CN110007325B - Rapid frame synchronization method for satellite-based enhanced L5 signal - Google Patents

Abstract

The invention provides a fast frame synchronization method for a satellite-based enhanced L5 signal, which comprises the steps that data acquired by an SBAS L5 receiver through a carrier wave are subjected to Manchester decoding to generate first-time decoded data; decoding the FEC by using a Viterbi decoding algorithm for the data subjected to Manchester decoding to generate original navigation message data; finding the position or offset of the frame identifier in the original navigation message data; and adjusting the epoch count in the receiver according to the offset of the frame header in the original data to enable the epoch count of the initial position of one frame of the navigation message to be equal to 0. The method has simple algorithm flow, improves the storage space and the operation speed, approaches the misjudgment rate to 0, and can meet the engineering requirements.

Description

A fast frame synchronization method for satellite-based enhanced L5 signal

technical field

The invention belongs to the field of satellite navigation, and relates to a signal frame synchronization method.

Background technique

At present, satellite navigation technology has been widely used in various fields and has become an important link in the development of the country. However, due to the limitations of technology and systems, it cannot meet the needs in some high-precision applications. Satellite-Based Augmentation System (SBAS) has the characteristics of high precision, wide range and low cost, and has become an important technical direction for the development of satellite navigation in various countries.

SBAS is to monitor the navigation satellite by a large number of widely distributed monitoring stations (the location is known), obtain the original measurement data and send it to the main control station. The main control station calculates various correction information such as ephemeris error, satellite clock error, ionospheric delay, etc., and injects it to the Geostationary Earth Orbit (GEO) satellite through the uplink injection station. Finally, the correction information is broadcast to the majority of users to improve positioning. precision.

In 2016, the United States released the "IWG Satellite-Based Augmentation System L5 Interface Control Document (SBAS ICD) Draft", adding L5 signals to achieve L1 and L5 dual-frequency tracking, which can effectively correct ionospheric delay errors, provide corresponding integrity protection, and ensure that System availability and continuity.

One frame of SBAS L5 signal message is 250 bits long. The first 4 bits of a frame are the frame identification, followed by the 6-bit information type number and the 216-bit data area, and finally the 24-bit CRC check. The navigation message frame format of SBAS L5 signal is shown in Figure 1.

The frame identification of SBAS L5 signal message adopts distributed arrangement. A complete 24-bit frame identifier is distributed in the first four bits of the 6 frames in the order of 01011100 0110 1001 0011 1010.

The forward error correction code (Forward Error Correction, referred to as FEC) of 1/2 coding rate is used for the original navigation message of 250bps, and a data stream of 500bps is generated. The specific circuit is shown in Figure 2. After Manchester Encoded (Manchester Encoded), it becomes a data stream with a rate of 1000bps, and after being synthesized with a spread spectrum code, it is transmitted through carrier modulation.

However, the traditional feature character matching method faces the distributed characteristics of SBAS L5 message frame identification, and has the shortcomings of large demand for data bits and low recognition rate.

SUMMARY OF THE INVENTION

In order to overcome the deficiencies of the prior art, the present invention provides a frame header search method based on CRC check, which can quickly and accurately complete frame synchronization.

The technical scheme adopted by the present invention to solve its technical problem comprises the following steps:

1) The data obtained by the SBAS L5 receiver through the carrier is first decoded by Manchester to generate the first decoded data;

2) The data after Manchester decoding is used to decode the FEC with the Viterbi decoding algorithm to generate the original navigation message data;

3) Find the position or offset of the frame identifier in the original navigation message data;

4) Adjust the epoch count inside the receiver according to the offset of the frame header in the original data, so that the epoch count at the starting position of one frame of the navigation message is equal to 0.

The step 3) uses the method of feature character matching to find the position or offset of the frame identification.

