Abstract: A method downsamples characters in a document by font substitution when the characters in the document are identified with sufficient reliability. Otherwise, the characters are downsampled by decimation. The method also allows downsampling by applying characters to a two-dimensional nonlinear filter and decimating.

Abstract: A method of admitting and routing switched virtual circuit requests in a network first finds a set of routing paths on which a requested VC may be routed by using a two step process. The method uses a cost function based on a parameter related to the number of hops in a subset of VC connections previously made in the network to determine potential routing paths on which the VC can be routed at a cost below a specified threshold. The method next checks to determine which potential routing paths comprise links and nodes with sufficient resources to accommodate the request. Paths satisfying both steps are output as a set of routing paths and then a second criterion is used to select a path from the set on which to route the request. In a distributed routing system, the inventive method uses a local network state to determine the cost function and the set of routing paths.

Abstract: A method regulates the admission control of, and requests for routing of, virtual circuits in a network by determining network resource requirements for the virtual circuits. In particular, the network resource requirements are based on a set of parameters used to control the flow of information from a communications device onto the virtual circuit. The requirements for network resources typically include buffer space requirements in network nodes and bandwidth requirements in network links, and the parameter used to control the flow of information are those associated with an access regulator. The network resource determination is made for the case where lossless performance in the network is required and in the case where statistical multiplexing with limited loss is allowed. Both constant bit rate and variable bit rate information is considered.

Abstract: A method of operating a network by sharing resources associated with various classes among calls in the various classes according to a state dependent reservation parameter. Nominal amounts of one or more resources are allocated to each call class. When a call of a class of service operating in the network requires resources in excess of those allocated to the class, resources allocated to other classes of service are advantageously shared with the class of service. The sharing is based on a reservation parameter associated with the class of service of the call. The reservation parameter is advantageously a function of the network state. The role of the reservation parameter is to protect underloaded classes (i.e., those classes not using all of their allocated nominal capacity) from excessive borrowing by overloaded classes (i.e., classes using more than their allocated nominal capacity).

Abstract: A method and apparatus aligns device components based on an analysis of an image of features of the components to be aligned. The image is analyzed for deviation in component alignment, and a signal, based on the deviation, is sent to a positioning apparatus to adjust the position of the components relative to each other. The image of the components is advantageously a top view which permits measuring deviations in alignment of features on the top surfaces of components. The components are advantageously bonded together after alignment, and the alignment accuracy of the bonded components is checked.

Abstract: A method and apparatus measures a deviation in alignment between two components of a product wherein each component is characterized by a respective pattern and wherein each pattern comprises respective elements. The centroid of each element in each respective pattern is indicated, and based on the centroid indications, the center of each respective pattern is indicated. The deviation in alignment between the two components is determined as a function of the indicated position of the centers of the respective patterns, and positioning of the components is controlled as a function of the indicated position of the centers of the respective patterns.

Abstract: A method identifies and models the distortion in a document using a two-dimensional nonlinear filter. The two-dimensional nonlinear filter is then used to improve the performance of optical character recognition systems in recognizing characters in documents that have been degraded, as for example by repeated photocopy reproductions or by fax transmission.

Abstract: In CELP coding, stochastic (noise-like) excitation is used in exciting a cascade of long-term and short-term all-pole linear synthesis filters. This approach is based on the observation that the ideal excitation, obtained by inverse-filtering the speech signal, can be modeled for simplicity as Gaussian white noise. Although such stochastic excitation resembles the ideal excitation in its global statistical properties, it contains a noisy component that is irrelevant to the synthesis process. This component introduces some roughness and noisiness in the synthesized speech. The present invention reduces this effect by adaptively controlling the level of the stochastic excitation. The proposed control mechanism links the stochastic excitation to the long-term predictor in such a way that the excitation level is inversely related to the efficiency of the predictor.

