Abstract: Techniques for precise removal of offset charge associated with the reset switch of an integration circuit are disclosed. Offset cancellation circuitry includes a single reset offset subtraction circuit and a replica integrator, which is configured identically to the integrators to be offset cancelled. An offset charge is generated by the circuitry and capacitively coupled to the target integrators. This generated offset charge causes voltage at the input node of each target integrator to substantially match the desired starting voltage level of the targeted integration process. Minimal additional space and circuitry is needed. All of the undesired offset charge is cancelled, without canceling any of the desired input current.

Abstract: Techniques that eliminate or otherwise reduce retroreflections associated with focal plane arrays (cooled and uncooled) and other tilted devices are disclosed. The device topology is designed so that incoming radiation is reflected outside of the incoming focusing optical cone. Planarization is used to eliminate concave geometries and angled sidewalls can be used to eliminate convex geometries, as such geometries cause retroreflection. The resultant structure is covered with reflective material (e.g., aluminum, titanium tungsten, or other standard metallization), which not only safely reflects light from the scene, but also prevents reflections from the underlying geometries of the via. Devices configured in accordance with the principles of the present invention require less tilt, thereby minimizing optical distortion.

Abstract: The present invention relates to an optical modulator array that uses stepped-well continuously tunable quantum well infrared modulators in order to accomplish electronic beam modulating. The present invention involves a coherent optical beam modulating device to steer an optical beam comprising: an optical modulator array, where said optical modulator array includes a stepped quantum well doped with electrons, wherein the modulator array affects operates as at least one of a phase modulator and a light intensity modulator base upon a voltage bias applied across the modulator array. The continuous tunable quantum well modulator includes asymmetry of the unit cell that allows transitions from the ground state to the second excited state that are normally forbidden in symmetrical quantum well infrared photodetectors.

Abstract: A reduced complexity Turbo multi-user detector (MUD) processing system in multiple access communications channels that decreases the likelihood of improper decoding of the final values of interest and decreases the computation complexity for each iteration, thereby allowing for a reduction in the number of iterations performed and lowers the overall complexity without negatively impacting performance. In one form the present invention comprises a multi-user detector coupled to two or more decoder sections, two ore more recoders, and a compare and adjust section in such a manner that data flows iteratively to correct for errors in a computationally efficient manner.

Abstract: A collapsible discone antenna is provided with an ultra wide band width by providing a collapsible conical skeleton cone, with the rods of the skeleton being provided with meander lines so as to effectively reduce the overall dimensions of the antenna by a factor of 2, with the antenna rods being electrically interconnected at their distal ends so as to eliminate performance degradation due to varying ground conductivities. A specialized feed configuration is used in one embodiment to feed multiple antennas stacked above a low band disc through the utilization of one or more coaxial lines which are wrapped around a ferrite toroid so that they may be passed up through the low-band disc without detuning the low band discone antenna. The use of the toroid inductor between the low-band cone and the low-band disc further reduces the low frequency cutoff of the antenna by markedly decreasing the VSWR at frequencies as low as 20 megahertz.

Abstract: By substituting a ring optical parametric oscillator for the linear cavity, one eliminates the effects of feedback solved in the prior art by the use of isolators and by the purposeful off-axis alignment of the pump laser beam. The result is an exceedingly robust system with state-of-the-art efficiency for maximizing output power while preserving beam quality.

Abstract: A read/write circuit for accessing chalcogenide non-volatile memory cells is disclosed. The read/write circuit includes a chalcogenide storage element, a voltage limiting circuit, a current-to-voltage converter, and a buffer circuit. The voltage limiting circuit, which is coupled to the chalcogenide storage element, ensures that voltages across the chalcogenide storage element will not exceed a predetermined value during a read operation. During a read operation, the current-to-voltage converter, which is coupled to the voltage limiting circuit, converts a current pulse read from the chalcogenide storage element to a voltage pulse. By sensing the voltage pulse from the current-to-voltage converter, the buffer circuit can determine a storage state of the chalcogenide storage element.

Abstract: A group of pixels obtained using an imaging device is converted into a group of pixels having an equal or lower count by receiving at least first and second input pixels having an initial intensity value, forming at least one intermediate intensity value from each of the first and second input pixels, and combining the intermediate intensity values formed from the first and second input pixels to form at least one output pixel. This allows the pixels formed to have improved signal-to-noise characteristics, and it reduces transmission rates. A feature of the present invention allows a distance between the imaging device and an object to be used to select the number of output pixels formed. Another feature of the present invention allows a change in distance between the imaging device and an object to be used to dynamically adjust the number of output pixels formed. A further feature of the invention allows a variable resolution image to be converted to a fixed resolution image.

Abstract: A method and system for training an audio analyzer (114) to identify asynchronous segments of audio types using sample data sets, the sample data sets being representative of audio signals for which segmentation is desired. The system and method then label asynchronous segments of audio samples, collected at the target site, into a plurality of categories by cascading hidden Markov models (HMM). The cascaded HMMs consist of 2 stages, the output of the first stage HMM (208) being transformed and used as observation inputs to the second stage HMM (212). This cascaded HMM approach allows for modeling processes with complex temporal characteristics by using training data. It also contains a flexible framework that allows for segments of varying duration. The system and method are particularly useful in identifying and separating segments of the human voice for voice recognition systems from other audio such as music.

Abstract: An imaging system comprising a lens, a detector array (e.g., focal plane array), a signal processing module and a shutter, wherein the shutter is positioned in front of the lens (between the lens and the scene being imaged). This front lens shutter mount configuration allows offset correction to compensate for internal radiant flux and other deficiencies associated with conventional systems.

