic approximation using a sinusoidal phase control filter (18). The signal (3) is then passed through an amplitude control filter (21) to remove magnitude ripple components.

Abstract: An electronic data filter system (10), which functions as an approximation of a perfect matched filter, receives a conventional square-wave digital input signal (12) and outputs a filtered output signal (18). Output signal (18) appears as the summed integral of input signal (12). Filter system (10) includes a rate detector (50), which receives input signal (12) and provides a tuning voltage (16), and a data filter (30) which provides output signal (18). Data filter (30) includes, arranged in series, a notch filter (80) and a low-pass filter (90), wherein such filters (80, 90) contain capacitive-inductive circuits having certain value ratios which are tunable responsive to tuning voltage (16). Signal degradation of the filtered signal (12, 18) is minimized by filter system (10), which reduces the bit error rate (BER) and noise bandwidth of the filtered signal (12, 18). In one embodiment, data filters (30) are cascaded in parallel to extend data filtering bandwidth.

Abstract: A system and method for controlling a multi-point matrix switch (13) to route signals in a communication system (11). The invention identifies source and destination ports (129, 133) in the communication system (11) and then polls identified ports (129, 133) to recognize those that are available for communication. Identified and recognized ports (129, 133) are displayed graphically, so that at least one path (139) between recognized source ports (129, 133) and associated recognized destination ports (129, 133) may be selected graphically by user (159) input. Selected path (139) information is then submitted to switch (13) to cause switch (13) to route signals through the selected paths (139) by connecting recognized source ports (129, 133) to associated recognized destination ports (129, 133).

Abstract: An electronic activation circuit (1) for automatically transitioning a guided missile from a non-tracking mode to a tracking mode. When the missile's tracking means comprises a gyroscope, the non-tracking mode corresponds to the gyroscope being caged. A parameter indicating target signal quality is extracted (2) from a missile carrier signal (3) generated by means, which may be on board the missile or associated with the device launching the missile, for measuring target signal quality. This signal (3) is compared (6) against a preselected reference threshold voltage (4) to determine whether the tracking mode is to be initiated. Temporary indications to recage during a preselected time delay latch period T are suppressed (8, 10). The quality threshold comparison (6) is hysteresis conditioned (46, 48, 50, 52) to attenuate the effects of noise on the input signal (3).

Abstract: A telescoped ammunition round comprises: a propellant charge having an axial cavity for supplying firing power for the ammunition round; a projectile housed within the axial cavity for being fired from a forward end of the ammunition round; a control tube means housed within the axial cavity for selectively covering an aft surface portion of the propellant charge axial cavity, the control tube means having a generally cylindrical axial bore substantially coaxial with the axial cavity, the projectile extending forward of the control tube means; coupling means for releasably securing the projectile to the control tube means; a control tube pistion having a generally cylindrical outer shape forming a sliding fit within the axial bore, positioned aft of the projectile; a booster charge within the axial bore, the control tube piston being forwardly movable in response to the firing of the booster charge and the projectile, in turn, being forwardly movable by forward movement of the control tube piston; a primer me

Abstract: A method for making a battery electrode includes roughening the surface of a substrate (10) that constitutes a precursor to the electrode, using an electrolytic solution (12) with electrical potential perturbations applied thereto. The substrate (10) of porous sintered nickel powder is first formed. The electrolytic solution (12) prefereably contains the pure metal that forms the electrode. Then all gases in and around the substrate (10) are preferably removed. Next, the substrate (10) is placed in the solution (12) for a predetermined amount of time. Potential perturbations are then applied to the substrate (10) and the solution (12). The potential peturbations vary between the voltages necessary for electrodissolution and electrodepositon of the substrate (10), and thus, cause the surface of the substrate (10) to be roughened as portions of the substrate (10) are dissolved into the solution (12) and then redeposited onto the substrate's (10) surface.

Abstract: Planar dual mode filters (30) are formed by a conductive resonator (20) having circular symmetry and two pairs of symmetrically oriented planar conductive leads (22, 26 and 24, 28). The conductive leads (22, 26 and 24, 28) are aligned colinearly with two orthogonal diameters (32, 34, respectively) of the circular conductive resonator (20) and are electrically isolated from said resonator (20). A perturbation (38) located on an axis (36) oriented symmetrically with respect to the two pairs of conductive lead (22, 26 and 24, 28) couples electromagnetic modes which are injected into the resonator (20) by the planar conductive leads (22, 26 and 24, 28). Higher order filter circuits can be realized by combining multiple filters (30) of the present invention. The filters (30) are amenable to printed circuit (microstrip to stripline) fabrication using superconductors for the conductive elements.

Abstract: An adaptive RADAR environmental signal filter (1) is described which selectively preprocesses RF RADAR input data prior to the data's utilization by a processor (8). The filter (1) consists of an array of parallel discriminator cells (165) which compare incoming RF RADAR data to a window parameterized by predetermined threshold values. The array of discriminator outputs are networked through combinational logic (59) in order to provide an output correlation signal (63) to an external processor (8). A discriminator enable latch (55) synchronizes the discriminator outputs as the outputs pass through the combinational logic (59).

Abstract: A remote waveguide flange clamp system (10, 11) detachably connects a first waveguide section (20) to a second waveguide section (30, 37). System (10, 11) is controlled remotely by a controller (60). A pair of elongated clamping levers (40) is mounted pivotally to pins (42) on bracket arms (39) of section (30, 37). Levers (40) rotate about pins (42) when a motor assembly (51) moves driving ends (47) of levers (40) outwardly or inwardly relative to section (30, 37). When ends (47) move outwardly, clamping ends (44) of levers (40) apply a compressive force to flanges (22, 32) of sections (20, 30) to provide a gasketed seal (80) inhibiting RF leakage therebetween. Assembly (51) includes a torque sensor (66) for continuously applying current to a motor (50) in assembly (51) until sensor (66) measures current substantially equaling a predefined value.

Abstract: A nine bit Gray code counter is constructed as a single integrated circuit such as a PLA and comprises D flip-flops, AND gates and Exclusive-OR gates. The array is programmed to provide a nine bit Gray code count at its outputs. The Q output of each flip-flop provides one bit in the Gray code count. The D input to each flip-flop is determined by the significance of the bit and the following equations: D0=A XOR Q0; D1=(A.multidot.Q0) XOR Q1; Dn=(A.multidot. . . . .multidot.Qn-2.multidot.Qn--1) XOR Qn; and Dmsb=(A.multidot.Q0.multidot.Q1.multidot. . . . .multidot.Qmsb-2 XOR Qmsb.