Abstract: This invention concerns a method for connecting optical fibres, comprising stages consisting in (i) effecting a controlled rough polishing of the interface (5) of the fibres (1) to be connected, (ii) steeping the ends of the fibres (1) in a concentrated solution hydrofluoric acid and (iii) aligning then connecting the fibres (1) using an index matching material (FIG. 4). The invention also concerns a variable optical attenuator and a system using this attenuator to measure the effect of the reflection coefficient on an optical line.

Abstract: A binary calculation circuit has a logic operator acting as an exclusive-OR gate generating a first intermediate signal which is an exclusive-OR of a first input and a carry-in input. An inverter generates a second intermediate signal. A second logic operator generates a first output bit which is a symmetrical exclusive-OR of a second input and both the first and second intermediate signals. A second output bit is a symmetrical trigger function of the first and second input, depending on the first and second intermediate signals, and is generated in a transmission gate. Since the carry-in signal passed via the first and second intermediate signals is applied directly to transistors of the transmission gate, carry propagation delay is reduced.

Abstract: A device for the separation and alignment of objects of the packages, parcels or similar kind, including a conveyor for the parcels brought in bulk to the inlet and for discharging the same parcels separated and aligned at the outlet, which includes at least one truncated cone opened at its ends, driven continuously in rotation about its axis which is inclined with respect to the horizontal and which includes in its inner surface a set of helical wings, axially offset with respect to one another and extending when taken together over at least one complete turn of the truncated cone, the parcels introduced at the upper end of smaller diameter at the inlet of the truncated cone being discharged at its lower end of larger diameter at its outlet.

Abstract: The invention relates to a quantization process in an encoding by transformation for transmission on a low rate link of an image signal, said image signal being processed in the form of blocks, each containing a set of digital data, each block being transformed to produce a transformed block of coefficients (Ci) representing the frequency components of the block, each transformed coefficients of a block being quantified during a scan by comparison with quantization levels, the first being called the threshold, the process being characterized in that the quantization is performed with a threshold (T) able to vary between one transfored coefficient (Ci) and the next (Ci+1) in the scan, depending on whether the transformed coefficient does or does not fulfil the conditions -T<Ci<+T.

Abstract: A method of generating a quasi-complex hologram using incoherent light and a conoscope system comprising a birefringent crystal (30) inserted between two circular polarizers (10, 20), the method comprising the following steps: (i) recording a plurality of different conoscopic holograms of the same object without relative displacement thereof and with respective different polarization configurations; and (ii) combining the various conoscopic holograms obtained in this way in order to remove the bias and the conjugate image. The invention also relates to a device for implementing the method.

Abstract: An information encryption method particularly intended for a communication system in which bidirectional multiplexed communications of the point-to-multipoint type are derived between a switching exchange and plural terminal stations. The switching exchange transmits first multiplexed words respectively destined for the terminal stations. The terminal stations supply second words which are multiplexed and transmitted towards the switching exchange. In the switching exchange, informations concerning transmission errors detected in the second words are derived and transmitted towards the terminal stations, and the first words are encrypted as a function of second words previously received free of transmission error in the switching exchange. In each of the terminal stations, the informations concerning the transmission errors are detected, and the second words transmitted from the terminal station and received free of transmission error by the switching exchange are memorized.

Abstract: A conventional demodulator for digital transmission includes means (12) for splitting the received HF signal into two identical auxiliary signals, mixers (20, 30) for transposing the auxiliary signals into in-phase and quadrature baseband components by multiplying them with in-phase and quadrature output signals from a local oscillator (14), and respective filter and amplifier means (22, 24; 32, 34) and analog-to-digital converters (26, 36) on the in-phase and quadrature channels. In addition, the demodulator includes estimation means (100) suitable for estimating faults on groups of samples (Y.sub.p,k, Y.sub.q,k) taken from the outputs of the analog-to-digital converters (26, 36), with the faults being estimated in the form of five parameters, and correction means (200) for correcting current samples (Y.sub.p, Y.sub.q) on the basis of the parameters estimated on an earlier group of samples.

