Abstract: A speaker identification system (10) employs a supervised training process (100) that uses row action projection (RAP) to generate speaker model data for a set of speakers. The training process employing RAP uses less memory and processing resources by operating on a single row of a matrix at a time. Memory requirements are linearly proportional to number of speakers for storing each speakers information. A speaker is identified from the set of speakers by sampling the speaker's speech (202), deriving cepstral coefficients (208), and performing a polynomial expansion (212) on cepstral coefficients. The identified speaker (228) is selected using the product of the speaker model data (213) and the polynomial expanded coefficients from the speech sample.

Abstract: A modular phased array antenna for the formation of simultaneous independently steerable multiple beams, the modular phased array antenna comprising a modular array including a plurality of sub-array modules combined together in close proximity, each one of the plurality of sub-array modules including a plurality of input modules, a layer of a plurality of radiating antenna elements, a plurality of stacked beamformers arranged in series and each connected to one of the plurality of input modules and to the plurality of radiating antenna elements in beam communication.

Abstract: A key management system includes a hierarchy (10) of independent key arbitration centers (KAC) for providing access to a user's session keys through key management centers (KMC). When a court order is issued for a user's session keys, a message requesting the keys is transferred down through hierarchy until a terminal KAC (16,36) is reached. Each KAC in the hierarchy adds its ID and signs (116) the message, verifying prior signatures (114). The user's ID is encrypted with the terminal KAC's public key. The terminal KAC engages in a blind key access procedure (129) with the KMC (18,38) to receive the user's session key. The key is provided encrypted with the requesting party's or agency's public key. Accordingly, privacy is assured because only the KMC and the requesting agency have access to the actual key value, and only the terminal KAC and requesting agency have access to the user's ID.

Abstract: A secure communications system (501) including a local digital terminal (510) coupled to an interworking function (538). The interworking function (538) couples the digital radio network (121) to a PSTN (542, 546, 548) and a remote analog communication terminal (552). The system (501) includes a novel method of establishing an end-to-end communication channel between local digital (510) and remote analog (552) terminals wherein the local digital terminal establishes (510) a direct digital channel (121) between itself (510) and the interworking function (538) and transmits a message describing its signaling capabilities to the interworking function (538). The interworking function (538) then trains its modem with the modem of the remote analog terminal (552) such that signaling capabilities of the local digital terminal (510) and the remote analog terminal (552) are not violated.

Abstract: A transceiver (10) in a first embodiment includes a zero intermediate frequency (ZIF) receiver (11), which achieves high throughput operation by reducing receiver latency time due to receiver response within a receiver pass band to carrier spectral components of a modulated carrier frequency. During a receive mode, a receiver local oscillator (LO) frequency is offset from the carrier frequency within the receiver pass band, preferably as determined by worst case errors in the carrier frequency and the LO frequency. During a transmit mode, the receiver LO frequency is modulated in such a way as to suppress the LO center frequency component in the modulated spectrum.

Abstract: A packet switching communication network (10) includes a number of switching nodes (16). The switching nodes include a switch (18) and at least one data processor (20). When selected packets (34), as identified by routing codes (40), arrive at one of the switching nodes (16), they are passed (58) to the data processor (20). The data processor (20) is configured to perform a network processing function, such as packet broadcasting (74) or error correction (60). In performing these functions, user data (38) conveyed by the selected packets (34) are processed to generate processed packets. The processed packets are re-introduced to the switch (18) and routed onto their destinations.

Abstract: A two-wire line and a four-wire line telephone system interface with each other through a hybrid (12). After a connection between the two and four-wire line systems is made, an echo canceler (10) provides training signals to the hybrid on the transmit path an generates echo cancellation signals on the receive path. An adaptive filter (16) is employed to provide the proper echo cancellation signals, while a controller (22) provides filter coefficients based on an error signal. The training signals include the convolution of an impulse function with a sin(x)/x function. Once the filter coefficients are determined for the particular call, the transmit path of the four-wire line system is switched in and voice data between the two systems my commence.

