Asymmetry technique for multipath mitigation in pseudorandom noise ranging receiver
This invention is directed to a method for estimating and compensating for multipath errors in a pseudorandom noise ranging receiver. The method exploits the asymmetry of the correlation function and proportionately relates it to the magnitude of multipath error. In a pseudorandom noise ranging receiver, the correlation function is obtained by generating the local pseudorandom noise sequences at different programmed non-uniform phases resulting in non-uniformly spaced correlators. Numerically controlled oscillators, code generators and shift registers are programmed to determine the correlation values at non-uniformly distributed points on a correlation function. Curve fitting is undertaken to determine the code phase at which the correlation function peaks. A proportionality constant is applied to the measure of asymmetry of the correlation function to determine the multipath error in the pseudorandom noise signal. A filter is used to detect and eliminate outliers.
1. A method of estimating the multipath error of pseudorandom noise signal in a pseudorandom noise ranging receiver, comprising the steps of:
(a) determining correlation values at programmable and non-uniformly distributed points on a correlation function, wherein each set of correlation values has a different phase with respect to a previous set of correlation values;
(b) curve-fitting the upper portion of the correlation function by polynomial interpolation for determining the code phase at which the correlation function peaks;
(c) determining a difference in area between a right section and a left section of the correlation function, the left section covering the area under the correlation function to the left of the correlation peak and the right section covering the area under the correlation function to the right of the correlation peak; and,
(d) applying a proportionality constant to the difference in area of the right section and the left section of the correlation function to determine the multipath error in the pseudorandom signal.
2. The method of claim 1 , wherein the step of applying a proportionality constant to the difference in area of a right section and a left section of the correlation function, comprises the steps of:
(a) transmitting a delayed predetermined calibrated multipath signal with a direct signal to the pseudorandom signal receiver and determining the correlation function and the shift in the correlation peak due to multipath signal, and thereafter determining the difference in area between the right section and the left section of the correlation function;
(b) determining a proportionality constant derived from the linear relationship between the shift in the correlation function peak and the area between the right section and the left section of the correlation function; and
(c) applying said proportionality constant to the difference in area of the right section and the left section of the correlation function to determine the multipath error in the pseudorandom signal.
3. A method of estimating the multipath error of pseudorandom noise signal in a pseudorandom noise ranging receiver, comprising the steps of:
(a) determining correlation values at non uniformly distributed points on a correlation function, wherein each set of correlation values has a different phase with respect to a previous set of correlation values;
(b) curve fitting an upper portion of the correlation function by polynomial interpolation for determining the code phase at which the correlation function peaks;
(c) determining a difference in area between a right section and a left section of the correlation function, the left section covering the area under the correlation function to the left of the correlation peak, and the right section covering the area under the correlation function to the right of the correlation peak;
(d) applying a proportionality constant to the difference in area of the right section and the left section of the correlation function to determine the multipath error in the pseudorandom noise signal; and
(e) removing outliers of the correlation function and filtering one or more acceptable multipath errors over multiple code tracking loop invocation periods up to a measurements generation period to improve the accuracy of the estimated multipath.
4. The method of claim 1 , wherein the step of determining the correlation values at non uniformly distributed points on a correlation function comprises programming the spacing between the correlators at the rate at which the code tracking loop is invoked thereby providing the flexibility for deriving the correlation values at different points on a correlator function at different times.
5. The method of claim 1 , wherein the step of conducting a curve fitting comprises screening the correlation values to detect outliers and erroneous data, thereby enhancing the robustness of the tracking method in weak signal strength conditions and in the presence of interferences.
6. The method of claim 1 , wherein the step of generating non-uniformly spaced correlators comprises the steps of:
(a) coupling a set of parallel numerically controlled correlators and corresponding pseudorandom noise sequence generators per satellite tracking channel;
(b) programming each numerically controlled oscillator for the entire set of numerically controlled oscillators per satellite tracking channel to trigger the corresponding pseudorandom noise sequence generator at the common programmed frequency; and
(c) programming each numerically controlled oscillator for the entire set of numerically controlled oscillators per satellite tracking channel to trigger the corresponding pseudorandom noise sequence generator with different phases, whereby the difference in phases of coarse acquisition code in each correlator in the channel gives rise to the different, programmable, non-uniform spacing between the correlators in the satellite tracking channel.