Satellite-borne ADS-B signal low signal-to-noise ratio receiving method

Abstract

The invention provides a satellite-borne ADS-B (Automatic Dependent Surveillance-Broadcast) signal low signal-to-noise ratio receiving method. The method comprises the following steps: inputting a data cache module; a frequency offset estimation and compensation module; a matched filtering module; matching a data caching module; a phase synchronization module; an amplitude estimation module; a bipolar conversion module; a correlation operation module; a peak detection module; a bit judgment module; a confidence ranking module; and a CRC violent decoding module. The invention provides a low-complexity ADS-B (Automatic Dependent Surveillance-Broadcast) coherent demodulation method by utilizing the spectral characteristics of an ADS-B signal. According to the invention, a specially designed block processing structure is adopted, and burst frame detection is carried out after coherent demodulation, so that the frame header detection performance is improved. According to the method, CRC violent decoding is combined, and finally, when Eb / No is equal to 6.5 dB, the frame accuracy exceeds 90%.

Description

Technical Field

[0001] The invention belongs to the technical field of satellite communications, relates to a satellite-borne ADS-B signal, and in particular to a satellite-borne ADS-B signal low signal-to-noise ratio receiving method. Background Art

[0002] The Automatic Dependent Surveillance-Broadcast System (ADS-B) is the main air traffic control surveillance technology identified by ICAO in the future. It uses an open broadcast method to periodically broadcast information such as the target's own position and speed. On the one hand, it is convenient for the air traffic control center to no longer rely solely on radar for control and surveillance, and on the other hand, other aircraft can also receive signals, providing effective data support for situational awareness and aviation safety. However, due to the limitation of ground stations, the monitoring range is generally limited to land and coastal areas.

[0003] Satellite-based ADS-B refers to the use of ADS-B receivers on satellites to expand the monitoring range to places that traditional ground-based equipment cannot cover, such as oceans, polar regions, and remote areas, to achieve global coverage. The Iridium NEXT satellite constellation is equipped with ADS-B payloads to expand global applications. The Micro-Nano Satellite Engineering Center of the National University of Defense Technology launched the "Tiantuo-3" satellite in September 2015. The ADS-B receiver it carries has a sensitivity of -93dBm through low-noise design and narrow bandwidth.

[0004] Due to the long distance between the satellite and the aircraft, the satellite-based ADS-B receiver has a weak signal and a low signal-to-noise ratio. According to calculations, the minimum processing power required is -102dBm signal, which is converted to E b / N o =7dB, the frame correct reception rate (PCR) needs to reach more than 90%. Land-based ADS-B receivers can only process signals with a minimum power of -90dBm, and generally use non-coherent demodulation, and the demodulation performance cannot meet the requirements of satellite-borne. Therefore, it is necessary to study high-sensitivity satellite-borne ADS-B reception methods.

[0005] The coherent demodulation of ADS-B signals needs to solve two problems: on the one hand, due to the Doppler effect, the low-orbit satellite link has a large frequency deviation (about ±40kHz), and the use of coherent demodulation to improve sensitivity requires carrier synchronization, including frequency deviation and phase deviation estimation and compensation; on the other hand, the ADS-B signal is a burst frame structure, and the start of the burst frame needs to be accurately detected before carrier synchronization and brute force decoding can be performed. The ADS-B coherent demodulation method currently proposed by the industry is generally more complex to implement, and uses incoherent burst frame detection, followed by carrier synchronization, which reduces the frame header detection performance. Summary of the invention

[0006] In view of the shortcomings of the prior art, the object of the present invention is to provide a low signal-to-noise ratio receiving method for satellite-borne ADS-B signals, so as to solve the technical problem that the satellite-borne ADS-B signal receiving method in the prior art is difficult to balance the complexity and frame header detection performance.

[0007] In order to solve the above technical problems, the present invention adopts the following technical solutions to achieve the above problems:

[0008] A method for receiving a satellite-borne ADS-B signal with a low signal-to-noise ratio, the method comprising the following steps:

[0009] Step 1: Input data cache module:

[0010] For the complex signal r[n] that is digitally down-converted to zero intermediate frequency, the sampled data is first cached and then read out according to the sliding window. The output length each time is (α+β)*N s The sampled data block.

[0011] Where:

[0012] n represents the sampling point number;

[0013] N s Indicates the number of sampling points for each ADS-B symbol;

[0014] α*N s Indicates the sliding window size;

[0015] β*N s Indicates the amount of overlap between blocks;

[0016] Step 2: Frequency offset estimation and compensation module:

[0017] For each sample data block obtained in step 1, perform a frequency offset estimation to obtain the frequency offset estimation value f cw , and perform frequency offset compensation on the entire data block to obtain the compensated sampling signal

[0018] Step 3, matched filter module:

[0019] Using N s / 2 point square wave is used as a matched filter to compensate the sampled signal in step 2 Perform filtering to obtain the matched filter output signal u[n].