Abstract: A honeycomb-like shell is provided for an acoustic projector used in sonar applications to increase the bandwidth of the projector, reduce its weight, and to provide increased power density and reduced shell costs. The shell has outer and inner layers, with the honeycomb structure therebetween. The honeycomb shells have application in slotted cylindrical projectors, flextensional projectors, inverse flex tensional projectors, and oval-shaped projectors in which the honeycomb structure replaces the solid shells, with the honeycomb providing the relatively high specific stiffness required for the acoustic properties of the projector. The honeycomb shell achieves the same bending stiffness of the solid shells with less weight through the utilization of radial stiffeners between the inner and outer layers.

Abstract: A method and system begin at a random node of a trellis model having a set of at least two axes, wherein one axis corresponds to time and a second axis corresponds to frequency, a set of M states, corresponding to a set of all possible frequencies that may be transmitted by the system, and branches leaving each state, that terminate at allowable transmit frequencies for a next frequency hop. The number of branches leaving each state is dependent on the number of encoded bits per hop. Next, B bits of data are passed to a buffer, where B is the number of bits that will be transmitted per frequency hop. The method and system then combine the B bits of data with information for previously selected frequencies to select a current frequency, transmit the selected frequency, and feedback the currently selected frequency information to be used as previously selected frequency information when selecting a next frequency.

Abstract: A circuit for accessing a chalcogenide memory array is disclosed. The chalcogenide memory array includes multiple subarrays with rows and columns formed by chalcogenide storage elements. The chalcogenide memory array is accessed by discrete read and write circuits. Associated with a respective one of the subarrays, each of the write circuits includes an independent write 0 circuit and an independent write 1 circuit. Also associated with a respective one of the subarrays, each of the read circuits includes a sense amplifier circuit. In addition, a voltage level control module is coupled to the read and write circuits to ensure that voltages across the chalcogenide storage elements within the chalcogenide memory array do not exceed a predetermined value during read and write operations.

Abstract: A multi-user turbo decoder combining multi-user detection and forward error correction decoding is disclosed that utilizes iterative decoding of received, interfering signals, and the construction of a decoding tree of the decoder is changed for each iteration of the decoding based on the previous conditional probability estimates of the value of the data bits of each signal making up the received, interfering signals. Before each iteration of multi-user decoding, a probability estimate is calculated that the value of the bit in a signal has a certain value for all of the data bits. Using the probability estimate a new decoding tree is constructed before each iteration of decoding such that the signal bit having the most reliable estimate is assigned to the lowest or root level of the tree.

Abstract: In a multi-user detection system in which interfering signals are purposely allowed to exist, a parameter estimation unit is provided which utilizes signal processing for determining the channel transfer function for each received signal, including the received power, phase of the oscillator, timing offset relative to the base station clock carrier frequency offset, and number of multipaths replicas and delays for each replica, with the system providing realtime uninterrupted estimates of these parameters required by the signal separation unit.

Abstract: A method for designing a cell-based ASIC device with multiple power supply voltages is disclosed. An ASIC chip image is made without applying power or ground buses to metal layer M1. All fast or high-power circuits are grouped together into high-power logic blocks and synthesized with high-power circuit macro libraries. All slow or low-power circuits are grouped together into low-power logic blocks and synthesized with low power circuit macro libraries. The associate power and ground buses are applied to metal layer M1 in each of the logic blocks. The logic blocks are placed on the ASIC so that different voltage groups are separated by at least one cell. The ASIC is then routed and tested before the mask is released.

Abstract: A circuit for accessing a chalcogenide memory array is disclosed. The chalcogenide memory array includes multiple subarrays with rows and columns formed by chalcogenide storage elements. The chalcogenide memory array is accessed by discrete read and write circuits. Associated with a respective one of the subarrays, each of the write circuits includes an independent write 0 circuit and an independent write 1 circuit. Also associated with a respective one of the subarrays, each of the read circuits includes a sense amplifier circuit. In addition, a voltage level control module is coupled to the read and write circuits to ensure that voltages across the chalcogenide storage elements within the chalcogenide memory array do not exceed a predetermined value during read and write operations.

Abstract: An apparatus for inspecting semiconductor module packages which includes a cylindrical base section, a truncated spherical section superimposed on the cylindrical base capable of being rotated and tilted on the cylindrical base section, and a tray section superimposed on the truncated section.

Abstract: A system is provided for rapidly ascertaining the position of a pulse train emitter such as a radar using multiple collectors without requiring more than one platform to measure the same pulse. Thus time-of-arrival measurements at a number of collecting platforms are performed, with the positions of the platforms being accurately ascertainable using GPS data, and with time synchronization between the spaced-apart collectors performed by utilizing atomic clocks. In the multi-ship case, geolocation can be performed on ten milliseconds of data as opposed to 30 seconds of data for measurements involving a single platform. The subject system is preferable to conventional time-difference-of-arrival geolocation systems because those systems require that each of the collecting platforms measure the same pulse from the emitter, which severely constrains the flight paths of the collectors, limits the amount of usable data, and increases the system's sensitivity requirements.

Abstract: A method of fabricating a dense pixel array comprising the steps of: (a) printing a photoresist mask and applying said mask to a semiconductor material substrate to form a masked area and an unmasked area on said substrate; (b) applying a photoresist material layer to the unmasked area of the substrate, then applying a metal layer over the photoresist material layer and the substrate, and then applying a solvent to remove the photoresist material layer and said metal layer applied over said photoresist material layer to leave a plurality of metal layers superimposed over the unmasked area of the substrate; (c) removing the substrate to a depressed substrate surface between the metal layers formed in step (b) to form a plurality of pixels each having an upper metal layer; (d) superimposing an insulative layer over each of the metal layers formed in step (c); (e) forming a hole in at least one of the insulative layers formed in step (d) so as to expose the metal layer under the insulative layer; and (f) superim