Abstract: An RF power detector (10) includes an RF detector circuit (12), a nonlinear feedback amplifier (14), a temperature compensation circuit (16) and a linear feedback amplifier circuit (18). The RF detector circuit converts an RF signal to a voltage representative of the RF signals power level. Nonlinear feedback amplifier (14) nonlinearly amplifies the voltage and compensates for the nonlinearities of the RF detector circuits detector elements. Temperature compensation element provides a temperature compensation signal to compensate for the temperature effects of the detector elements of RF detector circuit (12). The output signal of RF power detector (10) is a substantially linearly representation of the RF input signals power level. Nonlinear feedback amplifier (14) includes a nonlinear feedback circuit (30) with a nonlinear feedback element (32) in the feedback path of op-amp (42).

Abstract: An intelligent digital beam former (10) in conjunction with a satellite based array antenna (20) provides a plurality of dynamically controllable antenna beams (52) for communication with subscriber units on earth's surface. The location of a subscriber unit (90) requesting communication services is determined and an individual antenna beams is formed and assigned to the subscriber unit. The antenna beam tracks the subscriber units as the satellite and/or the subscriber unit moves. The digital beam forming coefficients are dynamically adjusted and controlled to help maximize the signal quality of communications with the subscriber units.

Abstract: Crosspolarized signals of unrelated baud rates transmitted through a communication channel (10) become depolarized due to channel distortions such as rain and antenna imperfections. The resulting interference is canceled in a Modified Adaptive Crosspole Interference Canceler (MAXPIC) receiver (50, 70) by adjustment the timing in the crosspolarization path to compensate for the differential delay. Near equivalent bit error rate (BER) performance is achieved for systems utilizing crosspolarized signals with independent baud rates. In one embodiment, the receiver uses a fractionally spaced finite impulse response (FIR) filter (78) that operates at an integer multiple of the direct channel signal baud rate. In another embodiment, a variable delay (54) is used to time-align the received crosspolarized channel signal with the crosspolarization interference contained in the received direct signal.

Abstract: A method (140) for tuning millimeter-wave FET amplifiers (20) during manufacture, through the application (144) of a gate bias voltage (52) so as to tune the FET (22) of the amplifier (20) to match an input circuit (24), and through the application (146) of a drain bias voltage (74) so as to tune the FET (22) of the amplifier (20) to match an output circuit (26), then measuring (150) the frequency response of the amplifier (20). This tuning method (140) is repeated (152) until a predetermined frequency response has been achieved. Once achieved, the predetermined frequency response is realized (154) by permanently fixing the gate bias voltage (52) and the drain bias voltage (74) at the determined values. This iterative method (140) of tuning amplifiers (20) is then repeated for all amplifiers (20) to be tuned.

Abstract: An extremely broad band high power termination (10) for microwave and millimeter frequency amplifiers combines a standard resistive low frequency termination (15) with a broad band high frequency absorptive element (13) using an absorptive material such as Eccosorb. A mid-band matching network (14) is provided between the resistive termination (15) and the Eccosorb absorptive element (13). The Eccosorb absorbs the energy of the higher microwave frequencies while the resistor absorbs energy at low frequencies. Accordingly, a much higher power handling capability in a compact planer environment is achieved. This termination (10) is suitable use for use in K-band power amplifier combiners (30) that require high isolation and high power handling capability of the isolated ports (35).

Abstract: In a constellation of orbiting satellites (12) in non-geostationary orbits, a failing satellite is replaced with a spare satellite from a spare orbit (53). The satellites (12) are in one of several orbital planes (14) that have an inclination angle with respect to the equator. An orbiting spare satellite located in a monitor orbit (55) is provided for each orbital plane (14). When a failing satellite in the operational orbit is detected, the spare satellite replaces the failing satellite by exchanging positions with the failing satellite. The failing satellite may remain in the monitor orbit to subsequently replace another failing ("sicker") satellite. Accordingly, the satellites that are the "healthiest" are maintained in operation. Cross-links (23) between satellites are maintained minimizing any disruption in communication service when a hole occurs in an orbital plane (14) as a result of a failure.