[0020] Step 4: Matching data cache module:

[0021] The matched filter output signal u[n] of step 3 is stored in the matching data cache, and the address T num =0.

[0022] Step 5, Phase Synchronization Module:

[0023] From the address T in step 4 num Start reading 120*N from the matching data cache continuously s sampling points, perform phase estimation and compensation in segments, and then output a phase-synchronized unipolar signal v[n].

[0024] Step 6, amplitude estimation module:

[0025] Utilize phase synchronization to output the end of δ*N s The sampled data is averaged to obtain the signal amplitude estimate Where δ represents N s Multiples of; N s Indicates the number of sampling points for each ADS-B symbol.

[0026] Step 7, bipolar conversion module:

[0027] Using the amplitude estimate obtained in step 6 Then the unipolar signal v[n] output in step 5 is converted into a bipolar signal w[n],

[0028] Step 8, related operation modules:

[0029] N s / 2 is the interval to extract 26 sampling points, and then compare them with the local frame header v in the bipolar signal w[n] obtained in step 7 uw Perform correlation operation and obtain the correlation value x[T num ].

[0030] Step 9, peak detection module:

[0031] If the correlation value x[T num ], if the address T num =α*N s -1, then search for the peak value from the correlation value x[n] obtained in step 8, where 0≤n≤(α*N s -1)) to get the position ρ corresponding to the peak; then, (ρ+8*N s ) is the starting address, N s / 2 is the interval, 224 sampling points are read out from the matching data buffer as the detected ADS-B burst frame y[n], where 0≤n≤223. Otherwise, address T num =T num +1, from address T num Start reading 120*N from the matching data cache continuously s sampling points, skip to step 4.

[0032] Step 10, bit decision module:

[0033] For each ADS-B symbol, bit decision and confidence calculation are performed based on the first half symbol and the second half symbol values ​​to obtain the decision bit z[n] and confidence p[n], where 0≤n≤111.

[0034] Step 11, confidence ranking module:

[0035] The confidence levels p[n] obtained in step 10 are sorted in monotonically increasing order to obtain the confidence level sequence q[n] and its position sequence π[n] in the original sequence p[n].

[0036] Step 12, CRC brute force decoding module:

[0037] Select the N bits with the lowest confidence, and select all combinations of 0 to 5 bits from them, calculate the CRC value after bit flipping, and use the bit sequence with CRC zero as the decoded ADS-B message d[n], where 0≤n≤111.

[0038] Compared with the prior art, the present invention has the following technical effects:

[0039] (I) The present invention utilizes the spectrum characteristics of ADS-B signals and proposes a low-complexity ADS-B coherent demodulation method.

[0040] (II) The present invention adopts a specially designed block processing structure and places burst frame detection after coherent demodulation, thereby improving frame header detection performance.

[0041] (III) The present invention combines CRC brute force decoding to achieve the final result in E b / N o When it is equal to 6.5dB, the frame accuracy rate exceeds 90%. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 Schematic diagram of the ADS-B signal burst frame structure.

[0043] Figure 2 Schematic diagram of the architecture for receiving and processing ADS-B signals with low signal-to-noise ratio.

[0044] Figure 3 Outputs a block diagram for the input data cache.

[0045] Figure 4 Schematic diagram of carrier phase segment tracking and compensation processing.

[0046] Figure 5 This is a schematic diagram of the output signal of the phase synchronization module.

[0047] Figure 6This is a schematic diagram of the output signal of the bipolar conversion module.

[0048] The specific contents of the present invention are further explained in detail below in conjunction with embodiments. DETAILED DESCRIPTION

[0049] It should be noted that, unless otherwise specified, all devices and techniques in the present invention are devices and techniques known in the prior art.

[0050] 1090ES is an ADS-B communication protocol based on the secondary surveillance radar S-mode transponder. It is an expansion of the 56-bit traditional S-mode short message. The operating frequency of the 1090ES data link is 1090MHz, the code element width is 1us, and the data transmission rate is 1Mb / s. A complete 1090ES ADS-B signal lasts for 120us and consists of two parts: an 8us header and a 112us data segment. The header consists of four 0.5us wide rectangular pulses located at 0us, 1us, 3.5us and 4.5us respectively. The data segment module contains a total of 112 bits of information, which is encoded using pulse position modulation (PPM). Specifically, each 1us bit time is divided into two 0.5us chips. If the high level appears in the previous chip, it means bit "1", and if the high level appears in the latter chip, it means bit "0".

