Abstract:
A method for demodulating a modulated signal, by: receiving a signal modulated in n-PSK or n-APSK comprising a succession of symbols organized in frames, each frame comprising a header followed by blocks of data symbols separated by blocks of pilot symbols, determining the phase of the headers and pilot blocks to predict the evolution of the signal phase, correcting the phase of the data symbols according to the evolution of the signal phase, and equalizing the data symbols corrected in phase using equalization coefficients evaluated thanks to estimated or known symbols of the signal, and pre-equalizing the header, pilot and data symbols, which is performed before determining the phase of the headers and pilot blocks, and using the estimated equalization coefficients to equalize the data symbols.
Abstract:
A process of correction of the spectral inversion for a receiver in a digital communication system: the process allows the reception in the receiver of a training sequence presumably known according to a modulation of type π/2 BPSK or MDP2. The process includes the following steps: Demodulating of the training sequence; Calculating of the differential correlation on a set of N received samples (Rn) and presumably sent (Sn) to generate a result; Using the result to detect the beginning of the frame and to order a spectral inversion in the chain of reception of the aforementioned receiver before launching the detection of the beginning of the frame. A receiver to process automatically the spectral inversion is also described.
Abstract:
A device for providing a digital error signal, for a timing correction loop of a digital demodulator for digital transmission by phase modulation or amplitude and phase modulation, the device successively receiving pairs of digital signals representative of the components of complex signals, and having circuitry for providing a difference signal representative of the difference between the modulus of the complex signal corresponding to the last received pair of digital signals and the modulus of the complex signal corresponding to the previously-received pair of digital signals; circuitry for providing a weighting factor which depends on the angle between the complex signal corresponding to the last received pair of digital signals and the complex signal corresponding to the previously-received pair of digital signals; and circuitry for providing the error signal proportional to the product of the difference signal and of the weighting factor.
Abstract:
An extractor for digital data transmitted at a first determined frequency (f0) through a video channel after a burst of 0s and 1s emitted at a first frequency (f0). A comparator (1) compares the input signal with a threshold level. A threshold level is provided by an up/down counter (12) operating at a frequency (F0) multiple of the first frequency, the up/down counting input of which is connected to the output of the comparator (1), and a digital/analog converter (16) receiving the output of the up/down counter and supplying the threshold level (V.sub.T).
Abstract:
The present invention relates to a method for correcting errors in a data frame including horizontal parity data enabling correction of errors in the rows of the frame based on horizontal syndromes calculated on the rows, and vertical parity data enabling correction of errors in the columns of the frame based on vertical syndromes calculated on the columns. The method includes the steps of calculating, on the fly, the horizontal and vertical syndromes of a current frame on the data of the current frame being received in a slow memory, storing these syndromes in a fast memory area, and, as the data of the next frame are being received in the slow memory, finding the values and positions of the errors of the current frame based on the syndromes stored in the fast memory area.
Abstract:
A Reed-Solomon decoder receives code sequences of M coefficients having a maximum value N, t of which can be corrected. The Reed-Solomon decoder includes 2t polynomial counters successively receiving the M coefficients of each code sequence, the polynomial counter of rank i (i=0,1 . . . 2t-1), providing the coefficient of the term of degree i of a syndrome polynomial. A circuit provides the coefficients of an error locator polynomial from the coefficients of the syndrome polynomial. Another circuit finds the roots of the error locator polynomial by successively trying values .alpha..sup.1 to .alpha..sup.M. The polynomial counter of rank i is preceded by a multiplier by .alpha..sup.(B+i)(N-M), .alpha..sup.B+i being the i-th root of the code generator polynomial.
Abstract:
A convolution decoder includes, for each state S of a shift register receiving an initial signal, an add-compare-select circuit which provides a one-bit decision for selecting either one of states 2S or 2S+1 as a state preceding the current state S. A decoding element traces back the memory according to a path indicated by the decisions stored in the memory in order to restore the succession of states of the initial signal. Each calculation cell associated with a state S further includes means for establishing a complex R-bit decision comprising, by decreasing weight, the one-bit decision of the calculation cell and the R-1 most significant bits of the complex decision established by the cell associated with the selected state 2S or 2S+1.
Abstract:
A circuit that extracts the synchronization signal from a composite video signal. The circuit includes a stage for aligning the low level of the synchronization signal interval with a reference voltage; a stage for detecting the signal suppression level and a comparator for comparing the aligned video signal with a value intermediate between the low level of the synchronization interval and the signal suppression level. The detecting stage includes a second comparator charging or discharging a capacitor depending on the polarity of the voltage difference across its inputs. The ratio Ic/Id between the values of the charging and discharging currents is selected (approximately equal to 8 in one embodiment) to obtain the suitable value Vsup at the second input of the comparator.
Abstract:
Curable, saturated polyester resins and epoxy resin cross-linking agents therefor, in association with an at least divalent metal oxide and a cross-linking catalyst, are useful binders for mechanically improved powder paints, e.g., those typically applied to metallic substrates by electrostatic deposition or fluidized bed spraying.
Abstract:
The invention relates to compounds based on saturated polyester and cross-linking agent. The compositions comprise a saturated polyester with terminal hydroxyl groups and a cross-linking agent consisting of a dianhydride with ester groups, the product of reaction between a trimellitic anhydride and a diol. Use of the compounds as a binder for powder paints applicable to the automobile, electrical household goods, agricultural machinery and lead industries.