Abstract: Disclosed is a modem including a transmitter having a convolutional encoder for transforming each time unit digital data into an expanded bit sequence having symbol-selecting bits and subset-selecting bits, the transmitter further providing modulation of a carrier signal and a receiver wherein a Viterbi decoder determines the maximum likelihood path through a trellis from a plurality of surviving paths using a wraparound memory wherein in each addressable location indexed to a current state of a surviving path there is stored a previous state of that surviving path.

Abstract: A modem receiver having a timing recovery circuit is disclosed. The timing recovery circuit includes circuitry for extracting timing information, multiplication circuitry responsive to the timing information for providing a clock envelope signal and processing circuitry for controlling the gain of the clock envelope signal as a function of the average power of the clock envelope signal.

Abstract: A method of encoding data into a frame, includes the step of providing a frame having a plurality of byte locations for storing a plurality of bytes each associated with one of the byte locations. The value of the initial byte location of the frame is stored in a temporary register. An address of a first byte location having zero value is stored in the initial byte location of the frame. A continuation bit associated with the address of said first byte location is set. The value of the initial byte is stored in the first byte location. The non-zero valued bytes of information are stored in their associated byte locations of the frame. When the initial byte location contains zero, the last location containing an address of a zero valued byte repeats the previous address having a zero valued byte.

Abstract: In a method and apparatus for transmitting data over either a DDS-I (74) or DDS-II (DDS-SC) network (34), a primary and secondary channel are accommodated preferably for providing both primary data capability and diagnostics for the network. Primary and secondary channel data are multiplexed with control information in a standard DDS-SC frame. The system timing circuitry is locked to the DDS data rate and means for providing encoding and decoding of bipolar violations is provided so that the digital service unit (DSU) is compatible with both types of digital data service. At power-up, a microcomputer (102) reads the DSU clock frequency and the voltage offset of operational amplifiers and provides adjustment through digital to analog converters (210,294).

Abstract: A method of providing B channel diagnostics to an integrated services digital network (ISDN) comprises the steps of monitoring error performance on the ISDN D channel and comparing the D channel error performance to a predetermined criterion as a measure of the B channel error performance. If the predetermined threshold is exceeded any of a number of possible actions can be taken by a network diagnostics controller including various loopback tests, alerts or network reconfiguration.

Abstract: This invention relates to an efficient CSMA/CD-based protocol for a local area network using an unidirectional global bus architecture, in which a ready station captures the bus for one packet time only if no other stations to its left transmit along with it. A modified version of the protocol provides a fairer access to the bus. This new scheme is a variable priority structure in which each station has a priority that changes dynamically with the state of the system. This is a contention scheme that behaves like a generalized and very flexible reservation protocol. A performance analysis of both versions of the protocol results in channel efficiency and transmission delay figures that are superior to those of the leading bus protocols.

Abstract: A method and apparatus for compression of data includes first encoding data characters according to a relative frequency of use table for each character to produce a frequency code. This frequency code is then compressed using a modified Huffman code. The code is modified by restricting the maximum word length to a predetermined number of bits. Different types of text can be encoded according to different frequency of use tables to maximize compression efficiency for various types of source data with minimum data processing.

Abstract: Stochastic Time Division Multiplexing (STOTDM) is disclosed. in STOTDM data packets are preceded by a flag word. Each data packet appears in a frame of data packets in a predetermined order. If there is no data to be placed in the packet location, only the flag is transmitted and the data location is allocated to other data. By using a fixed length frame, synchronous data packets can be placed near the beginning of the frame and packets which are not timing sensitive may follow them. A variable length frame may also be used wherein data from a plurality of sources are once again ordered and preceded by a flag word or bit. The flag may be used to indicate whether or not the packet location is filled or this may be determined by control packets. Variable length frames may be accommodated as long as the average bandwidth required to transmit the data does not exceed the channel bandwidth.

Abstract: An apparatus for cancellation of sinusoidal varying phase jitter in a data modem is implemented in firmware using microprocessor technology. An estimate of the frequency and phase of the phase jitter is computed in a first stage. Substantially independently, an estimate of the amplitude of the phase jitter is computed in another stage. These estimates are combined to form a composite estimate of the phase jitter which is utilized to cancel out the sinusoidal phase jitter in the demodulator. Digital phase locked loop technology (DPLL) is utilized to lock onto the phase jitter components. The capture range of the phase locked loop is dynamically altered during a training sequence to allow for capturing a wide range of jitter frequencies. During the training sequence the damping factor of the loop is gradually altered thereby substantially reducing the capture range and response time of the loop once jitter acquisition has occurred.

Abstract: A noninterruptive noise measurement technique wherein the absolute value of the amplitude error from a modem slicer is compared to a reference, the number of events in excess of said reference over a selected interval are counted and the range within which said count falls is determined as a particular signal to noise ratio value.

