Abstract: The present invention provides a method for recognizing connected and degraded text embedded in a gray-scale image. In accordance with the invention, pseudo two-dimensional hidden Markov models (PHMMs) are used to represent characters. Observation vectors for the gray-scale image are produced from pixel maps obtained by gray-scale optical scanning. Three components are employed to characterize a pixel: a convoluted, quantized gray-level component, a pixel relative position component, and a pixel major stroke direction component. These components are organized as an observation vector, which is continuous in nature, invariant in different font sizes, and flexible for use in various quantization processes. In this matter, information loss or distortion due to binarization processes is eliminated; moreover, in cases where documents are binary in nature (e.g.

Abstract: An Add/Drop circuit for use in optical wavelength-division multiplexed systems includes fiber coupled four-port polarized beamsplitters and fiber Bragg grating filters.

Abstract: A compact waveguide grating router is constructed by forming at least first and second free space regions, first and second sets of optical ports, and a grating array, onto a substrate. The grating array is coupled to the first free space region and the second free space region. A first plane defining a first reflective surface is introduced into the first free space region, and a second plane defining a second reflective surface is introduced into the second free space region. The first plane is oriented such that incident optical signals from the first set of optical ports are reflected to the grating array, and the second plane is oriented such that incident optical signals from the grating array are reflected to the second set of optical ports. The first and second planes are provided, for example, by cleaving, cutting, etching, sawing, or any other suitable technique. The first and second planes enable the input and output ports to be folded back into previously unoccupied space.

Abstract: A pair of SONET rings (12' and 14') can be interworked by providing at least two gateways (16.sub.m-1 and 16.sub.m) that are shared by each of the rings. Each shared gateway has the capability of transferring a block of optical information resident in a prescribed time slot in an interchange frame associated with one ring to a time slot in an interchange frame associated with the other ring so the block reaches its intended destination in that ring.

Abstract: An interactive video/computer pointing system utilizing a magnetic sensor to derive relative azimuthal information, and an inclinometer or accelerometer to provide relative angular elevation information. The azimuthal information is processed to yield an indication of any horizontal movement of the pointing device, and the angular elevation is processed to yield an indication of any vertical movement of the pointing device. This horizontal and vertical movement information is utilized to responsively control a video cursor, thereby enabling the user to point to and select various regions upon a video screen by manipulating the pointing device.

Abstract: Automatic call forwarding techniques for use in conjunction with a telephonic communications system such as a PBX or a central switching office. A telephonic device is associated with a smart card reader/writer and a communications unit RF transceiver. Subscribers are each provided with a smart card and an RF jacket transceiver adapted to partially and removably encase the smart card. In response to input received from the smart card reader/writer, the telephonic device instructs the communications system to forward incoming telephone calls for a given subscriber to the associated telephonic device. The communications unit RF transceiver periodically transmits an interrogation signal to the RF jacket transceiver. If the interrogation signal is received by the RF jacket transceiver, the RF jacket transceiver sends an acknowledgment signal to the communications unit RF transceiver, and incoming calls will continue to be forwarded to the associated telephonic device.

Abstract: In a communications network techniques are disclosed for optimizing node interconnection based upon the number of communications pathways to be provided by each of the nodes. For each node, the number of communications pathways to be provided by that node is represented in the form of a one-dimensional array. Image reconstruction techniques are applied to this one-dimensional array to generate a two-dimensional image. The one-dimensional array is used to generate vertical columns of the two-dimensional image, or alternatively, horizontal rows of the two-dimensional image. The two-dimensional image, comprised of a pixel array, is a mathematical representation of the communications network. Each pixel of the two-dimensional image represents a specific pair of nodes. Each pixel has a gray-scale value that signifies the number of communications pathways that are to be provided between this specific pair of nodes.

Abstract: Full channel service is obtained over a modified ISDN BRI interface by emulating at least one central office (CO) station terminal which provides access to a voice channel over a single digital subscriber line DSL for user station terminals whether or not they are ISDN BRI compatible. To this end, an interface is provided for the DSL that emulates an ISDN BRI CO station terminal and acts like a facility to user station terminals. More specifically, a plurality of CO station terminals are emulated on a single multi-point passive bus terminating the single DSL. Each of the emulated CO station terminals presents to the CO switching office the appearance of being a CO station terminal. The emulated CO station terminals are configured such that CO features are accessible. Information from the emulated CO station terminals is mapped into existing system structures where such structures are available and compatible with the information from the emulated terminal.

Abstract: Thin semiconductor die, approximately 0.004 to 0.007 inches thick, are positioned substantially on the neutral plane of a smart card, the neutral plane defined as the plane of substantially no mechanical strain during flexure of the smart card, thereby providing smart cards having improved resistance to mechanical flexure, and/or smart cards having improved RF performance.

Abstract: Certain material and objects can be characterized by a frequency-dependent absorption, dispersion, and reflection of terahertz transients in signals which pass illuminate the material or object. The present terahertz imaging system analyses that frequency dependence in the time-domain by collecting that transmitted signal propagating through the object and then processing the information contained in those signals for every point or "pixel" on that object. This is a non-invasive imaging technique that is capable of differentiating between different materials, chemical compositions, or environments.

Abstract: An improved method of speech coding for use in conjunction with speech coding methods wherein speech is digitized into a plurality of temporally defined frames, each frame including a plurality of sub-frames, and the digitized speech is partitioned into periodic components and a residual signal. For each of a plurality of sub-frames of the residual signal, the improved method of speech coding selects and applies a time shift T to the sub-frame by applying a matching criterion to (a) the current sub-frame of the residual signal, and (b) a sample-to-sample (subframe-to-subframe) pitch delay determined by applying linear interpolation to known pitch delays occurring at or near frame-to-frame boundaries of previous frames.The matching criterion is applied by minimizing .epsilon.

