COFDM demodulator with an optimal FFT analysis window positioning
The invention relates to a method for the COFDM demodulation of a signal received from a transmission channel. The inventive method includes performing the fast Fourier transform of the signal received in a window corresponding to a symbol, each symbol being associated with a guard time reproducing one part of the symbol; supplying a set of estimated values for the module impulse response; determining coefficients, each coefficient being obtained from the product of the aforementioned set and a filtering function (FE) for a determined relative position of the filtering function in relation to the set; determining the maximum coefficient and the corresponding relative position; and positioning the window as a function of the relative position, the filtering function including a central part (LMAX) which has a constant amplitude and a duration equal to the duration of the guard time and which is surrounded by non-zero decreasing edges.
1. A method of COFDM demodulation of a signal received from a transmit channel, comprising:
performing a fast Fourier transform of the received signal in a window corresponding to a symbol, each symbol comprising several carriers modulated in phase and/or in amplitude, some of which are pilots, and being adjacent to a guard interval reproducing part of the symbol;
providing a set of estimated values of the pulse response in module from the pilots;
determining coefficients, each coefficient being obtained from the product of said set and of a filtering function for a determined relative position of the filtering function with respect to said set;
determining the maximum coefficient and the corresponding relative position; and
positioning said window according to said relative position corresponding to the maximum coefficient,
the filtering function comprising a central portion of constant amplitude and of duration equal to the duration of the guard interval, surrounded with non-zero decreasing sides.
2. The method of claim 1 , wherein providing the set of estimated values of the pulse response in module comprises providing a set of estimated values of the frequency response of the transmit channel based on the pilots, and a step of transformation of said set of estimated values of the frequency response by inverse fast Fourier transform.
3. The method of claim 1 , wherein, for each relative position of the filtering function with respect to the set of estimated values of the module of the pulse response, the coefficient is determined based on the sum of the products of the module of the estimated values of the pulse response and of the filtering function.
4. The method of claim 1 , wherein the set of estimated values of the module of the pulse response is periodic, the filtering function having a total width smaller than the period of the set of estimated values of the module of the pulse response.
5. The method of claim 1 , wherein the filtering function is a stepped function.
6. The method of claim 5 , wherein each side comprises at least a first and a second adjacent steps each having a non-zero amplitude strictly smaller than the amplitude of the central portion, the double of the amplitude of the second step being greater than the sum of the amplitude of the central portion and of the amplitude of the first step.
7. The method of claim 6 , wherein the duration of the second step is smaller than the duration of the first step.
8. The method of claim 1 , wherein the filtering function is symmetrical.
9. A COFDM demodulator intended to receive a signal received from a transmit channel, comprising:
a circuit of fast Fourier transform of the received signal in a window corresponding to a symbol, each symbol comprising several carriers modulated in phase and/or in amplitude, some of which are pilots, and being adjacent to a guard interval reproducing part of the symbol;
a circuit for providing a set of estimated values of the module of the pulse response from the pilots;
a circuit for determining coefficients, each coefficient being obtained from the product of said set and of a filtering function for a determined relative position of the filtering function with respect to said set;
a circuit for determining the maximum coefficient and the corresponding relative position; and
a circuit for positioning said window according to said relative position corresponding to the maximum coefficient,
the filtering function comprising a central portion of constant amplitude and of duration equal to the duration of the guard interval, surrounded with non-zero decreasing sides.