Abstract: A variable field effect capacitive device suitable for providing different amounts of capacitance in response to control signals of different magnitudes. The device includes a pair of plate electrodes and a pair of control electrodes. A semiconductor region is located between the control electrodes. The plates each make Schottky contact to the semiconductor region to form a depletion region therein which changes shape in response to changes in the magnitude of the control signals.

Abstract: An improved method and structure for high voltage semiconductor devices capable of blocking voltages of the order of 1000 volts and greater is described. In a preferred embodiment, a blanket P layer is formed in an N.sup.- epi-layer on an N.sup.+ substrate. An annular groove is etched through the blanket P layer into the N.sup.- epi-layer. The bottom of the groove is doped N.sup.+ using the same mask as for the first groove etch. A second groove is formed inside of and partly overlapping the first groove and extending to a greater depth than the first groove, but not through the epi-layer. The second groove is filled with passivating material, metal electrodes are applied to the P.sup.+ region and the N.sup.+ substrate, and the devices separated at the N.sup.+ region lying outside the second groove in the bottom of the first groove. Excellent high voltage blocking characteristics are obtained with the same or fewer process steps and better yield.

Abstract: The performance of microwave coplanar and other circuit boards is improved by providing local matching elements for adjusting the characteristic impedance of local transmission line segments to better match the input or output impedance of the particular devices to which they couple. This is accomplished by providing a conductive bridge over the active lead of the transmission line, thereby lowering its characteristic impedance. This is very useful for narrow transmission line segments needed to make contact to small active device and/or MMIC's and which otherwise exhibit a substantial impedance mismatch with such small elements.

Abstract: An microwave analog frequency divider capable of ten percent bandwidth at 10 GHz has push-push amplifiers forming an injection locked oscillator. The input frequency f.sub.i is injected at the amplifier reference terminals. The amplifier control terminals are coupled by L's and the reference terminals are coupled by C's and the reference and control terminals form with the L's and C's a resonant circuit tuned approximately to the output frequency f.sub.o =f.sub.i /N, where N is the divisor. Parallel resonant circuits in the f.sub.i injection path have high impedance at f.sub.o to prevent loading of the oscillator. A control device in the f.sub.i injection path turns the oscillator on when f.sub.i is present and off when f.sub.i is absent. Matching networks are included at input, output and between the control device and the oscillator to obtain efficient operation.

Abstract: An N.times.M electrical connector suitable for use at microwave frequencies comprises a first set of N insulating parallel substrates each containing M notches with adjacent planar conductors and a second set of M parallel substrates having up to N spaced-apart conductors, wherein the notch-to-notch and conductor-to-conductor spacing of the first set matches the substrate-to-substrate spacing of the second set and vice versa, so that when the second set is pushed into the notches of the first set, each notch and associated conductor on the first set aligns with related conductors on the second set of "L" shaped relation. One or more spring contacts at each intersection provide positive electrical connection between the proximate conductors. The arrangement is especially useful for microwave systems since multiple signal lines and ground planes may be connected simultaneously without interference or substantial electrical discontinuity.

Abstract: A new way of determining correlation coefficients for stochastic codebook vectors for CELP coding of speech takes advantage of the sparsely populated nature of stochastic codebook vectors. N valued input signals (e.g., convolution vectors) to be correlated with N valued codebook vectors are fed to an N by N multiplexer or other selection means and the signal values either passed to an accumulator or not according to the state of N select inputs or other identification means determined from a memory store (e.g., an EPROM) whose entries correspond to the non-zero values of the codebook vectors. The accumulator output is the correlation of the codebook vector with the input signal. A sequencer steps through the entire codebook to provide correlation values for each vectors. The results are used to determine the optimum stochastic codebook vector for replicating the particular speech frame being analyzed.

Abstract: A new way of determining autocorrelation coefficients for adaptive codebook vectors for CELP coding of speech simplifies and improves the accuracy of the autocorrelation coefficient determination for the situation where the codebook vector length being analyzed is less than a speech frame length. This is important in synthesizing short pitch period speech. Copy-up of the shortened codebook vector to equal the frame length is not needed and autocorrelation coefficient errors associated with copy-up are avoided. The improved system relies on calculating autocorrelation coefficients of the first (shortest) vector and then obtaining subsequent autocorrelation coefficients for successive vectors of increasing length by a simple end correction technique until the vector length equals the frame length. The autocorrelation coefficients are scaled by multiplying them by the ratio of the frame length to the vector length.