Described step 3) comprises the following steps:

Step 3.1) obtain 500 original data sequences after Viterbi decoding;

Step 3.2) Match the nth 4-bit frame identification in the 500-bit original data sequence, if n is greater than 6, end step 3);

Step 3.3) If the nth frame identifier is found and the offset is less than 250, then jump to step 3.4), otherwise add 1 to the n value and jump to step 3.2);

Step 3.4) carry out CRC verification and judgment;

Step 3.5) If the CRC check is correct, then find the frame header, otherwise jump to step 3.3) and continue to search for the frame header in the remaining original data;

Step 3.6) Calculate the position of the frame header in the 500-bit original data.

Described step 4) comprises the following steps:

Step 4.1) obtain 2400 bits of data, carry out Manchester decoding to obtain 1200 bits of data;

Step 4.2) 1200 bits of data are carried out Viterbi decoding;

Step 4.3) in 600 bits of data, search frame header based on CRC method;

Step 4.4) Jump to step 4.8) if the frame header is found, otherwise continue to search;

Step 4.5) to 1200 bit dislocation processing, carry out VCP decoding;

Step 4.6) jump to step 4.8) if the frame header is found, otherwise jump to step 4.7); if it is the second inverted data bit search, then end step 4);

Step 4.7) Invert the 2400-bit data, and repeat the above steps 4.1) to 4.6);

Step 4.8) Complete the epoch adjustment so that the position epoch count at the beginning of a frame is 0.

The beneficial effects of the invention are: the algorithm flow is simple, the storage space and the operation speed are improved, the misjudgment rate is close to 0, and the engineering needs can be met.

Description of drawings

Figure 1 is the SBAS L5 message format;

Figure 2 is the L5 FEC register value;

Fig. 3 is the frame synchronization process block diagram of the navigation receiver;

Fig. 4 is the search frame header algorithm flow chart based on CRC check;

Figure 5 is a flow chart of the realization of the SBAS L5 frame synchronization project.

Detailed ways

The present invention will be further described below with reference to the accompanying drawings and embodiments, and the present invention includes but is not limited to the following embodiments.

The frame synchronization process of SBAS L5 provided by the present invention is shown in Figure 3. The data obtained by the receiver through the carrier needs to undergo four processes: Manchester decoding, Viterbi decoding, frame header search and epoch adjustment.

a) Manchester decoding

The data obtained by the SBAS L5 receiver through the carrier is first decoded by Manchester to generate the first decoded data. The Manchester decoding rule is executed according to the general rule that 10 is 0 and 01 is 1.

b) Viterbi decoding

After Manchester decoding, the FEC is decoded by the Viterbi decoding algorithm to generate the original navigation message data. The Viterbi decoding adopts the code rate of 1/2 and generates the parameters of the polynomial G1 (171) G2 (133) to complete the software decoding.

c) Search frame header

To find the position or offset of the frame identification in the original navigation text data string, the method of matching characteristic characters is generally adopted.

For the distributed frame header of SBAS L5, the traditional characteristic character method requires at least 6 frames of original data, that is, 1500-bit data in 6 seconds. Considering the boundaries of the frame IDs, at least 1504 bits of data are required to contain the complete 6 frame IDs. Due to the distributed 4-bit frame identification, many misjudgments will be encountered in the 1500-bit original navigation message data sequence, and the complexity of the algorithm is greatly increased.

The invention adopts the frame identification search method based on CRC check, so long as there is a complete frame of the navigation message data sequence, the frame identification can be accurately found. Taking into account edge cases, the minimum required data bits is 500 bits.

In order to ensure that the CRC check can accurately identify a whole frame of data, the missed detection rate of the CRC check, that is, different data have the same check, is analyzed.

For general CRC coding in the form of (n, k), k is the length of the binary sequence to be coded, n is the length of the coded binary sequence, then r=n-k is the order of the generator polynomial.

