The article investigates the quality of algorithms for esimate the main parameters of wideband signals in the problem of over-the-horizon radar under conditions of dispersion distortions in the Earth’s ionosphere. The boundary values of the variances of the estimates of propagation delay, frequency shift and phase, wideband phase-shift keyed signals are investigated. It is shown that dispersion distortions of the radar signal lead to a correlation between the estimates of the delay and frequency shift. The curves of the dependence of the frequency-time ratio depending on the length of the used pseudo-random sequence and the degree of dispersion distortions are obtained. Analytical curves of the Cramer-Rao lower bounds are constructed for the variances of estimates and the values of the correlation coefficients of the estimated parameters in the presence of dispersion distortions, which are confirmed by the results of simulation.

The detector is an integral part of the device for receiving and processing radio signals. Signals that have passed through the ionospheric channel acquire an unknown Doppler shift and are subject to dispersion distortions. It is necessary to carry out joint detection and parameter estimation to improve reception quality and detection accuracy. Modern hardware base developing makes it possible to implement a device for joint detection and evaluation of signals based on standard processors (CPU) and graphic processors (GPU). The article discusses the implementation of a signal detector that allows for real-time operation. A comparison of implementations of algorithms for estimating the Doppler frequency shift through multiplication by a complex exponent and the fast Fourier transform (FFT) is performed. A comparison of computational costs and execution speed on the CPU and GPU is considered.

Analytical expressions for the Cramer-Rao boundaries are obtained in the article when jointly evaluating a set of parameters for a broadband ionospheric-distorted signal of an arbitrary shape, namely: the steepness of the dispersion characteristic, Doppler frequency shift, delay and phase shift. The corresponding expressions for covariance functions and correlation coefficients are obtained. It is shown that the presence of dispersion distortions leads to additional, previously absent, statistical relationships of estimates of signal parameters. The correctness of analytical expressions is confirmed by the results of simulation. The reported study was funded by RFBR, project number 19-29-06055.

When using wideband signals, the Digital Down Conversion (DDC) operation is an important operation. The development of a modern hardware base allows the implementation of DDC devices based on standard processors (CPU) and graphics processing units (GPU). The article discusses an implementation based on Overlap And Save algorithm. Comparison of computational costs and execution speed on CPU and GPU is carried out. The estimation of the maximum number of devices implemented on various models of CPU and GPU has been made.

Algorithms for estimating the symbol rate of signals with modulation types M-PSK and M-FSK are presented. The algorithm for estimating the symbol rate of the M-PSK signals is based on the Fourier transform of the squares of the modules of the complex envelope of the signal. The algorithm for estimating the symbol rate of M-FSK signals is based on the separation of the spectrum of the original signal into separate subcarriers. Algorithms were simulated. It is shown that the algorithm based on separation of subcarriers is less sensitive to noise in the communication channel.

The article proposes a computationally effective scheme of the dispersion distortions compensation device, based on a filter bank using the Farrow filter. The possibility of constructing a tracking compensator for dispersion distortions of broadband signals when transmitting them through the ionospheric channel is considered. The tracking compensator allows the estimation of dispersion distortions and their compensation. The results of simulation of the proposed algorithms are presented: the obtained estimate and the true value of the frequency dispersion parameters.