Abstract:
A calibration method and system for reducing modulation errors in a telecommunication transmitter apparatus includes providing a pair of test signals, which are substantially in quadrature to each other, and to repeat an estimation loop. The estimation loop starts with generating a modulated signal by modulating the test signals (the modulation introducing a modulation error) and continues by obtaining a squared signal corresponding to the square of the modulated signal. A transformed signal corresponding to the squared signal in the frequency domain is then calculated. The estimation loop further includes calculating an error indicator, indicative of the modulation error, according to the modulus of the transformed signal. A compensation, to be applied to the test signals for counterbalancing the modulation error, is calculated according to the error indicator and is then applied to the test signals. The estimation loop is repeated until the error indicator reaches a threshold and an indication of the compensation resulting from a last iteration of the estimation loop is stored (for the application of the compensation to each pair of operative signals managed by the telecommunication transmitter apparatus during operation thereof).
Abstract:
In order to execute, as a function of a received signal (r), a procedure of channel estimation in a transmission channel with memory in a telecommunications system, there is envisaged an operation of estimation of a delay spread associated to said channel, said operation of estimation comprising calculation of a root mean square value (τ rms ) of delay spread by means of a step of evaluation of crossings with a threshold level of a quantity associated to a transfer function of said channel. Said step of evaluation of crossings comprises evaluating a mean number of crossings (λ 0 ) of the real and imaginary parts of said channel transfer function with a threshold level corresponding to the zero level. Preferential application is to OFDM telecommunications systems and in particular wireless systems according to the IEEE 802.11a WLAN standard or the HyperlanII WLAN standard.
Abstract:
The Doppler spread associated to a transmission channel with a gain represented by a random process, is estimated by transmitting on the channel a digital signal (DPCCH I,Q ), which comprises at least one pilot signal, which in turn comprises fields of known symbols, and estimating (22), on the basis of the pilot signal (DPCCH I,Q ), the channel so as to generate a signal indicating the aforesaid gain. There is then detected (22) the zero-crossing rate (η) of the aforesaid signal during a given time interval, and there is also estimated (28) the signal-to-noise ratio (SNR) associated to the channel. The bandwidth (f̂ D ) of the aforesaid random process is estimated (26) according to a reference quantity (I (2) D ), which comprises:
a first term (η̂ 2 π 2 ), representing an estimate (η̂) of said zero-crossing rate (η); and a second term ( I N (2) - η ˆ 2 π 2 I N (0) 2SNR ), which includes said signal-to-noise ratio of the channel.
Abstract:
A method for the estimation of the transfer function of a transmission channel in a receiving system of UMTS type envisages the computation of a plurality of channel coefficients, included among known channel coefficients corresponding to pilot symbols, through the reiteration of an interpolation algorithm, capable of calculating an intermediate point (Z, f(Z)) between a first extreme and a second extreme of a determined interval, the first extreme being formed by at least two known points and the second extreme being formed by at least one known point, the intermediate point to be calculated having as abscissa (Z) the abscissa value of the mean point between the points defining the interval rounded off to the integer closest to the first extreme, and having as ordinate (f(Z)) the arithmetic average between the ordinate of the known point of the second extreme and the ordinate of a point, chosen between the two known points of the first extreme, having a distance from the intermediate point equal to the distance between the intermediate point and the known point of the second extreme.
Abstract:
A method is described for designing a new prunable S-random interleaver (I) class to be used as a constituent part of turbo codes (C1, C2). With respect to previously proposed solutions the method has the advantage of being prunable to different block sizes while exhibiting at the same time, for any considered block size, performance comparable with the optimized "ad hoc" S-random interleavers. Another advantage is that, as for every S-random interleaver, the design rules are independent of the constituent codes and of the puncturing rate applied to the turbo code. Therefore, these interleavers potentially can find applications in any turbo code scheme (C1, C2) that requires interleaver size flexibility and code rate versatility, thanks to the advantage of requiring a single law storage (i.e. one ROM storage instead of several ROMs) from which all the others are obtained by pruning, without compromising the overall error rate performance.