Abstract: Low optical loss and simplified fabrication are achieved by a nonlinear reflector which incorporates one or more semiconductor quantum wells within an n half-wavelengths strain relief layer (where n is an odd integer greater than zero) that is formed on a standard semiconductor quarter wave stack reflector. Growth of the half-wavelength layer is controlled so that dislocations are formed in sufficient concentration at the interface region to act effectively as non-radiative recombination sources. After saturation, these recombination sources remove carriers in the quantum well before the next round trip of the optical pulse arrives in the laser cavity. The nonlinear reflector is suitable for laser modelocking at the high wavelengths associated with many currently contemplated telecommunications applications and provides, at such wavelengths, an intensity dependent response that permits it to be used for saturable absorption directly in a main oscillating cavity of a laser.

Abstract: In a speech coding system which encodes speech parameters into a plurality of frames, each frame having a predetermined number of bits, a predefined number of bits per frame are employed to transmit a speech parameter delta. The speech parameter delta specifies the amount by which the value of a given parameter has changed from a previous frame to the present frame. According to a preferred embodiment disclosed herein, a speech parameter delta representing change in pitch delay from the present frame to the immediately preceding frame is transmitted in the present frame, and the predefined number of bits is in the approximate range of four to six. The speech parameter delta is used to update a memory table in the speech coding system when a frame erasure occurs.

Abstract: A signal processing system processes an input signal in real time. The system includes a light valve array for selectively attenuating light incident thereupon in response to electronic control signals, a light source for providing illumination, a light-sensitive array for producing electronic output signals representing a processed version of the input signal in response to light incident thereupon, and control circuitry for generating and applying electronic control signals to the light valve array in response to the input signal. The electronic control signals are generated from the input signal and are adapted for implementing an arbitrarily-selected signal processing operation. The light valve array is situated between the light source and the light-sensitive array, such that the light source is selectively attenuated by the light valve array, and such that the selectively attenuated light is incident upon the light-sensitive array.

Abstract: Improved RF switching techniques operate in conjunction with an existing cellular telephone having an RF amplifier, an antenna, receiver circuitry, a duplexer, and a battery. The duplexer has a first port adapted for connection to the receiver circuitry, a second port adapted for connection to the RF amplifier, and a third port adapted for connection to the antenna. The improved RF switching techniques are characterized by a first switch, a second switch, and a battery of less than 6.0 volts. The first switch, coupled to the RF amplifier, connects the RF amplifier to any one of the antenna and the second port of the duplexer. A second switch, coupled to the antenna, connects the antenna to any one of the RF amplifier and the third port of the duplexer. During transmit-only operation, the first switch connects the RF amplifier to the antenna.

Abstract: Digital video transmission rate matching techniques are disclosed. A bit stream rate matching apparatus includes a bit rate increasing device and a bit rate reducing device. The bit rate increasing device converts a video bit stream having a first transmission rate to a video bit stream having a second transmission rate wherein the first transmission rate is less than the second transmission rate. The bit rate reducing device converts a video bit stream having the second transmission rate to a video bit stream having the first transmission rate. The bit stream rate matching apparatus is useful in the context of a multimedia conference where a first endpoint device employs the first transmission rate and a second endpoint device employs the second transmission rate.

Abstract: A circuit for maintaining the magnitude of a radio frequency (RF) signal at least one predetermined level employs a detection circuit coupled to an RF signal generator by a four port coupling device. In one embodiment of the present invention, an impedance inverter in series with a diode is connected between the load termination of the coupling device and ground. The series combination of the impedance inverter and diode approximates a short circuit as the output signal power approaches a predetermined low amplitude threshold and approximates an open circuit as the output signal power approaches a predetermined upper amplitude threshold so that gain is provided to signals received by the detection circuit below a first amplitude level and attenuation is provided to signals received by the detection circuit above a second amplitude level.

Abstract: A telephone instrument with computing means and a memory that stores a number of dialed numbers. The computing means follows a process that is responsive to the "re-dial" button of the telephone instrument and to the "#" and "*" buttons on the dial pad. The process provides a "backup" capability that points to successively older entries in the memory, thereby allowing a user to dial out older entries, and also provides a search function that searches through the memory for a specified number.

Abstract: Signal reconstruction methods reconstruct a missing portion of a signal from a first known portion of the signal preceding the missing portion, and a second known portion of the signal succeeding the missing portion. The first known portion of the signal is represented using a first autoregressive model and the second known portion of the signal is represented using a second autoregressive model. These methods construct a signal estimate from the first and second autoregressive models, thereby permitting the reconstruction of any signal having a missing portion, irrespective of whether or not the length of the missing portion is greater than the short-term stationarity duration of the signal.

Abstract: Methods of establishing multimedia conferences between a multimedia server and a plurality of endpoint devices. One or more of these endpoint devices are enhanced in that they include a memory adapted to store one or more telephone numbers, each telephone number corresponding to a respective endpoint device, and a user interface coupled to the memory for selecting a telephone number stored in the memory. The methods are performed in a substantially continuous manner, signifying that a third endpoint device is added to a preexisting conference between a first and a second endpoint device without substantially interrupting this preexisting conference. Therefore, a seamless multimedia conference is provided between the first and second enhanced endpoint devices even when a third endpoint device is added to the conference.