[0051] The 112 bits of information contained in the data segment can be divided into five parts: DF field, CA field, AA field, ME field and PI field. For the ADS-B signal transmitted by the S-mode transponder, DF is set to 17 (binary 10001), the CA field represents the capability and status of the transmitter, the AA field is the globally unique address assigned by ICAO to each aircraft, and the 56-bit ME field contains the aircraft's location, speed, heading, intention and other flight information. The last 24 bits are the cyclic redundancy check (CRC) code, which is used to detect whether the data transmission is wrong. The ADS-B signal burst frame structure is as follows: Figure 1 shown.

[0052] From a theoretical analysis, the sensitivity equation of the ADS-B receiver is:

[0053] S min =-174(dBm / Hz)+NF(dB)+E b / N o (dB)+10lgR b (bps)

[0054] Where:

[0055] S min Indicates the sensitivity of the ADS-B receiver;

[0056] NF represents the noise factor of the receiver, which is generally equal to 2dB in engineering;

[0057] E b Represents binary bit energy;

[0058] N o represents the noise power spectral density;

[0059] E b / N o Represents the ratio of binary bit energy to noise power spectral density;

[0060] R b Indicates the equivalent bit rate.

[0061] E b / N o Corresponding to the signal-to-noise ratio when PCR is 90%. Considering that the ADS-B signal is PPM modulated, the pulse width is equal to 0.5us, and the sensitivity is characterized by the peak power, the equivalent bit rate R b Therefore, if the satellite-borne ADS-B receiver sensitivity is to reach -102dBm, the demodulation threshold E b / N o It needs to be less than or equal to 7dB, and the key points are carrier synchronization and frame header detection.

[0062] Specific embodiments of the present invention are given below. It should be noted that the present invention is not limited to the following specific embodiments, and all equivalent changes made on the basis of the technical solution of this application fall within the protection scope of the present invention.

[0063] Example:

[0064] This embodiment provides a method for receiving a satellite-borne ADS-B signal with a low signal-to-noise ratio. Figure 2 As shown, the method comprises the following steps:

[0065] Step 1: Input data cache module:

[0066] For the complex signal r[n] that is digitally down-converted to zero intermediate frequency, the sampled data is first cached and then read out according to the sliding window, such as Figure 3 As shown, the output length each time is (α+β)*N s The sampled data block.

[0067] Where:

[0068] n represents the sampling point number;

[0069] N s Indicates the number of sampling points for each ADS-B symbol;

[0070] α*N s Indicates the sliding window size;

[0071] β*N s Indicates the amount of overlap between blocks;

[0072] Step 2: Frequency offset estimation and compensation module:

[0073] For each sample data block obtained in step 1, perform a frequency offset estimation to obtain the frequency offset estimation value f cw , and perform frequency offset compensation on the entire data block to obtain the compensated sampling signal

[0074] In this step, the ADS-B signal is regarded as an approximate single-carrier signal, and the frequency offset is estimated by delay correlation of the signal, and then the frequency offset is compensated for the data.

[0075] In step 2, the frequency offset estimate f cw for:

[0076]

[0077] Where:

[0078] d represents the delay value;

[0079] k represents the sampling point number used for parameter estimation;

[0080] L represents the number of sampling points used for parameter estimation;

[0081] r[] represents a complex signal;

[0082] arg[] represents the complex number argument operation function;

[0083] conj() represents the conjugate operation function of complex numbers;

[0084] start indicates the starting position.

[0085] Specifically, in this embodiment, α=80, β=40, N s = 8. The sliding window outputs a sampling data block of length 960 each time, and the last 320 sampling points are used for frequency offset estimation, that is, start = 640, d = 8, L = 320.

[0086] Step 3, matched filter module:

[0087] Using N s / 2 point square wave is used as a matched filter to compensate the sampled signal in step 2 Perform filtering to obtain the matched filter output signal u[n].

[0088] Step 4: Matching data cache module:

[0089] The matched filter output signal u[n] of step 3 is stored in the matching data cache, and the address T num =0.

[0090] Step 5, Phase Synchronization Module:

[0091] From the address T in step 4 num Start reading 120*N from the matching data cache continuously s sampling points, perform phase estimation and compensation in segments, and then output the phase-synchronized (real part) unipolar signal v[n].

[0092] In this step, for the matched filter output signal u[n], due to the error in frequency offset estimation, the residual frequency offset will cause a linear increase in the carrier phase, and the carrier phase needs to be tracked during the demodulation process. Here, the segmented phase estimation and compensation technology is used. Each segment contains block sampling points, and the accumulated phase error Δθ in the estimated segment is k , and phase compensation is performed on this segment of data to complete carrier phase synchronization. Figure 4 shown.

