Abstract: The method of the present invention transforms descriptor vectors that characterize items partitioned into groups into a space that discriminates between those groups in a well defined optimal sense. First data is generated that represents a differences between the groups of descriptor vectors. Second data is generated representing variation within the groups of descriptor vectors. A set of component vectors is then identified that maximizes an F distributed criterion function that measures differences of descriptor vectors between groups relative to variations of descriptor vectors within groups. A statistic is generated for subsets of the component vectors. For each particular subset of component vectors, a probability value for the statistic associated with the particular subset is calculated. The subset with the minimum probability value is selected. Finally, one or more of the descriptor vectors for the items are mapped to a space corresponding to the selected subset of component vectors.

Abstract: A fast trace log component for tracing the operation of one or more processes in a multi-processing operating system is described. When instantiated, the component allocates a common buffer within an area of memory and trace information supplied by a process is written to the common buffer. The component includes program codes for obtaining a next address within the buffer at which to write the trace information and for incrementing the next address to an address in the buffer after the trace information; and program codes for flushing the buffer if the trace information would exceed the end of the buffer.

Abstract: A data processing system comprises means for identifying and replacing instructions to jump to functions having known prolog instructions with modified jump instructions, means for storing the known prolog instructions, and means for retrieving the known prolog instructions when such modified instructions are found and for supplying the known prolog instructions for processing. A compiler or preprocessor is arranged to detect and modify the jump instructions. A logic module is arranged to intercept the modified instructions, retrieve from its storage the prolog instructions, and supply the prolog instructions for processing. The compiler or preprocessor is further arranged to detect and modify the first instruction of known epilog code. A logic module is arranged to intercept the modified instruction, and to retrieve and supply for processing the epilog instructions.

Abstract: The present invention utilizes a set of D descriptors for each of N items. A value K′ representing a number of descriptors, and a value K for a number of items that should support an hypothesis are generated, preferably via user input. Collections involving K or more items for which there is an association involving a selection of values across at least K′ of the D descriptors are identified, and preferably reported to the user.

Abstract: The invention describes methods for locating a treatment device disposed within a living body by means of magnetic fields that are produced by Barkhausen jumps, principally from amorphous tag wires with high permeability that exhibit reentrant flux reversal. When wires of this type are attached to concealed treatment devices such as catheters, interrogation or scanning of the tag wire by a low frequency ac magnetic field affords an accurate means for locating the treatment devices using a sensor coil to detect the magnetic field signal from the wire locating tag. The strength of the field detected by the position of a sensor coil with respect to the locator tag is used to determine the location of the tag. A favorable signal to noise detection ration is obtained as the signal emitted by the wire is at a very high frequency compared to that of the frequency of the interrogation field.

Abstract: A method and system of processing within a Java Virtual Machine on a computer system, a native function call instruction of a dynamic parameter set type and contained in a Java byte code application. The method comprises the steps of; interpreting the byte code representing a native function call instruction of the dynamic parameter set type; storing the parameters of the variable parameter set in an array; using the computer system to process the native function call as a fixed parameter set type using the array as the fixed parameter set. Before the function is called the computer system transfers the parameters from the array to a stack used by the native function.

Abstract: The invention provides a method and apparatus by which the usage of an existing ring network is extended by implementing spatial reuse of the network bandwidth. By spatial reuse we mean that specific portions of the ring network may reuse the same bandwidth. Essentially, the invention is based on the realization that each station rarely needs to communicate with all the other stations that share its ring. Under normal conditions, a station requires the ability to send messages only to a subgroup of the total number of stations. If the ring network is thought of as a circle and if the communicating stations are contiguous, then these stations occupy only an arc on the circle. Outside this arc, the bandwidth assigned to these stations is not in use. Accordingly, the bandwidth can be reassigned to stations which form other, non-overlapping arcs.

Abstract: A method is disclosed for managing the communication of data packets between a mobile data radio terminal and a plurality of fixed base stations in a data communications network. In a system of a plurality of base stations (each including a set of cellular telephone voice transceivers), wherein a first and second base stations are coupled through a home mobile data gateway (MDG) to the data communications network and a third base station is coupled through a server mobile data gateway to the data communications network, a method and apparatus which establishes in the home mobile data gateway, a forwarding address to a server mobile data gateway, sends forwarding information from the home mobile data gateway to the server mobile data gateway, and allocates with the server mobile data gateway, a new channel at the third base station, for the mobile data radio terminal.

Abstract: A method and system for generating a rules-based computer program employs an interactive workstation to display input data format screens. These screens are edited on the workstation to define the formats of all permissible items of input data to the program and the formats are stored. The workstation then displays rule entry screens in decision table format on which the program designer defines the rules relating to the processing of associated input data items. The completed decision tables are stored and together with the stored input data formats are compiled to produce computer program code for run-time execution or interpretation.

Abstract: A method and apparatus for the interactive modification of a production schedule using a graphical user interface. Each production task is represented by one or more primitive tasks. A primitive task is defined as that part of a production task for which the production amount or load remains unchanged. The primitive tasks are shown graphically on a display device as rectangular objects and are assigned attributes that are used to establish modification rules governing the way in which the graphical shapes change in response to the user's manipulation of the rectangular object representing one primitive task. A user can manipulate any primitive task through the graphical interface, and thereby modify the shapes of other primitive tasks which belong to the same production task in accordance with the modification rules.