Abstract: A wideband server is defined by a main processor (10) communicating over a DMA channel (15) with a hard disk (16) and also with outlet cards (2-1 to 2-16). Each outlet card has its own processor which controls alternating mode to-processor access to two buffer memories A and B. While one of the buffer memories is delivering musical data to a user, the other is being filled, and vice versa.

Abstract: The radio system embodying the invention comprises a central station and a terminal stations or passive end. Data signals are transmitted between data processing units which are respectively connected to the central and terminal stations. The radio system is essentially characterized in that the terminal station comprises a square two-dimensional antenna and a microwave switcher so as to phase modulate, by means of a data signal, a microwave transmitted from the terminal station to the central station. This phase modulated microwave is produced from an incident microwave transmitted by the central station and issued to the terminal station by the square two-dimensional antenna. The design of the terminal station is very simple and compact and can be included, as well as the corresponding data processing unit, in a contact-free chip card.

Abstract: An echo cancelling device for use between a line receiving an incoming signal and a line transmitting an outgoing signal, for cancelling out echo, comprising a plurality of processing channels connected in parallel relation and assigned to successive mutually adjacent sub-bands of the spectral band of the outgoing signal, each channel having: a first analysis band-pass filter receiving the echo-affected signal to be transmitted, whose output is connected to the additive input of a subtractor; a second analysis band-pass filter, identical to the first filter, receiving the incoming signal and feeding an adaptive filter delivering an estimated echo value in the sub-band to the subtractive input of the subtractor; and a synthesis filter, symmetrical with the analysis filters and whose output feeds the transmission line.

Abstract: A double heterostructure stack comprising confinement layers (CC) enclosing active layers (CA) is formed on a substrate (S), e.g. a N+ type gallium arsenide substrate. Selective etching is performed so as to lay bare the confinement layers at different depths, e.g. CC4, CC3, CC2. The confinement layers initially receive contact layers (CP) made of P-type gallium arsenide. Regions R1, R2, and R3 are formed through the contact layers to constitute junctions with the uppermost active layers (respectively CA1, CA2, CA3). Valleys (V10, V21, V32) are formed to isolate the above-defined elementary stacks. After a metal contact layer has been formed, and after the end surfaces have been optically prepared, a multi-wavelength laser device is obtained.

Abstract: For filtering a bi-dimensional image signal in the form of data representing pixels obtained by column per column scanning of bands each having a height equal to a fraction of the height of the image, the transform of each pixel is computed in turn. For that, a convolution product using MxN coefficients is carried out on the pixels of a zone of an image, each column of each M pixels. The image is fractionated into mutually adjacent blocks each representing N columns of M pixels and the transforms of all pixels of a same block are simultaneously computed in MxN cycles by: parallel computation, during a same cycle, of all partial convolution products of only one of the filtering coefficients and of those pixels which provide a partial product which intervenes in the computation of the transforms of all pixels of the block, within a "window" of the picture which contains the block. The cycle is repeated for each coefficient in turn. All partial products obtained during the successive cycles are summed.

Abstract: A first switching network embodying the invention is a subscriber installation having a chain structure. In the network, the switching is distributed over a plurality of switches each connecting a terminal to the network. A second embodiment is a business network having a star-chain structure and comprising a plurality of sub-networks of chain type. The sub-networks are star connected to central switches which operate in a load sharing mode. According to the method embodying the invention, the terminals connected to the network assume themselves the responsibility for determining communication routes through the network. During a communication set-up phase, caller and callee terminals transmit call packets which are disseminated to all terminals in order to measure two so-called "distances" respectively between the caller and the callee and between the callee and the caller. These distances are after included in data packets for routing said packets.

Abstract: A differential encoder for speech encoding, typically in a 16 kHz wide band, has a self adaptive predictive filter, comprising a comparator receiving, on an additive input, successive samples of a signal to be encoded and, on a subtractive input, an estimated signal delivered by a prediction loop which receives the error signal from the output of the comparator, the prediction loop comprising a self-adaptive predictive filter, a gain control circuit and a multiplier which receives the outputs of the predictive filter and of the gain control circuit and whose output is connected to the subtractive input.

