Abstract: Transport frames to be transmitted on a communication channel are formed from coded-signal frames. Each coded-signal frame comprises one or more sets of bits to be protected against transmission errors. A respective error detection code is calculated for a subset of bits, and placed in a respective transport frame along with this subset of bits. Some at least of the transport frames contain a plurality of subsets of bits, emanating from different coded-signal frames and accompanied by the corresponding error detection codes.

Abstract: A base station forms a time-division multiplexing physical control channel comprising a set of presynchronization timeslots each carrying the same synchronization signal, and a set of synchronization supplement timeslots respectively associated with the presynchronization slots. Each synchronization supplement slot carries the same synchronization signal representing system information and furthermore a signal identifying said synchronization supplement slot, thereby enabling to complete the synchronization achieved on the basis of the presynchronization slots.

Abstract: The invention concerns a method for detecting voice activity in a digital speech signal, in at least a frequency band, for example by means of a detecting automaton whereof the status is controlled on the basis of an energy analysis of the signal. The control of said automaton, or more generally the determination of voice activity, comprises a comparison, in the frequency band, of two different versions of the speech signal one of which at least is a noise-corrected version.

Abstract: A transmitter produces radio signal bursts in periodic time slots allocated to a time-division multiplexed channel. The radio signal of each burst is made up of digital symbols comprising two training sequences enabling the receiver to estimate demodulation parameters and information symbols which the receiver can estimate by a demodulation applied using the estimated parameters. The two training sequences are placed at the start and the end of the burst so that the receiver uses them to demodulate the received signal in the order of the symbols and demodulate the received signal again in the reverse order of the symbols.

Abstract: Runs of loss cable are disposed in succession along a zone of radio coverage and coupled with feeder means from base stations of a cellular mobile communication network. The feeder means apply first radio frequency signals from a first base station to a first cable run and second radio frequency signals from a second base station to a second cable run, adjacent to the first run. They also apply at least part of the second radio frequency signals to the first cable run, preferably attenuated relative to the first radio frequency signals as applied to the first cable run, creating a zone of partial coverage between adjacent cells in order to facilitate operation of an automatic cell changeover during calls.

Abstract: A transmitter outputs a radio signal formed from a first baseband signal incorporating a synchronization signal. The recipient obtains a second baseband signal from the received radio signal, detects the synchronization signal in a portion of the second baseband signal in order to compute synchronization parameters and/or parameters for estimating a radio transmission channel, and uses the computed parameters to demodulate another portion of the second baseband signal and extract the transmitted information from it. The synchronization signal contains a synchronization pattern selected by the transmitter on the basis of signalling information to be supplied to the second station. The receiver searches several patterns in the second baseband signal in order to obtain the signalling information depending on the detected pattern.

Abstract: The invention concerns a method for conditioning a digital speech signal(s) processed by successive frames, which consists carrying out a harmonic analysis to estimate the pitch on each frame where it has a speech activity, and in oversampling at an oversampling frequency (fe) which is a multiple of the estimated pitch.

Abstract: The filtering coefficients of a frequency domain stereophonic echo canceller are adapted by a method which takes account of the cross-correlation between the input signals relating to the two channels. In particular, the adaptation process takes account of the coherence function, reducing the problems commonly encountered with stereophonic cancellation schemes when relatively correlated input signals occur.

Abstract: The method determines a deviation between a modulation frequency, combined with a first baseband signal to form a transmitted signal and a frequency used by the receiver to form a second baseband signal from a received signal, in order to adjust the frequency used by the receiver. Knowing a time segment of the first signal and the time position of a corresponding segment of the second signal, whereby the segments comprise N samples at a sample frequency Fe, a frequency transform Y(f), of size N, of the product of the complex conjugate of said segment of the first signal by the corresponding segment of the second signal is calculated, and the frequency deviation &Dgr;f which makes the frequency transform Y(f) as close as possible to C.

Abstract: A digital speech signal processed by successive frames is subjected to noise suppression taking account of estimates of the noise included in the signal, updated for each frame in a manner dependent on at least one degree of vocal activity. A priori noise suppression is applied to the speech signal of each frame on the basis of estimates of the noise obtained on processing at least one preceding frame, and the energy variations of the a priori noise-suppressed signal are analyzed to detect the degree of vocal activity of said frame.