Abstract: A wireless data communications network utilizes a frequency synthesizer to achieve a systematic frequency hopping arrangement wherein frequency is incremented by two times the channel spacing from lowest frequency to highest frequency. When adjusting the frequency from highest frequency to lowest frequency a one times the channel spacing frequency hop is utilized for the first and half and last hop and two times the frequency spacing is utilized otherwise. In this manner, alternating frequencies are selected so that the same frequency is only utilized once during any complete cycle and the size of frequency transitions is held to a minimum allowing a narrow bandwidth filter and inexpensive implementation of the phase locked loop portion of the synthesizer.

Abstract: An all digital video synchronization technique wherein a microprocessor tracks the incoming video signal and synchronizes its operation by stopping at a certain point in the sync signal and restarting upon occurrence of a horizontal sync pulse.

Abstract: A digital voice communication apparatus having improved echo suppression includes a transmission gain detection circuit for detecting a near talker signal and a receiver gain detection circuit for detecting a far talker signal. An echo detecting circuit detects which of the near talker and far talker signals has a lesser value and suppresses the gain of the one detected to have the lesser value. A far talker suppressing circuit detects whether the most recent peak gain in this near talker signal is greater than a threshold and is greater than the gain of the far talker signal. If this condition exists the far talker signal is suppressed. A near talker signal suppressing circuit detects whether the most recent gain of the far talker signal is greater than a threshold and is greater than the gain of the near talker signal. If so, the gain of the near talker signal is suppressed.

Abstract: A secure snap together univeral housing arrangement according to the present invention includes identical upper and lower housing members having latch receiving apertures on the upper and lower surfaces. A plurality of circuit cardguides are disposed within the housing and include latching members which engage the latch receiving apertures. The circuit cardguides are preferably also identical and are used to fix various circuit cards in place. Front and rear panels snap into place using resilient latches also and include ledges along the upper and lower edges which engage lips on the upper and lower housing members to provide for secure engagement of the upper and lower housing members together. Assembly may be accomplished without use of any specialized tools and disassembly requires the use of a simple tool to pry the front panel out of engagement with the remainder of the housing.

Abstract: In a method and apparatus for processing color video signals, a switching network controlled by a control computer initially applies a composite video signal to a single analog to digital converter for storage in a memory. The composite signal is then repeatedly retrieved from the memory in order to be sequentially decomposed into required color characteristics such as red, blue and green. The decomposed signal is also applied to the input of the switching network so that the control computer can alternately multiplex each of the three decomposed signals into the memory. In this manner, the composite red, blue and green information is sequentially stored in memory for manipulation by a signal processing computer. This eliminates the need for multiple analog to digital converters operating in parallel.

Abstract: A method and apparatus for compression of data includes first encoding data characters according to a relative frequency of use table for each character to produce a frequency code. This frequency code is then compressed using a modified Huffman code. The code is modified by restricting the maximum word length to a predetermined number of bits. Different types of text can be encoded according to different frequency of use tables to maximize compression efficiency for various types of source data with minimum data processing.

Abstract: Apparatus employing microprocessor control for reconfiguring the strapping status of local and remote data modems automatically through a front panel display built into the modem.

Abstract: A circuit arrangement is provided for switching serial data packets through a network destined for one of a plurality of possible outgoing lines. Minimal delay is achieved by shifting the data through a shift register having length equivalent to the destination address of the incoming serial data packet. The shift register addresses a memory which in turns controls a switch network so that the incoming packet is switched with minimal delay to an appropriate outbound line. By utilizing random access memory to translate from destination address to switch position, the system may be altered to correct for changes in the overall network caused by network failures or expansion network or to allow dynamic load balancing by directing data through the switch to a control computer which in turn rewrites the memory.

Abstract: Apparatus for controlling RTS/DCD simulation in a data receiver by detecting, after the received data transmitted for assignment to a particular data terminal, the presence of a two-part code having a repetitive stream of binary 1's or 0's (or repeating sequences thereof) as the first part; and for thereafter detecting the presence of a preselected coded sequence of bits within a preselected number of succeeding bits received as the second part of said two-part code. The preselected two-part code is advantageously chosen to be one whose probability of occurrence in customer data is extremely remote.

Abstract: An apparatus for use in transmitting compressed digital voice data includes an analyzer for compressing digitized voice data according to an algorithm which produces a root mean square of the voice data. An analog to digital converter is used to convert analog voice signals to digital data as an input to the analyzer. An AGC circuit controls to analog voice signal to keep it within predetermined limits for digitization, with the AGC setting being determined by the root mean square of the digitized data. The digitized data includes data representative of the root mean square adjusted by the inverse of the change induced by the AGC. A speech synthesizer expands the digitized data at a remote location to convert it back to analog speech signals.