Abstract: A demultiplexing apparatus is disclosed which comprises a pulse generator (4), a phase control device (3), and an optical switching device (1), e.g., a NOLM. The phase control device (3) controls the pulse generator (4), whose pulse initiates switching operations in the optical switching device (1) depending on the magnitude of an electric signal derived from the output (8) of the optical switching device (1). The phase control device (3) includes a noise source (26) which disturbs the pulse in a defined manner.

Abstract: A cellular mobile radio system comprises cells divided into P sectors each having its own timebase. The system uses time division multiple access and accordingly transmission is effected in bursts each transmitted during a time slot, or channel time slot, of predetermined rank relative to a reference of the timebase of the sector concerned. N time slots form a frame which repeats periodically. A control channel is reserved for each sector for transmitting signalling and control data, a different control channel being associated with each sector. The timebase of each sector is shifted a predetermined number of time slots relative to the timebases of the other sectors of the same cell. The offsets are such that the control channels relating to each sector are conveyed at different times.

Abstract: A (BA) for picture and speech information.The video-telephone answering set (BA) stores only particular, essential single pictures, preferably in synchronism with the speech, in a storage means (HS). By suitable signalling, the caller can determine the single pictures to be stored. For the storage means (HS), a semiconductor memory is proposed.

Abstract: In digital transmission systems, adapted data units travel at a high throughput rate over large parts as a stream of fixed length data packets, called cells in this instance. The present method permits the measurement of characteristic magnitudes of the cell stream with respect to the respective individual connection to which they belong. It is based on storing from each arriving cell the identification data of the individual connection to which it is allocated, including its relative position in the cell stream. The memory is interrogated during a cell cycle time. A selection is made about which memory spaces within a defined area contain a predetermined identification data belonging to an individual connection. As a function of the selected memory spaces, the determined result, together with the magnitude and the position of the interrogated area, provides a decisive statement of the measured value.

Abstract: The present invention discloses a so called Discrete Multi Tone or DMT signal processor which is coupled to a transmitting end of a communication line and which reduces the need for the clipping of the amplitude of the output signal OS'. The signal processor includes the cascade connection of a vector transformation circuit VTC, a modulator MOD and a peak detection circuit PDC. The VTC circuit is provided with a plurality of input vectors IN whereon a vector transformation is performed according to a signal at a control terminal (CT) of the peak detection circuit PDC which is indicative of the fact that the amplitude of the output signal OS provided by the modulator MOD exceeds a predetermined threshold value.

Abstract: A data processing system is provided that consists of a connection of a first neural network (N.sub.1) with at least one other neural network (N.sub.21, N.sub.22, . . . , N.sub.2n). The first neural network (N.sub.1) and the at least one other neural network (N.sub.21, N.sub.22, . . . , N.sub.2n) is an associative memory. First input data (E.sub.0) are supplied to both the first neural network (N.sub.1) and also to at least the one other neural network (N.sub.21, N.sub.22, . . . , N.sub.2n) Data (E.sub.11, E.sub.12, . . . , E.sub.1n) which are evaluated by at least the one other neural network (N.sub.21, N.sub.22, . . . , N.sub.2n), are supplied as further input data (E.sub.1) to the first neural network (N.sub.1).

Abstract: Telephone calls passed from an operator service exchange (3) on to a local operator position (1) are received at the latter and information is issued to the caller. The principle components of the position (1) are (A) a connection (6) to the position desk controller (4) of the operator service exchange (3), (B) a speech connection from the position desk controller which terminates in a headset, and (C) a local personal computer (5) which, on the one hand, produces a display and controls the speech connection and, on the other hand, provides access to the required information. This position (1) is operated remotely, in that the local personal computer (5) is remote-controlled by a remote personal computer (8) through an ISDN telephone connection and in that, at the same time, the speech connection is extended from the position desk controller (4) through the local personal computer (5) and a telephone network to a remote operator position (2).

Abstract: The frequency of a first optical clock signal that is approximately sinusoidal and has a frequency f.sub.0 is doubled by all-optical means. A non-linear optical loop mirror (NOLM) is used as an optical modulator, being controlled by the first clock signal at f.sub.0 to modulate a continuous second signal introduced at the signal input of the NOLM. The modulated signal at the frequency 2f.sub.0 is recovered at the output of the NOLM. Two wavelengths are advantageously used for the first and second signals, these two wavelengths having substantially the same group velocity in the NOLM. In an important implementation the peak-to-peak power of the first control signal is much less than the power needed to maximize the switching effect of the NOLM on the second signal. The invention also concerns a soliton regenerator using this optical doubling method. Applications include fiber optic telecommunications, in particular using solitons.

Abstract: Speech recognition of Lombard-induced speech at a high rate of recognition is shown using a neural network (NN) utilizing nonlinear imaging characteristics. In a training phase, systematic parameter changes of Lombard-induced speech are trained to a neural network. In a speech recognition phase, imaging of Lombard-induced speech patterns to Lombard-free speech patterns takes place through the trained parameter changes.

Abstract: Recognition of speech with successive expansion of a reference vocabulary, can be used for automatic telephone dialing by voice input. Neural and conventional recognition methods are performed in parallel so that during training and configuration of the neural network, a conventional recognizer operating according to the dynamic programming principle has available newly added word patterns as references for immediate use in recognition. Upon completion of the training and configuration, the neural network takes over the recognition of the now expanded vocabulary.

