Abstract: A method of imaging an integrated circuit using a focused ion beam system is presented. According to the method an integrated circuit is imaged in plan-view using a focused ion beam system. Circuit information is then extracted absent processing. In another embodiment, a method and system for imaging an entire IC automatically without removing the IC from the imaging system and requiring minimal operator intervention is presented. The method employs a focused ion beam system to image an exposed layer of an integrated circuit and then to etch a portion of the exposed layer in situ. Imaging and etching are repeated until substantially the entire integrated circuit is imaged. A processor is used to assemble the layers into a three-dimensional topography of the integrated circuit. Because of known relationships between layers, the mosaicing is facilitated and the final topography is more reliable than those produced by currently known computer implemented methods.

Abstract: The invention comprises the insertion of anchors into suitably located wall cavities, each anchor being suspended by cables extending up through the wall top and into the roof space; then making fast each anchor by the injection into the cavity above the anchor of a hardening, adhering liquid “plug” of sufficient size and strength, pressing down upon the anchor and strongly adhering to the confining internal surfaces of the cavity. Especially in cases where there are no wall cavities, or none readily accessible, the invention encompasses anchoring the roof to or through the exterior surfaces of suitably located walls by means of long vertical straps; these extend up through the ceiling sufficiently into the roof space, taking the place of the cavity anchors and cables.

Abstract: A method of resource management at computer controlled telephony hardware outside the Public Switched Telephone Network (PSTN) operated by a service provider for establishing a call between a service subscriber and another party by facilitating a PSTN voice connection between the service subscriber and the computer controlled telephony hardware, and a PSTN voice connection between the other party and the computer controlled telephony hardware involves determining a resource management index associated with the service subscriber, determining a call treatment using the resource management index, and applying the call treatment to the call.

Abstract: A flexible partition suspended from a ceiling of a gymnasium or the like, includes a drive cylinder which extends the length of and intermediate of the curtain forming the partition. A cylinder is provided with an internal drive motor, and a reaction device is provided externally of the curtain such that the motor device when activated will drive the cylinder to wind-up the upper and lower portions of the curtain. In one embodiment, the upper portion is a single curtain panel and is attached to the periphery of the cylinder at an axial line. The lower portion includes two spaced curtain panels which are connected to the periphery of the cylinder at diametrically opposite lines thereon. In another embodiment both the upper portion and lower portion of the curtain include two curtain panels, so as to form a full-size double curtain. In such an arrangement, the flexible partition has a spaced-apart double panel configuration at least at its lower portion, to provide a better acoustic insulation.

Abstract: A remote traffic data acquisition and intelligent vehicle highway system for highway vehicles is provided. In-vehicle devices compute time-related vehicle locations on a digitized road network map using information received from a global position system (GPS) and transmit the time-related vehicle locations to a traffic service center. The traffic service center collects the data from all equipped vehicles that travel the roadway system in an area within range, processes the data and provide a real-time traffic forecasts. The in-vehicle devices receive the digitized road network map as well as the real-time traffic forecasts and provide route guidance and related services for the drivers using the traffic forecast information. The traffic forecast is based on projections from normal traffic conditions retrieved from archived data adjusted by factors related to real-time situations. The system provides a practical and economic solution for an intelligent highway vehicle system.

Abstract: A method and apparatus for providing voice communications between two parties using computer controlled telephony hardware which is separate from the PSTN is described. The voice communications may be voice over Internet or PSTN voice connections, or any combination of the two. The apparatus includes a WEB server, a computer telephony server, a Voice over IP gateway and an operations, administration and maintenance server. The apparatus is suitably installed and operated by a service provider such as an Internet Service Provider. The advantage is unexcelled control over call setup, transfer and conference functions. A further advantage is the rapid, unfettered development of new services without compatibility issues with the PSTN.

Abstract: A variable power supply technique for use with a high efficiency line driver uses a signal peak amplitude to determine a start of a rise time for a power supply to begin supplying operating voltage to the line driver. This ensures that the voltage supplied to the line driver essentially tracks the output signal. A slew start delay circuit detects when the signal rises above a level that can be supported by a current power supply, and determines a variable delay for switching on a power supply to supply a higher voltage to the line driver. The advantages include reduced power usage, less heat dissipation, and the ability to select a primary power supply that outputs a lower voltage.

