METHOD OF PROCESSING A DIGITAL SIGNAL DERIVED FROM A DIRECT-SEQUENCE SPREAD SPECTRUM SIGNAL AND A RECEIVER

• US 20100189163A1
• Filed: 01/26/2010
• Published: 07/29/2010
• Est. Priority Date: 01/27/2009
• Status: Active Grant

• Current Assignee: U-Blox AG (u-blox Holding AG)
• Original Assignee: U-Blox AG (u-blox Holding AG)
• Inventors: Bürgi, Clemens, Waelchli, Grégoire, Baracchi, Marcel
• Granted Patent: US 8,295,324 B2
• Time in Patent Office: Days
• US Class Current: 375/142
• CPC Class Codes: G01S 19/29 carrier including Doppler, related G01S19/246 takes precedence
• CPC Class Codes: G01S 19/30 code related G01S19/246 takes precedence
• CPC Class Codes: H04B 1/70752 Partial correlation
• CPC Class Codes: H04B 1/7085 using a code tracking loop, e.g. a delay-locked loop
• CPC Class Codes: H04B 2201/70715 with application-specific features

Abstract

Complex digital data values derived from a DSSS signal, in particular, a GNSS signal, are delivered to a general purpose microprocessor at a rate of 8 MHz and chip sums over eight consecutive data values spaced by a sampling length (T_S), each beginning with one of the data values as an initial value, formed and stored. For code removal, each of a series of chip sums covering a correlation interval of 1 ms and each essentially coinciding with a chip interval of fixed chip length (T_C), where a value of a basic function (b_m) reflecting a PRN basic sequence of a satellite assumes a correlation value (B_m), is multiplied by the latter and the products added up over a partial correlation interval to form a partial correlation sum. The partial correlation interval is chosen in such a way that it essentially coincides with a corresponding Doppler interval having a Doppler length (T_D) where a frequency function used for tentative Doppler shift compensation and represented by a step function (sine, cosine) is constant. The partial correlation sums are then each multiplied by the value assumed by the frequency function in the corresponding Doppler interval and the products added up to form a correlation sum.

Claims

1. A method of processing a digital signal derived from a direct-sequence spread spectrum analog signal containing at least one encoded binary sequence consisting of repetitions of a characteristic basic sequence which extends over a basic interval to determine a multiplicity of correlation sums, the method comprising:

providing a correlation sequence corresponding to the basic sequence which consists of correlation values of +1 and −

1 each pertaining to one of a series of consecutive chip intervals having a constant chip length,

deriving from the digital signal a data sequence consisting of data values which follow upon each other separated in each case by a sampling interval, the sampling intervals having a constant sampling length which is not greater than one half of the chip length such that a plurality of consecutive data values is contained in a chip interval in each case, and

correlating the data sequence with a plurality of copies of the correlation sequence, the copies having different phase positions with respect to the data sequence, and forming a correlation sum comprising products of correlation values and data values over a correlation interval in each case,

wherein a sequence of overlapping chip sums is formed by calculating, for every data value, at least one chip sum with the data value as an initial value, where the chip sum is in each case a linear combination of data values in a set consisting of the initial value and immediately subsequent data values, wherein the summation number of data values contained in the set multiplied by the sampling length differs in each case by at most the sampling length from the chip length,

a series of subsequences of chip sums is formed, each subsequence selected from the sequence of chip sums, with the position of the initial value of a first chip sum as a starting point, in such a way that

at every end point of an interval beginning at the starting point of a first subsequence and having a length corresponding to a multiple of a resolution length which is not greater than half the chip length, at least where the end point is situated within a chip length from the said starting point, there is a starting point of a further subsequence whose position differs from the position of the said end point by not more than one half of the sampling length, and

each subsequence has, for every end point of an interval beginning at its starting point and having a length corresponding to a multiple of the chip length, an initial value whose position differs from the position of the said end point by not more than one half of the sampling length, and

wherein for every subsequence, every chip sum is multiplied with one corresponding correlation value of the correlation sequence and the products summed up over a partial correlation interval to form a partial correlation sum which is then used for the determination of one of the correlation sums.

2. The method of claim 1, wherein the distance separating the positions of two consecutive initial values of a subsequence equals in each case the sampling length multiplied by a step number which assumes either a lower step number value or a higher step number value which are the integers closest to the quotient resulting from the division of the chip length by the sampling length.

