A Multipath Mitigation Technique for BOC - CiteSeerX

0 downloads 4 Views 414KB Size Report
PRN (direct prompt)”, and is a compromise between low complexity and high ... Jinghui Wu and Andrew G. Dempster, is with the School of Surveying and Spatial ...... and for University of Westminster in London as a Lecturer, Senior Lecturer ...

> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < ACF of BOC-BOC Discriminator of BOC-BOC(E-L) (0.3Chips spacing) Discriminator of BOC-BOC(E-L) (0.5Chips spacing)

2

Discriminator Functions

2

1.5 1 0.5 0 -0.5 -1 -1.5

-1

-0.5

0

0.5

1

time delay in Chips Fig. 1. The ambiguous BOC-BOC (E-L) discriminator functions (dash dot line) constituted by the outputs of Early-armed and Late-armed correlators with 0.3 chip (dash line) and 0.5 chip (solid line) spacing. The output of the Prompt-armed correlators is the ACF (dash dot line) of BOC (1, 1).

Such a new signal structure can provide potential benefits but also introduces challenges. Because of the subcarrier, BOC (1,1) signal has a wider signal spectrum compared to the traditional GPS BPSK (Binary Phase Shift Keying) C/A signals, and a sharper/narrower peak at the symmetrical centre of the auto-correlation function(ACF) (dash dot line in Fig. 1). This is desirable for better multipath rejection performance when the receiver is operating under the popular classes of multipath mitigation solutions, originally developed for GPS BPSK C/A signals (e.g. generating shaped reference code waveforms for correlation operation [7]). However, the symmetrical secondary peaks are also generated, located at both sides of the central peak which is used as the only indication of perfect alignment between the incoming PRN code and the locally generated PRN code. As a consequence, by adopting existing techniques without modification, the resulting DLL discriminator function (dash line and solid line in Fig. 1.) may have more than one zero-crossing point (representing zero code-phase error between incoming and locally generated codes), among which only the centre one corresponds to the zero-code-phase-error inside the tracking loop. The others lead to biased pseudorange measurements. To avoid the ambiguous zero-code-phase-errors as the feedback to the Numerically Controlled Oscillator (NCO) inside the tracking loop and maintain the signal multipath rejection property, different methods have been proposed. From the view of solving ambiguity, they can be generally categorized into 4 types: Self-adjusting method, BPSK- like method, shaping method, and BOC-PRN method. Self-adjusting methods (e.g. [8]) usually provide a mechanism to estimate and correct the location of the prompt peak of the ACF so that it can prevent the tracking loop falling into false lock when it is implemented using the traditional discriminators such as BOC-BOC Early minus Late discriminator (i.e. subtraction between the early and late version

> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < discriminator with different gate width (W1

(0, 1) chips) is plotted. The wave form of reference waveform is also illustrated in

Fig. 12. W0 gref (t)

1

0 t -1 - W0 - W0 4 2

W0 4

W0 2

Fig. 5. Bipolar symmetric strobe pulse for “gating function” definition W0=0.1Chips W0=0.3Chips

pseudorange error in meters

20 15 10

In-phased Multipath Error

5 0 -5 -10

Out-phased Multipath Error

-15 -20 -25 0

0.2 0.4 0.6 0.8 relative multipath delay in Chips

Fig. 6. Multipath effects on “gating function”,

0.3

1

=0.5, Infinite BW.

W0=0.1Chips W0=0.3Chips

0.2 Normalized CCF

8

0.1 0 -0.1 -0.2 -0.3 -1.5

-1

-0.5 0 0.5 time delay in Chips

1

1.5

Fig. 7. Ambiguous “gating function” discriminators. The false nodes (zero-crossing) located at the +0.5Chips delay can lead to biased tracking, Infinite BW.

> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)
REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT)

Suggest Documents