Abstract: A radio telecommunication network (10) includes satellites (12) which project cells (30) on the surface of the earth. Mobile units (16) are alerted to incoming calls when they receive ring signals. Ring signal coverage parameter values are selected in response to differentiating data (46) which cause different ring signals to be activated for mobile units (16) which appear to be proximately located. An illumination area ring signal coverage parameter may have a value selected in response to differentiating data (46) so that larger illumination areas are used for higher speed mobile units (16"). A transmission power ring signal coverage parameter may have a value selected in response to differentiating data (46) so that higher power ring signals are used in specific situations identified by the differentiating data (46).

Abstract: A telecommunication network (10) includes satellites (12) in which cellular base station switches (42) are located. Each switch (42) is hardware configured to require the use of relatively few interconnection segments but would suffer a significant amount of call blocking if operated in a conventional manner. A new call setup process (96) is performed whenever a request to setup a new call is received at the switch (42). The process (96) defines new connections through switch 42 for ongoing calls while simultaneously defining a new connection for the new call. In addition, the process (96) forces the new connections to use only those interconnection segments that are reasonably required to implement the new connections. Segments that are not required are deactivated (128) to reduce power consumption.

Abstract: Quasi-stripline filters (20) are tuned during manufacture to achieve consistent frequency response. A filter structure (20) which accommodates the method (60) is provided. The microstrip filters (30) are fabricated by forming conductive filter patterns (22) upon non-conductive (dielectric) substrates (24), and then bonding those substrates (24) to conductive (metallic) substrates (28). These filters (30) are then entunneled with conductive housings (34) electrically and physically joined to the conductive plates (28). Housings (34) of varying cross-sectional areas (40) are chosen to produce the desired frequency response, thus converting the microstrip filters (30) into quasi-stripline filters (20) with a specific frequency response.

Abstract: A communications network (10) includes nodes (18) in which switched channelizers (32) decompose wideband signals into narrowband signals and route the narrowband signals to specified demodulators (34). The switched channelizers (32) serve as building blocks which may be coupled together in parallel and in series. Switch portions (28) of switched channelizers (32) may couple together so that the switched channelizers (32) form interconnected switched channelizer groups (54). Within an interconnected switched channelizer group (54), any channel from any wideband signal may be routed to any demodulator (34).

Abstract: A system (28) provides communication services to mobile units (24). Mobile units (24) perform a unit-based location process (46, 48) to determine their locations. The system (28) performs a system-based location process (64, 66) to independently determine mobile unit locations. The system-based process may determine location less precisely than the unit-based process. A location selection process (104) evaluates the unit-determined location in view of a system-determined location error region to decide whether the unit-determined location is reliable. If the unit-determined location is reliable, it is used (120) to qualify communication services to be provided to the mobile unit (24). If the unit-determined location is unreliable, the system-based process is repeated (116) to improve the system-determined location precision, and the resulting system-determined location is used (118) to qualify communication services to be provided to the mobile unit (24).

Abstract: A satellite assembly (60, 68, 70) is formed from any number of bus modules (22) which have a substantially common shape and interior space volume. Each bus module (22) includes a structural frame (20) which is part of a structural subsystem and at least one and possibly all of a propulsion subsystem (28), a power subsystem (30), a thermal subsystem (32), an attitude, orientation and control subsystem (34), a telemetry, tracking and control subsystem (42) and a payload subsystem (44). Within each bus module (22), the subsystems are substantially non-redundant. Bus modules (22) attach together permanently or temporarily through attachment mechanisms (36). Permanent attachment is used to form large and/or redundancy within the satellites. Temporary attachment is used to increase the structural rigidity of individual bus modules (22) for launch purposes, then the assembly (60) is decomposed into individual satellites.

Abstract: A series capacitance is included in both surface ground conductors of coplanar waveguide on an MMIC die. The capacitance compensates for the inductance of bond wires used for transitioning from microstrip line to coplanar waveguide. The capacitance is chosen to resonate with the inductance of the bond wires at the desired frequency. As a result, virtually all microstrip mode ground signal currents are transitioned from the ground plane of the microstrip line to the surface ground conductors of the coplanar waveguide. Microstrip modes affecting the coplanar waveguide are significantly reduced improving isolation between ports of the MMIC die.