Abstract: Low rate telemetry transmission between a transmission side and a receiving side is accomplished at the transmission side on digital information in a channel at a predetermined bit rate by selecting a predetermined frame thereof and outputting a selected one of at least two digital information elements desired for transmission with and in addition to those in the frame by means of a key generator which outputs a known key to the transmission and receiving sides and defined by a sequence of digital information that uniquely identifies the selected digital information elements by first processing the digital information element with the key in a first reversible mathematical manipulation to provide a first output and secondly performing a second reversible mathematical manipulation of the first output and the digital information in the frame to obtain an encoded digital information sequence which is then transmitted from the transmission side and received by the receiving side where it is decoded by reversing t

Abstract: The invention relates to an 8-PSK differential coder for trellis-coded modulation and to a corresponding pragmatic decoder, enabling 90.degree., 180.degree., and 270.degree. phase ambiguities to be lifted. The invention applies in particular to transmitting images at high rate by radio link.

Abstract: A software structure comprises structural objects each comprising at least one service access point to which a message addressed to the object may be sent. Each structural object is associated with other objects by relationships which relate it structurally relative to the other objects. Communication between objects takes place by transmission of messages between a service access point module of an originator object and a service access point module of a recipient object via a communication service object receiving messages to be sent from the originator objects and supplying them to the recipient objects. The communication service contains data defining the relationships between objects and refers thereto at the same time as to the recipient object identity in the header of a message in determining the identity of the structural object to which the message must be sent.

Abstract: The apparatus of the invention regenerates solitons in-line by synchronously modulating the solitons using a non-linear optical loop mirror as an optical modulator, the NOLM modulator being controlled by a clock signal recovered from the soliton signal by clock recovery means which may be all-optical means or electro-optical means. The NOLM includes an input coupler which may be a 50/50 coupler or an asymmetrical coupler. Said input coupler may be a 2x2 coupler or a 3x3 coupler. In a particular embodiment, the regenerating system further includes a plurality of optical amplifiers and a plurality of "guiding" filters. Application to long-distance soliton optical telecommunications.

Abstract: The optical amplifier combiner arrangement is used in a tree-like optical network (APON) comprising a cascade connection of dedicated branches, of the arrangement and of a common branch. The arrangement is coupled between a plurality of optical network users and an optical line terminator, via the dedicated branches and the common branch respectively. The network enables upstream transmission of information signals from the optical network users to the optical line terminator. The optical amplifier combiner arrangement includes for each branch of the dedicated branches (1) an optical amplifier to amplify an information signal with a gain value and to thereby generate an amplified information signal with a predetermined power level, (2) an optical on/off switch coupled between the optical amplifier and an optical combiner to pass the amplified information signal when the information signal is present and to interrupt the branch when the information signal is not present.

Abstract: To allocate time slots to terminal stations (T1 . . . Ti . . . Tn) for transmission of upstream burst (UB) in a network system wherein a central station (CS) is coupled to a plurality of terminal stations (T1 . . . Ti . . . Tn), the central station (CS) downstream transmits access grant information (AGI) which forms part of downstream frames (DE). Each terminal station (Ti) is equipped with a cyclic local grant counter (LGC) which generates a local grant counter value between zero and a predetermined limit, and which is synchronized with a cyclic master grant counter (MGC) included in the central station (CS). The upstream time slots are bounded by two zero crossings of the cyclic local grant counter (LGC) and a terminal station (Ti) is allowed to transfer an upstream burst (UB) in such a time slot provided that this time slot is allocated to the terminal station (Ti) via previously received access grant information (AGI).

Abstract: In a system for transmitting information via a time-varying transmission channel, learning information is transmitted in addition to wanted information to enable the transmission channel to be estimated at the receiving end. The system estimates if it is necessary to transmit learning information given the current variations of the transmission channel and transmits learning information only if this is deemed necessary.

Abstract: LAN's are customarily used for pure data communication. Voice communication via a LAN is only possible with the constraint of other subscribers being subjected to long waiting periods during the relatively long occupation time. The present LAN expands the functionality of work stations connected to the LAN, by using ATM technology to meet the communication requirements of an office environment. A network node or a conventional work station (3) is expanded by means of a unit which performs switching technology functions, so that terminals of the most different types (4, 5, 6, 7) can access the LAN via interface circuits (41, 51, 61, 71), resulting in a private network with distributed telecommunication installations. In this way, real-time connections with different bandwidth requirements, such as for example video conferences and pure voice transmission, are made possible. The thus expanded ATM-LAN is suitable for all types of communications taking place in an office.

Abstract: A method of encoding/interleaving a source sequence of data symbols to be transmitted to produce at least two bursts to be sent made up of said data symbols to be transmitted and redundancy symbols writes the data symbols to be transmitted into a base matrix. It stores the base matrix in a first quadrant of a send extended matrix. It carries out first and second encoding steps using the columns and rows, respectively, of the base matrix, generating first and second redundancy symbols which are stored in second and third quadrants of the send extended matrix and constitute additional rows of the base matrix. It reads the send extended matrix row by row (or column by column), each row (or column) of the send extended matrix constituting a burst to be sent.

Abstract: The invention concerns a method for the providing of messages for subscriber information for services in a communication network, as well as a service control facility (SCP1), a service support system (IP1), a service access arrangement, a service switching point and a service system for the execution of this method. The messages for subscriber information, for instance, voice announcements or speech texts, are managed centrally from the service control facility (SCP1). Only by need are the messages for subscriber information loaded into the service support system (IP1) from the service control facility (SCP1, SCP2). One of the messages loaded into the service support system (IP1) is only sent to a subscriber of the communication network subsequent to a transmitted demand message from the service control facility (SCP1) to the service support system (IP1).