Abstract: An apparatus for secure communications contains a controller for automatically selecting one of several data ciphering means utilizing one of a plurality of ciphering algorithms common to transmitting and receiving terminals. A transmitting means for transmitting encrypted data, and a receiving means for receiving encrypted data are coupled to the plurality of ciphering means. The control means automatically determines which of the ciphering means to employ for any given secure communication. The method for establishing a secure communications link includes the steps of exchanging a first message for determining a common key generation and ciphering method and comparing a further shared message for validation of communications terminal security.

Abstract: A logarithmic amplifier includes amplifier stages having an input transmission line and first and second output transmission lines. The input and output transmission line are coupled by multiple amplifier elements distributed along the transmission lines. One output transmission line forms a high gain low compression path and the other a low gain high compression path. The output transmission lines of each stage are coupled to a combiner from whence the logarithmically amplified output signal is obtained. The logarithmic amplifier stages are readily constructed in MMIC form and multiple stages may be easily cascaded to provide a very large dynamic range.

Abstract: A new way of CELP coding speech simplifies the recursive loop used to poll code adaptive book vectors by reducing the number of autocorrelation operations that must be performed with the K vectors of the codebook each having N entries. Autocorrelation is initially performed for only a small number P<<N autocorrelation coefficients in each codebook vector and the values found are used to scan through all the codebook vectors looking for those S vectors (S<K) which give the best match to the input speech. The autocorrelation function for the S vectors is then recalculated for R entries (P<R.ltoreq.N) in the codebook vectors to determine which of the S codebook vectors and associated gain gives the best match to the input speech.

Abstract: An improved process is described for depositing TiW/TiWN/TiW/Au metallization which provides superior adhesion properties, excellent barrier properties and which is suitable for use with metal line widths of the order of one micron or smaller. It is important in order to obtain these properties to ensure that the layer immediately underlying the gold layer by substantially pure TiW deposited in a nitrogen free sputtering atmosphere. To this end, the gas supply manifolds and deposition chamber are purged and the chamber evacuated following deposition of the TiW layer and prior to deposition of the TiWN layer underlying the gold layer. A final TiW layer is also conveniently placed on top of the gold layer to act as an etching mask.

Abstract: The synthesizer includes an output PLL having a divide-by-N1 divider in its feedback loop. The output PLL is couples through frequency offset circuitry to receive a reference signal from a driver PLL having a divide-by-N2 divider in its feedback loop. Another divide-by-N1 divider coupled a reference oscillator to the driver PLL. The reference oscillator provides another reference signal. As a result, the setting for N1 controls the course frequency tuning and the setting for N2 controls the fine frequency tuning of the synthesizer which provides any one of a plurality of selectable predetermined output frequencies. The adjacent selectable frequencies are closer together than the frequencies of the reference signals. The synthesizer has a simple configuration and provides a high degree of output frequency resolution, fast acquisition and low noise.

Abstract: An improved high frequency dielectrically isolated (DIC) transistor (100) or integrated circuit is obtained by providing a highly doped single crystal semiconductor region (112) coupled to the device reference terminal (16') and extending between front (98) and rear (61) faces of the semiconductor die. This allows the reference terminal (16', 116) to be coupled to the package ground plane without use of wire bonds, thereby lowering the common mode impedance. The desired structure is formed in connection with DIC devices (100) by etching first (66) and second (77) nested cavities into a single crystal substrate (60). The cavities (66) form protruding islands (821, 822) of single crystal semiconductor having a height (80+68) about equal the final die thickness (110) and which, after conventional DIC processing using an oxide isolation layer (86) and a poly handle (88), are exposed by grinding away the poly handle (88) to expose the highly doped, single crystal reference terminal feed-through (112).