A given information sequence mi ( _i =1∼k) of length k can be expressed as the following polynomial:

m(x)=m _k +m _k-1 x+m _k-2 x ² +…m ₁ x ^k-1 (1)

The generator polynomial g(x) can be expressed as the following polynomial:

The polynomial m(x) x ^r is divided by the generator polynomial g(x), and the remainder R(x) is obtained as follows:

R(x)=pr +pr _-1 x+pr _-2 x ² +...p ₁ x _r ^-1 (3)

Then p ₁ p ₂ ... _pr constitute the CRC check code sequence.

For the k data sequence to be encoded, there will be ^2k different expressions, and for a generator polynomial of degree r at most ^2r different remainders. Then theoretically, there will be 2 ^k /2 ^r = 2 ^kr different data sequences sharing a remainder, and there is a probability of misjudgment. The larger the value of kr is, the more data sequences will be misjudged, but the probability of misjudgment is only related to the value of r, which is 1/2 ^r .

The L5 navigation message takes the form of (250, 226), and its generator polynomial is:

in,

Therefore, the probability of wrong judgment generated by L5 frame header search is 1/2 ²⁴ , which is close to 0, which fully meets the actual needs of the project.

The specific process of searching the frame header is shown in Fig. 4 .

Step 1: Obtain the 500-bit original data sequence after Viterbi decoding.

Step 2: Match the nth 4-bit frame identifier in the 500-bit data sequence, if n is greater than 6, jump to step 7.

Step 3: If the nth frame identifier is found and the offset is less than 250, then jump to step 4, otherwise n+1 and jump to step 2.

Step 4: Carry out CRC check and judgment.

Step 5: If the CRC check is correct, find the frame header, otherwise jump to step 3 to continue searching for the frame header in the remaining original data

Step 6: Calculate the position of the frame header in the 500-bit raw data.

Step 7: End.

d) Adjust the epoch.

The epoch count inside the receiver is adjusted according to the offset of the frame header in the raw data. Make the epoch count at the start of one frame of the navigation message equal to 0.

The navigation receiver tracking loop algorithm generally uses the Costas phase detector. Because it is insensitive to the 180-degree phase reversal of the I and Q signals, the navigation data bits may be reversed. The misalignment caused by the uncertainty of the starting G0G1 position of the bit stream input to the Viterbi decoder leads to incorrect decoding results.

In view of the two kinds of uncertainties in the navigation message data in practical engineering applications, it is necessary to search for the frame headers in sequence in four cases. (1) Viterbi decoding G0G1 without data bit reversal does not produce misalignment; (2) Viterbi decoding G0G1 without data bit reversal produces misalignment; (3) Data bit reversal Viterbi decoding G0G1 does not produce misalignment; ( 4) Data bit inversion Viterbi decoding G0G1 produces dislocation.

Considering the loss of data bits in decoding and the boundary alignment characteristics of data storage in engineering, 2400 bits of data are selected to complete frame synchronization. Ideally, there are 600 bits of data after Manchester and Viterbi decoding, which is greater than the minimum data requirement of 500 bits.

The specific process of project realization is shown in Figure 5, and the steps are as follows:

Step 1: Obtain 2400-bit data, perform Manchester decoding to obtain 1200-bit data;

Step 2: perform Viterbi decoding on the 1200-bit data;

Step 3: Search the frame header based on the CRC method in the 600-bit data;

Step 4: Skip to step 8 when the frame header is found, otherwise continue to search;

Step 5: Perform VCP decoding on 1200-bit dislocation processing;

Step 6: Go to Step 8 if the frame header is found, otherwise go to Step 7. If it is the second inverted data bit search, go to step 9;

Step 7: Invert the 2400-bit data, and repeat the above steps 1-6;

Step 8: Complete the epoch adjustment so that the position epoch count at the beginning of a frame is 0;

Step 9: End.

Implementation Effect:

The frame synchronization method described in the present invention has been implemented by programming in C language on the TI DSP 6657 platform. The following table measures the key block times during frame synchronization.