Abstract: A multi-media network bus (MMN-Bus) and Controller (MMNC) serve as a distribution medium between a single high-speed network and multiple user network interfaces (UNIs). The MMNC allocates multiple isochronous channels and packet channels for distribution to the users over the MMN-Bus. The MMNC sends information in standard time segments or master cycles. Following transmission of a start signal, a series of control blocks is sent. The first type of control block is used for communication from the MMNC to each UNI. The second type of control block is used for transmissions from each UNI to the MMNC. Both types of control blocks are used for call setup, call acknowledgement, channel allocation, call end and packet data request. Following transmission of both sets of control blocks, a signal is sent to indicate the beginning of a sequence of isochronous transfers. Each UNI tracks the number of signals, identifying those signals which it should either receive or transmit.

Abstract: A method and system for call set-up and channel resource allocation between a plurality of parties in a multi-media conference call that consists of receiving notice that a Sender (the initiator of the conference) wants to commence a video conference. At that point, the number of channels required by the proposed call is determined. Subsequently, each requested participant is identified as available, busy, or no answer. If the receiver is available and sufficient channels are available, the call is set-up and the present members of the conference are notified of what channel each should transmit on and what channel(s) to listen to for the other participants. As each receiver is added, the Sender may conference with the present Receiver. If it is decided at this point, that the receiver will not participate in the conference, the call to that receiver may be terminated, and the procedure continues for other receivers until the full conference is set up. Individuals may leave the conference at any time.

Abstract: In manufacturing packaged modules for computer systems, many modules require sealing a lid to a lower substrate by using solder around the outer edges of the lower substrate. When the solder is heated, it often flows inward towards the chips on the substrates (solder intrusions) and may cause shorts. This invention solves the problem of solder intrusions by using a graphite lid seal fixture during lid seal. The fixture is comprised of lower and upper parts. The substrate, which has a sealband area, rests on and is held in place by the lower part of the fixture. The upper part of the fixture rests on the lower part. The lid, which has solder on its lower surface which is slightly smaller but otherwise generally matches the sealband area of the substrate, fits in a hole in the upper part and rests on the substrate.

Abstract: An improved method of testing in a distributed environment which is comprised of a Control Program residing in a Control Machine. The Control Machine also contains the central repository of information to control test execution in the Test Machines. The Control Program forwards instructions to a particular Test Program, residing in a Test Machine. The instructions are executed on that machine, and results are reported back to the Control Program. The Control Program verifies whether the results are correct. Depending on the results of the verification, the Control Program sends the test machine further instructions (to continue the test, stop the test, etc.). Logging the results of each test operation, keeping track of the tests performed, and coordinating the test cases are all performed on the Control Machine, by the Control Program.

Abstract: A method of scanning and detecting format code changes is disclosed. The method is preferably utilized in a data processing system which includes a central processing unit, a random access memory, a bulk storage device, a display device and a printer. The method scans a document for format controls, detects a format control, retrieves information about the format characteristics in effect before and after that control, organizes that format characteristic information in a menu and displays that menu to the user. This method minimizes the amount of time spent editing format controls in a large document by fully informing the user of the underlying format characteristics associated with each format control.

Abstract: A system and method for recognizing characters in cursive script is provided in which the script is scanned to detect word boundaries and words are then segmented into characters. This is accomplished by segmenting the script to form an initial portion, the segmentation being performed with reference to its position relative to a word boundary. This initial portion is then compared with a set of reference portions. Subsequent portions of the script are taken in sequence and compared with reference portions until a character is identified with an uncertainty less than a predetermined threshold value. A new initial portion is then segmented, with the new initial portion chosen on the basis of the average width of the character identified and the comparison process repeated to identify the next character.

Abstract: The disclosed system facilitates the user in electronically ordering items from suppliers. The system is comprised of an Electronic Catlogue and an Electronic Requisition facility. The Electronic Catalogue includes a Public Catalog and a Private Catalogue. The Public Catalog is stored on a publicly available database for access by customer/Requestors. The Private Catalogue is resident on a Customer's computer system and may contain unique pricing data based on pricing agreements. The Electronic Requisition facility is used by the Customer/Requestors to electronically create purchase requisitions based upon the information provided in the catalogues and route the requisitions through the appropriate approval process within he enterprise. Requisitions are then processed through the customer's procurement system and transmitted electronically as purchase orders to Suppliers.

Abstract: A method for storage and retrieval of both time-oriented versions and view-oriented versions of engineering change information in which the engineering change information progresses through a set of status conditions and access to the data by different user groups is conditioned upon the status of the information. Version control software logic enables users to create versioned objects by logical key grouping of data elements. The version control logic acts upon the logical keys and special versioned attributes of these objects for the proper specification and selection of object instances during creation, update or retrieval processing. Insert and extract sequence numbers are automatically generated for both historical preservation of previous engineering change information and efficient retrieval of the currently effective designs.

Abstract: A method and system for dynamically controlling the horizontal scrolling of data contained in a stored table of arbitrary size in a windowing environment. The list of data to be scrolled through is stored in memory with certain columns designated as sticky. Responsive to input from the user, a process is invoked to either scroll left or right through data contained in the stored list. A "must display" column is determined based on the current display in the window and the direction of scrolling. The sticky columns that can fit with the "must display" column are determined and placed on the display. Sticky columns that cannot be placed on the display with the "must display" column due to window size limitations are temporarily suppressed. Additional non-sticky columns are placed on the display on a space-available basis including the placement of a partial column that fills the display.

Abstract: A method and system for the definition and tracking of multiple unique product configurations for configuration management in which customer contracted views of the product are defined by applying temporary changes to the base product configuration. A product serial number effectivity is assigned to each unique customer specified product configuration and to the base product configuration in order to maintain multiple time-oriented versions of the product, thus supporting simultaneously both customer-specified and design-specified engineering changes to the base product. A product configuration identifier provides non-redundant data storage for each unique product configuration and can refer to noncontiguous ranges of product serial numbers. The configuration entry identifier in a product configuration table is used to peg detail component requirements to serially numbered products in order to correlate the product configurations as built with the product configuration as designed.