Abstract: A MOSFET having a back-side source contact and top-side gate and drain contacts is provided by a structure comprising superposed N.sup.+, N-, P-, N.sup.+ regions arranged between top and bottom surfaces of the semiconductor die. In a preferred implementation, two trenches are etched from the top surface to the P-, N.sup.+ interface. A buried P-, N.sup.+ short is provided in one trench and a gate dielectric and gate electrode are provided over the sidewall P- region exposed in the other trench. This creates a vertical MOSFET in which the N.sup.+ substrate forms the source region shorted to the P- body region in which the channel is created by the gate. Superior performance is obtained in RF grounded-source circuit applications.

Abstract: Improved resistance to electrical instability of opto-isolators subjected to large stand-off voltages is obtained by coating the semiconductor light sensing element with a high resistivity layer of amorphous silicon while leaving most of the surface PN junction perimeter and nearby regions free of metal. The amorphous silicon prevents mobile ions in the encapsulation, which are driven to the detector surface by the stand-off voltage, from inverting or modulating the conductivity of the detector surface and causing instability. The amorphous silicon also makes it possible to leave most of the light sensitive PN junctions and nearby regions free of metal, thereby simplifying design of complex IC detector chips and increasing sensitivity.

Abstract: An improved method and system is described for allocating manufacturing or process resources having multiple constraints thereon to meet various time varying manufacturing or service demands having multiple parameter requirements.The demand requirements are expressed as a multi-dimensional time ordered array of vertices D.sub.q [t,r,j], for each demand q, wherein t is time and r identifies the physical requirements associated with the q.sup.th demand and j is an integer index running from 1 to J wherein J is the total number of times wherein r has differing values. The available resources are expressed as a multi-dimensional time ordered array of resource vertices R.sub.p [t,c,i] for each resource p, where c expresses the physical capacities associted with the p.sup.th resource, t is time and i is an integer index running from 1 to I where I is the total number of times wherein c has differing values.A logical system is provided for comparing R.sub.p and D.sub.q to determine when and how D.sub.

Abstract: The transmission line is coupled to a charging circuit and to a series circuit including a terminating resistor, the source and drain electrodes of a MOSFET and laser diode. Trigger signals are applied to the gate electrode of the MOSFET to rapidly discharge the transmisson line to provide currents of about 10 amps through the laser diode to provide pulses of light having durations of about 3 nanoseconds.

Abstract: An improved damper diode is obtained by replacing the usual step junction (P.sup.+ -Nu-N.sup.+) structure of the prior art with an epitaxial double sided Pi-Nu structure (i.e., P.sup.+ -Pi-Nu-N.sup.+) in which the thickness and impurity concentrations of the Pi and Nu regions are substantially equal and have a combined thickness about equal to the prior art Nu region for the same voltage. Improved transient response (TFR), better transient energy absorption capability (UIS) and lower forward transient turn-on peak over-shoot voltage (TOPO) is obtained for the same or higher reverse breakdown voltage (BVR), in the same or smaller die size.

Abstract: An improved contact is obtained to power devices having a raised dielectric region adjacent the die contact region by providing a contact dimple on the otherwise flat metal lead used for die contact. The dimple is arranged above the die contact and soldered thereto. The radius of curvature and depth of the dimple is adjusted so that the contact lead is far enough away from the edge of the surrounding raised dielectric at the edge of the die contact to provide a laterally concave outward air-solder interface in that location. This prevents solder creep onto the dielectric surface and avoids die edge shorting. Several dimple shapes are described.

Abstract: Distortion introduced in phase modulated RF signals by amplifiers and other signal processors is substantially reduced by modifying complex modulating signals I,Q in a way that compensates for the amplifier phase error .phi..sub.e as a function of amplifier power P. In a preferred embodiment, I, Q are combined to calculate the amplifier power level and scaled by a parameter, e.g., K1=-d.phi..sub.e /dP, to provide factor B so that compensated signals I"=AQ+BI and Q"=AI-BQ, where A is a constant, can be determined. When an RF carrier modulated by I", Q" is passed through the distorting amplifier the amplifier induced distortion is cancelled. Substantial reduction in bit error probability is obtained. The method can also be used for post-distortion correction.