Abstract: An automatic gain control method and apparatus for modem receivers. The automatic gain control circuit includes a programmable loop gain for scaling a digital signal Y to a first prescribed level during a first mode of operation and to a second prescribed level during a second mode of operation; and filters and converters for converting the scaled signal Y into an analog gain control signal for input to the analog AGC. The gain control circuit and method of operation provides control over the parameters of the programmable loop gain such that during start-up initialization the signal Y is scaled to the first prescribed value and during steady state operation the signal Y is scaled to the second prescribed value. This is accomplished by changing the gain parameter of the programmable loop gain.

Abstract: A system of energy detection for detecting a valid signal in a modem to enable power savings in receiver and transmitter circuits. The method of energy detection includes the steps of: (a) processing a prescribed number of signal samples (e.g. typically between 512 and 2048); (b) determining an average energy level for the prescribed number of signal samples; (c) determining a gain of an amplifier based on the processed signal samples; (d) comparing the average energy level with a prescribed energy level threshold; and (e) comparing the gain of the amplifier with a prescribed amplifier gain threshold. If the average energy level is greater than the prescribed energy level threshold and if the gain of the amplifier is less than the prescribed amplifier gain threshold then a valid signal is declared and the transmitter of the modem is activated.

Abstract: A method and apparatus for timing recovery in modem receivers. The timing recovery circuit includes a voltage controlled oscillator for controlling the voltage controlled sampling frequency of an analog-to-digital converter. The oscillator generates a timing clock that is dependent on an average phase error signal calculated from Nyquist signals of the input signal. A phase detector circuit is used for generating an instantaneous phase error signal of the in-phase and quadrature-phase signals. A digital loop filter receives the instantaneous phase error signal over time to generate the average phase error signal. The average phase error signal is conditioned further (after conversion to analog) by an analog loop filter such that the average phase error signal adjusts the timing clock generated by the oscillator. The low pass filter provides control of the acquisition and steady state operations by changing the gain and pole parameters of the filter.

Abstract: A method and apparatus for pre-establishing and managing Switched Virtual Circuits (SVCs) and Switch Cross-Connects (XCs) to support rapid setup of real-time services in Asynchronous Transfer Mode (ATM) networks are described. SVCs and XCs are pre-established and cached in accordance with communications application-specific rules and maintenance algorithms for each destination and service characteristics in the backbone network. The intelligent management of cached SVCs and XCs permits the backbone network to meet the tight delay budget of delay-sensitive communication applications. With proper management of cached resources, connection setup can be rapidly effected even during periods of high demand. The advantage is a connection setup mechanism having adequate granularity to ensure both efficient network usage and rapid response to service requests.

Abstract: A method of timing recovery convergence monitoring in modems using an average phase error signal. The method involves continuously tracking the peak of the average phase error. The average phase error is compared to a dynamic threshold (i.e. a threshold that can change over time based on changes to the detected peak at a given time). Convergence is declared when the average phase error remains less than the threshold over a given length of time (i.e. after processing a prescribed number of consecutive samples).

Abstract: A method and apparatus for adjusting the amplitude of a received discrete multi-tone analog signal contaminated by radio frequency interference from radio stations in the AM band are described. The amplitude is adjusted by controlling an analog gain controller in order to utilize substantially all of a dynamic input range of an analog-to-digital converter to reduce the quantization noise of the output digital signal. Data discrimination is truly improved and the capacity of usable discrete multi-tone channels is increased.

Abstract: A forwarding table data structure and a memory optimization process that involves variable multi-stage lookups. The data structure for an address translation system includes, a plurality of blocks, each of the plurality of blocks includes a contiguous set of data records; each data record includes a pointer and an action indicator; if the action indicator is equal to a prescribed value the pointer represents an address of a translation target in the data structure; and if the action indicator is not equal to the prescribed value the pointer represents an address of a selected one of the plurality of blocks in the data structure. The data structure is built based on the input data set (e.g. IP routing table) to be memory optimized. The number of stages can be flexibly determined based on memory restrictions, input data set etc. Since the data structure is essentially “customized” optimal routing speeds can be obtained.