Abstract: Using a group of spectral components from an audio signal, a coder produces a digital data stream (&PHgr;) including the quantification indices (QE0m) of at least some of the spectral components (E0m). The coder, or a transcoder situated downstream, selects at least one pair of components exhibiting a maximum correlation out of the group of spectral components, and includes in its digital output data stream identification information for each pair selected. At least some of the quantification indices are included in the output data stream for only one of the components for each pair selected. The decoder will use these to obtain the suppressed indices for the other component of the pair.

Abstract: A spectral subtraction is effected including: a first subtraction step in which overestimates of the spectral component of the noise are taken into account, to obtain spectral components of a first noise-suppressed signal; the computation of a masking curve by applying an auditory perception model on the basis of the spectral components of the first noise-suppressed signal; and a second subtraction step in which a respective quantity depending on parameters including a difference between the overestimate of the corresponding spectral component of the noise and the computed masking curve is subtracted from each spectral component of the speech signal in the frame. The result of the spectral subtraction is transformed into the time domain to construct a noise-suppressed speech signal.

Abstract: The station includes a coded speech signal source having a given digital rate, a modulator for producing a radio signal from the coded speech signal provided by the source, and a redundancy encoder having a coding rate of the form K/N, where K and N are integers such that 0<K<N. An operating mode of the station is selected from at least a first operating mode where the modulator directly receives the coded speech signal and converts it into a radio signal occupying a determined fraction of time on a carrier frequency, and a second operating mode where the redundancy encoder receives the coded speech signal, and the modulator receives the output signal from the redundancy encoder and converts it into a radio signal occupying N/K times the determined fraction of time on a carrier frequency.

Abstract: The detection of symbols of a sequence on the basis of an observation signal is performed in accordance with a soft output Viterbi algorithm using a trellis. In order to assign a likelihood to a discrete estimation of a symbol, the difference between the metrics of the optimal path determined in accordance with the Viterbi algorithm and the metrics of a competitor path, which is optimal among all the paths leading to a different decision for the symbol in question is calculated, the metrics being considered on the whole length of the trellis.

Abstract: On the basis of a portion of the signal, a coder selects a parameter vector belonging to a reference library containing 2Q vectors each designated by an address of Q bits. This coder, or a transcoder located downstream, forms a digital data stream (&PHgr;) containing an index deduced from the address of the vector selected. For each rate value corresponding to an integer p≧0, the index contained in the digital data stream is formed of Q−p bits which, completed by p bits of predetermined positions, define indices of Q bits representing a group of 2p addresses including that of the vector selected from the reference library. The decoder receiving the digital data stream, or a transcoder located upstream, is capable of reconstructing appropriate parameter vectors on the basis of the truncated index. Thus very fine steps are obtained in the adjustment of the transmission rate.

Abstract: A recognition test is applied to a segment of speech produced by a speaker. An examination is carried out to see whether a first optimum model, with which the recognition test has matched the spoken segment of speech, satisfies a criterion of likelihood. If the criterion of likelihood is satisfied, this first optimum model is selected. If not, a repetition of the speech segment is requested. The recognition test is applied to the segment of speech repeated by the speaker. This second optimum model is selected if it satisfies the likelihood criterion. If not, and if a combination of the results of the two recognition tests satisfies at least one criterion of combined selection, one of the models which the two recognition tests have matched with the spoken segment of speech and with the repeated segment is selected.

Abstract: The synchronous serial bus between a main processing unit and a peripheral unit includes a data line and a clock line. Strobe pulses presented by the main processing unit on the data line while it holds the clock line at a given logic level characterise transfer cycles on the bus. The main processing unit can thus run write or read cycles in registers of an interface of the peripheral unit. A direct transfer mode, wherein the strobe pulse is transmitted at the beginning of the cycle without specifying an address, is provided to enable the main processing unit to have a fast access to certain locations previously specified. The data and clock lines of the bus may be shared with those of another synchronous bus.

Abstract: A recognition test matches a speech segment supplied to the system with a set of parameters associated with a reference and memorized in a dictionary. A provisional version of each set of parameters to be memorized in the dictionary in association with a reference is estimated on the basis of one or more segments of speech, after which the provisional version is memorized in the dictionary in association with the reference. At least one repetition of the speech segment is submitted to a recognition test, after which depending on whether it has matched the speech segment with the provisional version, the provisional version is modified and the modified provisional version is memorized.