Abstract: A method of fabrication of a p+nn+ diode and a bipolar transistor which incorporates this diode, uitilizing the effect of neutralization of donor atoms by atomic hydrogen consists in forming a layer of Group III-Group V material of n+ type on a substrate of Group III-Group V material, locally implanting acceptor ions with a view to forming, in the surface, the p+ region of the diode adjacent to an intermediary n+ region forming two metal contacts at the surface of the layer of Group III-Group V material situated in proximity to one another and with respect to the p+ and n+ regions of the diode, the portion of the intermediary n+ region contiguous with the p+ region being without of metal contact, and hydrogenating the structure, to form in that portion an n region, with the other portion of the intermediary region being provided with a metal contact being constituting the n+ region of the p+nn+ diode.

Abstract: The coding and decoding devices embodying the invention are connected by two paths, a main path and an auxiliary path, of a variable rate digital transmission network. When the main path is capable of dispatching the totality of the information rate that the coding device is able to issue, only a main coder and a main decoder, respectively of the coding and decoding devices, are active. When the main path is not capable of dispatching the totality of the picture information, an auxiliary coder and an auxiliary decoder, respectively of the coding and decoding devices, are also active and the auxiliary path dispatches the part of the information rate that cannot be dispatched by the main path. A decoder and a subtracter are provided in the coding device so as to estimate the part of information not transmitted by the main path.

Abstract: The circuit embodying the invention comprises a pointer memory memorizing J=16 decremented write pointers associated with J=16 buffer queues. A channelling word supplied conjointly to incoming data and indicating the buffer queue in which this data is to be written, addresses the corresponding decremented write pointer. The write pointer and the channelling word form a write address for a buffer memory used as a medium for the J=16 queues. A comparator detects equality between the write pointer and the read pointer and in this case inhibits the acceptance of the incoming data which would overlay data not yet read. Logic means associated with the comparator control the updating, by possible incrementation, of the write pointer values contained in the pointer memory, after each data read in the buffer queues. This circuit is particularly well suited for use in an asynchronous self-channelling packet time-division switching system.

Abstract: A device for real time computing of a sum of terms having the form .SIGMAiT).qe((n-i)T) is suitable for use in an echo canceller for a data transmission system using code 2B 1Q. The device first transforms each of the successive estimated values of qe into transformed values qde which are powers of two, by adding or subtracting 1. The successive elementary products each obtained by a shift of h by an amount selected in response to the respective value of qde. The elementary products are summed in a register.

Abstract: A double heterostructure stack comprising confinement layers (CC) enclosing active layers (CA) is formed on a substrate (S), e.g. a N+ type gallium arsenide substrate. Selective etching is performed so as to lay bare the confinement layers at different depths, e.g. CC4, CC3, CC2. The confinement layers initially receive contact layers (CP) made of P-type gallium arsenide. Regions R1, R2, and R3 are formed through the contact layers to constitute junctions with the uppermost active layers (respectively CA1, CA2, CA3). Valleys (V10, V21, V32) are formed to isolate the above-defined elementary stacks. After a metal contact layer has been formed, and after the end surfaces have been optically prepared, a multi-wavelength laser device is obtained.

Abstract: A two-port switched capacitor filter network of the type comprising two input terminals (4, 6), two output terminals (8, 10) and at least three integrator circuits (12, 14, 16) connected therebetween, each integrator circuit including two inputs provided with capacitors which are switched under the control of two disjoint clock states, said integrator circuits being interconnected to constitute a sampling filter having a selected amplitude/frequency response. The two-port network includes two other integrator circuits (36, 38) whose respective inputs are fitted with capacitors switched under the control of two disjoint clock states, said other integrator circuit being interconnected with each other and with the first-mentioned integrator circuits in order to linearize the phase/frequency response of said filter.