Abstract: The invention is used to shape noise in time domain and frequency domain coding schemes. The method advantageously uses a noise weighting filter based on a filterbank with variable gains. A method is presented for decoding an encoded signal based on received side information and on a masking matrix derived from the masking properties of speech. In particular, the encoded signal is separated into subband signal components, each of which is multiplied by a corresponding gain value based on the masking matrix. These multiplied subband signal components are then combined to produce a decoded signal.

Abstract: A method maintains information associated with items in a database of limited memory which information is used to generate representations of the information such as high-biased histograms. In a first embodiment of the inventive method, information associated with all items with sales above a threshold, together with approximate counts of the items, is maintained. Appropriate choice of a threshold limits the amount of information required to be maintained so as to generate accurate representations of the information with high probability. In a second embodiment of the inventive method, information used to generate a high-biased histogram is maintained within a fixed allotment of memory by dynamic adjusting a threshold which threshold is used to determine a probability with which information is retained in the database.

Abstract: A method and apparatus is disclosed for conformance testing of protocols specified as a collection of communicating finite state machines (FSMs). A guided random walk procedure is advantageously used to cover a substantial number of transitions in the component FSMs to determine if the FSM output corresponds to the protocol as specified.

Abstract: An integrated method for congestion control uses access regulator to control the admission of information from a communication device into a network according to a predetermined function characterized by a set of parameters. A node within the network sends a signal indicating a level of congestion in the node to the access regulator, and at least one of the parameters is adjusted in response to the signal. The parameter may additionally be adjusted according to other parameters which are selected according to stability and transient response criteria.

Abstract: The invention is used to shape noise in time domain and frequency domain coding schemes. The method advantageously uses a noise weighting filter based on a filterbank with variable gains. A method is presented for computing the gains in the noise weighting filterbank with filter parameters derived from the masking properties of speech. Illustrative embodiments of the method in various coding schemes are illustrated.

Abstract: A method and apparatus generates a channel codeword based on codewords with arbitrary block digital sums. Respective portions of the channel codeword are generated based on respective sets of input symbols, and the channel codeword is generated from the portions in accordance with another set of input symbols. The potions are advantageously codewords, comprising symbols, generated by selecting, for each set of input symbols, a codeword from a codebook and by adapting, as for example by ordering or by inverting symbols in, the codewords to form the channel codeword.

Abstract: The invention discloses a method and apparatus for assigning link "distance" metrics that result in near optimal routing for a network formed of nodes (routers) and links, where each link has a capacity associated with it, and where source-destination flows are given. The routing optimality is measured with respect to some objective function (e.g., average network delay).

Abstract: A new method for high quality speech coding, Timing-Frequency Interpolation (TFI) which offers advantages over conventional CELP (code-excited linear predictive) algorithms for low rate coding. The method, provides a perceptually advantageous framework for voiced speech processing. The general formulation of the TFI technique is described.

Abstract: A method for simplifying computer protocols is disclosed which offers advantages in reducing the memory and processing requirements for invoking the protocol. In accordance with the present invention, a protocol is represented as a collection of communicating finite machines. The subset of services offered by the protocol desired by a user is also specified as a finite state machine. The method finds the thinned version of the protocol by comparing the finite state machine representation of the protocol with the finite state machine representation of the subset of desired services and keeps only that part of the protocol necessary for providing the desired subset of services.

Abstract: A method for sending and displaying customized video images with telecommunications devices is disclosed. In one embodiment of the invention, a video-caller identification scheme is described in which in-band signalling is used to transmit video data.

Abstract: A method of network multicasting using transport layer protocols, suitable for use in ATM networks, is disclosed. The protocols advantageously combine the acknowledgments of individual destinations along the underlying multicast tree to prevent acknowledgement implosion. The inventive method also prevents unnecessary retransmission by performing local multicasts.

Abstract: A family of extended Generalized Viterbi Algorithm (GVA) and Soft Output Viterbi Algorithm (SOVA) decoders is disclosed. In one embodiment structural aspects of a SOVA are used to develop a new implementation for a serial GVA. Another aspect of the invention provides a low complexity SOVA that generates a reliability measure for decoded information using a list output of the GVA. In a further aspect, reliability information from a SOVA is used to develop a low complexity GVA.