Shortwave Channel Propagation Characteristics Of Research, And Methods Of Measurement

For contemporary HF radio communication systems the ionospheric channel exhibits different multi-path delay, fades, Doppler spread and Doppler Shift. Unlike other medium, the way by which the ionosphere affects the transmission of HF radio waves is unique and complicated. The multiple ionization layers result in multi-mode and multi-hopped ionospheric propagation. The unstationary channel state is deteriorated by the ionospheric disturbance and time-varying electron density irregularities. Additionally, some other physical phenomena such as attenuation, refraction, diffraction and scattering also prevent the improvement on the capacity of large channel on HF links.Thus, a need has arisen for a system capable of measuring HF channel characteristics in order to increase the available data rates on the occurrence and magnitudes of these phenomena. Experimental measurements are commonly used to specify the detailed characteristics in the 3-30 MHz HF band. The measurements have usually been made over actual ionospheric links, to obtain on-the-air data and to apply statistical processing. These measurements mostly aim for complex impulse responses because they are essential to characterization of the channel.Firstly, through channel sounding, some classic dispersion parameters such as delay spread and doppler spread are evaluated. Two methods are proposed: one is pulse compression sounding technique using PN sequence and cross-correlation, the other is based on OFDM pilot-symbol-aided channel estimation. Pulse compression and the integration of multiple pulses will yield typical signal processing gains of 27 dB (depending on the sequence length). The point is with the advent of PN sequence's fine autocorrelation property the pulse compression technique can provide a good resistance against inter-symbol interference and narrowband interference, which would severely affect the performance of the sounding technique based on channel estimation. OFDM channel estimation also suffers from inter-carrier interference. However, the performances of both methods are sensitive to white noise. So AWGN suppression is adopted. 4 different noise floors are calculated to filter noise. Through analysis and simulation, the noise floor's filtering effect on both delay and doppler dispersion parameters are compared. It is proved that taking the mean power of PDP as the noise floor has a better performance.In addition, MIMO ionospheric channel measurements determine the statistical measure of the covariance of 4 subchannels when the antenna array is 2×2. Since the data rate available for HF communication systems is comparatively low, it would be very useful if the capacity improvements inherent in MIMO could be exploited at HF application. In order to assess this possibility and in order to gain some insight into the optimum design of HF MIMO systems, the signal correlation for spaced antennas on ionospheric links are explored. Similarly, two methods are proposed: one is pulse compression sounding using orthogonal sequences, the other is based on MIMO-OFDM channel estimation. Under some specific constraints, the latter can get similar correlation measurement results with the advantage of the simplicity of implementation.Finally this dissertation presents the design and realization project of the sounding systems currently being developed. The system will use pulse compression technique to determine the following real time channel parameters: multipath dispersion, Doppler shift and Doppler spread. Another system will make use of MIMO-OFDM channel estimation to specify the covariance of 4 subchannels on HF MIMO links. Limitations and accuracy of these measurements are discussed. Considering the requirements that are set by channel condition and the limitation set by laboratory measurement devices, implementation of the two systems based on FPGA and DSP is presented respectively.