Abstract: A method and apparatus for filtering interference (e.g. transmitter feed-through and RF) in a modem receiver. The method of reducing transmitter feed-through (Tx-FT) interference at a receiver of a modem to obtain a desired receive (Rx) signal having a spectrum at a carrier frequency for input into a QAM demodulator includes attenuating the Tx-FT interference and passing the Rx signal at the carrier frequency using an analog filter. The output signal from the analog filter is converted to a digital representation by a analog-to-digital (ADC) converter. A digital band pass filter is used to pass the Rx signal at the carrier frequency and to attenuate the Tx-FT interference further. A mixer is then used to shift the carrier frequency of the spectrum of the Rx signal to an intermediate frequency. The mixer introduces at least one spectral transmitter feed-through image Tx-FT(I) and a receive signal image Rx(I).

Abstract: A method and apparatus for transferring data associated with synchronous transfer mode (STM) calls through an asynchronous transfer mode (ATM) network are described. The apparatus includes a voice interface control unit, a signaling interface and a TDM peripheral device associated with each edge node in the ATM network. The voice interface control unit controls the ingress and egress of STM calls to the network. The signaling interface sends and receives common channel signaling messages associated with calls originating in the STM network and passes message content to the voice interface control unit. The TDM peripheral device converts STM network and passes message content to the voice interface control unit. The TDM peripheral device converts STM calls to ATM cells, and vice versa.

Abstract: An apparatus and a method for completing long distance telephone calls over the Public Switched Telephone Network (PSTN) using an automated Internet Operator are disclosed. The apparatus includes a long distance service provider node connected to the Internet and a call connection control computer connected to the Internet and interfaced to a toll switch connected to the PSTN. The method permits a calling party to request the automated Internet Operator to complete long distance call connections by connecting to the long distance service provider node and identifying himself. The caller completes a call request form which is uploaded to the long distance service provider node and passed to the call connection control computer for processing. The call connection control computer instructs the toll switch to dial the calling party, dial the called party and conference the two lines.

Abstract: A forwarding table data structure and a memory optimization process that involves variable multi-stage lookups. The data structure for an address translation system includes, a plurality of blocks, each of the plurality of blocks includes a contiguous set of data records; each data record includes a pointer and an action indicator; if the action indicator is equal to a prescribed value the pointer represents an address of a translation target in the data structure; and if the action indicator is not equal to the prescribed value the pointer represents an address of a selected one of the plurality of blocks in the data structure. The data structure is built based on the input data set (e.g. IP routing table) to be memory optimized. The number of stages can be flexibly determined based on memory restrictions, input data set etc. Since the data structure is essentially "customized" optimal routing speeds can be obtained.

Abstract: The invention provides a method and apparatus for scaling transmitter power to a prescribed level (e.g. at or below the maximum average transmitter power supported by the communication system) for a full range of QAM index levels. In particular, in an n-level quadrature amplitude modulation (QAM) communication system having a digital-to-analog converter and an n-QAM signal having an average transmitter power, for transmission on a communication channel, the method of adjusting the average transmitter power of the signal includes selecting a reference QAM index from a prescribed set of QAM index levels (e.g. 4, 16, 64, 256) where the composite signal is processed with a maximum average transmitter power supported by the communication system. The next step involves generating a power adjustment coefficient for each of the QAM index levels from the prescribed set of QAM index levels based on the reference QAM index level.

Abstract: A method of controlling asymmetrical port access in a common-memory (CM) multi-port data packet switch and a high-capacity flexible access rate CM data packet switch are described. The method involves providing a common-memory-access control table which defines a unique contiguous band of CM access time slots allocated to each port of the packet switch. Each access time slot is identified by a time slot identifier. Time slot identifiers are generated at a regular interval. The time slot identifiers are generated so that CM access time for each port is substantially equally spaced to minimize delay jitter. The packet switch may have an equal number of input and output ports, or more output ports than input ports. In the high-capacity CM packet switch, several CM switch modules are interlinked by middle buffers in a folded configuration to provide a packet switch having Terabit output capacity.