3. The method of claim 2, wherein the step numbers are in each case calculated prior to the formation of the partial correlation sum, and encoded by a single bit stored in a memory unit.

4. The method of claim 2, wherein the summation number for a chip sum used for forming the partial correlation sum equals in each case the step number used to calculate the distance separating the positions of the initial value of the chip sum and the initial value of the subsequent chip sum of the subsequence.

5. The method of claim 4, wherein, where applicable, chip sums are formed with each of the two step number values used as a summation number in each case.

6. The method of claim 1, wherein the summation number for every chip sum equals the integer closest to the quotient resulting from the division of the chip length by the sampling length.

7. The method of claim 1, where the resolution length equals the sampling length or a multiple thereof.

8. The method of claim 1, where the analog signal is a GNSS signal and a residual carrier which reflects at least one Doppler shift of a component of the analog signal is superposed on the data sequence, and, for tentative removal of the Doppler shift,

multiplying each of the digital data values by a value of a frequency function which assumes one of a finite number of constant values in each of a series of consecutive Doppler intervals, the Doppler length of each of which is at least twice the chip length in each case and

for each Doppler interval, choosing a sequence of consecutive chip intervals, which overlap with the Doppler interval at least by half, as a partial correlation interval, the corresponding partial correlation sum is established and the products of each partial correlation sum and the value of the frequency function assumed by the same in the Doppler interval added up over the correlation interval for the determination of the correlation sum.

9. The method of claim 8, wherein a sum over all partial correlation sums pertaining to a value of the frequency function is formed and then multiplied by the said value in each case before the products are added up.

10. The method of claim 8, wherein the data values are complex values and the frequency function is a complex step function approximating a complex exponential function.

11. The method of claim 10, wherein each of the two components of the frequency function is encoded by a number comprising between one bit and five bits.

12. The method of claim 8, wherein, before being multiplied with a partial correlation sum or a sum of partial correlation sums the value of the frequency function is in each case read from a memory unit.

13. The method of claims 8, wherein the Doppler intervals have each the same Doppler length.

14. The method of claim 1, wherein the correlation interval has the length of a basic interval.

15. A receiver comprising a front-end circuit with mixers and A/D converters for determining, from a direct-sequence spread spectrum analog signal containing at least one encoded binary sequence consisting of repetitions of a characteristic basic sequence which extends over a basic interval, a multiplicity of correlation sums, in deriving, from the said analog signal, a digital signal consisting of data values which follow upon each other separated in each case by a sampling interval, the sampling intervals having a constant sampling length which is not greater than one half of the chip length such that a plurality of consecutive data values is contained in a chip interval in each case, and for providing a correlation sequence corresponding to the basic sequence which consists of correlation values of +1 and −

1 each pertaining to one of a series of consecutive chip intervals having a constant chip length,

the front-end unit further comprising FIR filter circuits for forming a sequence of overlapping chip sums by calculating, for every data value, at least one chip sum with the data value as an initial value, where the chip sum is in each case a linear combination of data values in a set consisting of the initial value and immediately subsequent data values, where the summation number of data values contained in the set multiplied by the sampling length differs in each case by at most the sampling length from the chip length, and

the receiver further comprising a base-band processor with a microprocessor unit and a memory unit configured for correlating the data sequence with a plurality of copies of the correlation sequence, the copies having different phase positions with respect to the data sequence, and forming, in each case, one of the correlation sums comprising products of correlation values and data values over a correlation interval, in that

a series of subsequences of chip sums is formed, each subsequence selected from the sequence of chip sums, with the position of the initial value of a first chip sum as a starting point, in such a way that

at every end point of an interval beginning at the starting point of a first subsequence and having a length corresponding to a multiple of a resolution length which is not greater than half the chip length, at least where the end point is situated within a chip length from the said starting point, there is a starting point of a further subsequence whose position differs from the position of the said end point by not more than one half of the sampling length, and

each subsequence has, for every end point of an interval beginning at its starting point and having a length corresponding to a multiple of the chip length, an initial value whose position differs from the position of the said end point by not more than one half of the sampling length, and where

for every subsequence, every chip sum is multiplied with one corresponding correlation value of the correlation sequence and the products summed up over a partial correlation interval to form a partial correlation sum which is then used for the determination of one of the correlation sums.