Research On The Measurement Of Time Difference Of Arrival For X-ray Pulsar-based Navigation

Pulsar is a kind of neutron star with stable self-rotation and strong magnetic field. Satellites orbiting the earth or in the interstellar space can receive periodical X-ray pulse signals, whose period stability is as good as a cesium atomic clock. Timing, position-determination, attitude-determination and navigation based on X-ray pulsars can offer accurate time, position, attitude, velocity and orbit information for the spacecrafts in near-earth orbit, geostationary orbit, ellipticorbit or interplanetary orbit. XPNAV (X-ray Pulsar-based Navigation) system can also offer a uniform time-space coordinate for the‘Beidou’satellite navigation system or other satellite nets. Without the aid of accessorial satellites or other radar stations, XPNAV system enjoys the ability of concealment and self-survival in military affairs. TDOA (Time Difference of Arrival) measurement, a key technology in XPNAV system, decides the presivion of XPNAV’s timing and position-determination.The paper focuses on a few key points of TDOA measurement in XPNAV system, and the most distinctive parts are listed as follows:1. Based on‘hard’multi-channel dedispersion technology, the algorithm of‘soft’multi-channel dedispersion is presented by deducing the relationship of the power spectra using IDFT (Inverse Discrete Fourier Transform) and spectrum windows. In contrast, the algorithm of‘soft’multi-channel dedispersion is more flexible than Incoherent dedispersion and more real-time than Coherent dedispersion.2. Non-linear and non-gaussian characteristics of the average pulse profiles of pulsar signals are analyzed using bispectra technology. The results show that there are nonlinear phase couplings and non-gaussian elements between the two frequency axes of the bispectrum charts. Then one-dimension selected line spectra algorithm for extracting pulsar signals’characteristics is proposed. The algorithm makes use of feature Core and Domain to represent the generality of bispectrum and its distribution domain. And the variance of signal feature vector and the distance between two kinds of signal feature vectors are defined, based on which one-dimension line spectral vector with best interclass separability is redefined. The results of recognition experiments show the algorithm of one-dimension selected line spectra is suitable for extracting pulsar signals’characteristics. And compared with selected bispectra, selected line spectrum has the maximum interclass separability from the point of view of the whole one-dimension feature vector.3. Faint pulse detection technology is presented, based on Consecutive 1(1/2) spectra transformation. 1(1/2) spectra transformation on pulsar signals is done with spectra window sliding,and then the ratio between the spectra value of pulsar signal and that of noise is defined as the criterion of pulsar pulse’s absence. At last, the influence from the dispersion on the detection performance is analyzed. Results of experiments show that 1(1/2) spectra of pulsar signals are so different from that of noises from the same direction, and the detection performance on single pulsar pulse is better than that of FFT.4. The algorithm of transformation from atomic time in the satellites orbiting the earth or in the interplanetary orbit is studied. Then the transferring algorithm of observation stations from satellites to SSB (Solar System Barycenter) is studied. The influences from Doppler delay effect, Parallax effect and Sun-Shapiro delay effect on the measurement precision of TDOA are analyzed, and the influences from the position errors of satellites on Parallax effect and Sun-Shapiro delay effect are analyzed too. The iterance algorithm of TDOA’s absolute measurement is presented, whose initial value is worked out by the estimated scope of satellite’s position. At last, the convergence ability of the iterance algorithm is verified by experiments.5. For reducing the computation complexity of TODA period ambiguity resolution for XPNAV, a new relation formula of full-period number is presented, whose factors are decided by the characteristics of X-ray pulsars and whose constant is decided by TDOA phase shifts. Then the fast ambiguity resolution algorithm is presented. Theoretical and simulation analysises show that the new method can reduce the complexity of computation obviously. Meanwhile the new relation formula of ambiguity resolution has constants, which can be solidified into hardware or be stored into memory unit, so that the new ambiguity resolution algorithm is easy to implement in engineering.6. In order to improve time offset measurement precision of pulsar integrated pulse profile and enhance the capability of noise suppression, time offset measurement algorithm is proposed, based on bispectra for pulsar integrated pulse profile. Time offset measurement algorithm based on bispectra can be deduced by contrasting self-bispectrum and cross-bispectrum and solving the extremum formula. The results of experiments show that time offset measurement algorithm based on bispectra can restrain additive Gaussian white noise, and its precision is better than one based on power spectrum proposed by Taylor.