Run 6 frame synchronizations, and the time statistics are shown in Table 1 below:

Table 1 Time consumed by 6 frame synchronizations (unit is nanoseconds)

The key functions or function modules in frame synchronization do time statistics as shown in Table 2:

Table 2. Key functions and module time measurements (in nanoseconds)

The time statistics of the decoding function after frame synchronization are shown in Table 3:

Table 3 Decoding function time measurement after frame synchronization (unit is nanoseconds)

Through the comparative analysis of Table 1, Table 2 and Table 3:

(1) The telegram data of the analog source SBAS L5 signal is received, and the frame synchronization is all successful, which verifies the accuracy and achievability of the frame synchronization of the patent.

(2) 1200-bit Manchester decoding is about half the time of 2400-bit Manchester decoding. But when the 2400-bit Manchester decoding occurs when the first bit is discarded, the time will be extended from 280US to 425US.

(3) In the six frame synchronization process, it is best to find the frame header twice, which takes about 30MS. The worst is to find the frame header 4 times, which takes about 60MS.

(4) 600-bit Viterbi is about half the time of 1200-bit Viterbi.

(5) Viterbi decoding is the longest process in frame synchronization.

Through the above comparative analysis, it can be seen that the advantages of the present invention are:

(1) The CRC-based frame synchronization method proposed by the present invention only needs 2400-bit telegrams to complete frame synchronization, which is 3.5 seconds faster than traditional 6000-bit original data to achieve frame synchronization.

(2) The present invention proposes a frame synchronization method for SBAS L5 signals. Compared with the frame header matching algorithm of 6 characters, the algorithm flow is simpler, and the storage space and operation speed are improved by about 2 times.

(3) The present invention proposes a frame synchronization method for the SBAS L5 signal, and the probability of misjudgment is 1/224 close to 0, which meets engineering needs.

Claims (2)

1. A fast frame synchronization method for a satellite-based enhanced L5 signal, comprising the steps of:

1) data acquired by an SBAS L5 receiver through a carrier wave is subjected to Manchester decoding to generate first-time decoded data;

2) decoding the FEC by using a Viterbi decoding algorithm to generate original navigation message data after Manchester decoding;

3) finding the position or offset of the frame identifier in the original navigation message data; the method comprises the following steps:

step 3.1) obtaining 500 bit original data sequence after Viterbi decoding;

step 3.2) matching the nth 4-bit frame identification in the 500-bit original data sequence, and ending the step 3 if n is more than 6);

step 3.3) if the nth frame identification is found and the offset is less than 250, jumping to step 3.4), otherwise, adding 1 to the value of n and jumping to step 3.2);

step 3.4) performing CRC check judgment;

step 3.5) finding out a frame header if the CRC is correct, otherwise, skipping to the step 3.3) continuing to search the frame header in the rest original data;

step 3.6) calculating the position of the frame header in the 500-bit original data;

4) and adjusting the epoch count in the receiver according to the offset of the frame header in the original data to enable the epoch count of the initial position of one frame of the navigation message to be equal to 0.

2. The fast frame synchronization method for a satellite-based enhanced L5 signal according to claim 1, wherein: the step 4) comprises the following steps:

step 4.1) 2400 bit data is obtained, and Manchester decoding is carried out to obtain 1200 bit data;

step 4.2) carrying out Viterbi decoding on the 1200-bit data;

step 4.3) searching frame headers in the 600-bit data based on a CRC method;

step 4.4) if the frame header is found, jumping to step 4.8), otherwise, continuing searching;

step 4.5) processing 1200 dislocation, and carrying out VCP decoding;

step 4.6), jumping to step 4.8) if the frame header is found, otherwise, jumping to step 4.7); if the data bit search is the second time of negation, ending the step 4);

step 4.7) inverting the 2400-bit data, and repeating the steps 4.1) to 4.6);

step 4.8) completes the epoch adjustment to make the epoch count of the position where a frame starts 0.

Images