[0093] In this step, since the ADS-B signal is regarded as an approximate single-carrier signal, demodulation processing is not required during carrier tracking, but maximum likelihood estimation is performed directly.

[0094] In this embodiment, the phase synchronization module output signal schematic diagram is as follows: Figure 5 shown.

[0095] In step 5, when the phase estimation is performed segment by segment, the phase estimation of the kth segment is:

[0096]

[0097] Where:

[0098] Δθ k represents the phase deviation of the kth segment;

[0099] n represents the serial number of the sampling point;

[0100] block indicates the number of sampling points;

[0101] k represents the sequence number of the segment;

[0102] u[] represents the matched filter output signal;

[0103] In this embodiment, block=160 is set. Using the estimated phase deviation Δθ k, phase compensation is performed on the 160 sampling points in the kth segment, thereby completing the phase error correction.

[0104] Step 6, amplitude estimation module:

[0105] Utilize phase synchronization to output the end of δ*N s The sampled data is averaged to obtain the signal amplitude estimate ; where δ represents N s Multiples of; N s Indicates the number of sampling points for each ADS-B symbol.

[0106] In this embodiment, the coherent detection includes steps 6 to 9, namely, an amplitude estimation module, a bipolar conversion module, a correlation operation module and a peak detection module. Each time a phase synchronous output v[n] sampling data block is received, a coherent detection is performed.

[0107] In step 6, for the T num The unipolar signal v[n] of samples is averaged by the last 320 samples to get the signal amplitude estimate

[0108] Step 7, bipolar conversion module:

[0109] Using the amplitude estimate obtained in step 6 Then the unipolar signal v[n] output in step 5 is converted into a bipolar signal w[n],

[0110] In this embodiment, the schematic diagram of the output signal of the bipolar conversion module is as follows: Figure 6 shown.

[0111] Step 8, related operation modules:

[0112] N s / 2 is the interval to extract 26 sampling points, and then compare them with the local frame header v in the bipolar signal w[n] obtained in step 7 uw Perform correlation operation and obtain the correlation value x[T num ].

[0113] In step 8, the relevant value

[0114] Where:

[0115] n represents the sampling point number;

[0116] w[] represents a bipolar signal;

[0117] v uw=[1,0,1,0,0,0,0,1,0,1,0,0,0,0,0,0,1,0,0,1,0,0,1,0,1,0,1,1,0].

[0118] Step 9, peak detection module:

[0119] If the correlation value x[T num ], if the address T num =α*N s -1, then search for the peak value from the correlation value x[n] obtained in step 8, where 0≤n≤(α*N s -1)) to get the position ρ corresponding to the peak. Then, (ρ+8*N s ) is the starting address, N s / 2 is the interval, 224 sampling points are read out from the matching data cache as the detected ADS-B burst frame y[n], where 0≤n≤223; otherwise, address T num =T num +1, from address T num Start reading 120*N from the matching data cache continuously s sampling points, skip to step 4.

[0120] Step 10, bit decision module:

[0121] For each ADS-B symbol, bit decision and confidence calculation are performed based on the first half symbol and the second half symbol values ​​to obtain the decision bit z[n] and confidence p[n], where 0≤n≤111.

[0122] Step 11, confidence ranking module:

[0123] The confidence levels p[n] obtained in step 10 are sorted in monotonically increasing order to obtain the confidence level sequence q[n] and its position sequence π[n] in the original sequence p[n].

[0124] Step 12, CRC brute force decoding module:

[0125] Select the N bits with the lowest confidence, and select all combinations of 0 to 5 bits from them, calculate the CRC (Cyclic Redundancy Check) value after bit flipping, and use the bit sequence with CRC zero as the decoded ADS-B message d[n], where 0≤n≤111.

Claims

1. A method for receiving satellite-borne ADS-B signals with low signal-to-noise ratio, characterized in that: The method comprises the following steps: Step 1: input data cache module; Step 2: frequency offset estimation and compensation module; Step 3, matched filter module; Step 4: Matching data cache module; Step 5, phase synchronization module; Step six, amplitude estimation module; Step 7, bipolar conversion module; Step 8, related operation module; Step nine, peak detection module; Step 10, bit decision module; Step 11, confidence ranking module; Step 12: CRC brute force decoding module.

2. The low signal-to-noise ratio receiving method for satellite-borne ADS-B signals according to claim 1, characterized in that: The method comprises the following steps: Step 1: Input data cache module: For the complex signal r[n] that is digitally down-converted to zero intermediate frequency, the sampled data is first cached and then read out according to the sliding window. The output length each time is (α+β)*N s The sampling data block; Where: n represents the sampling point number; N s Indicates the number of sampling points for each ADS-B symbol; α*N s Indicates the sliding window size; β*N s Indicates the amount of overlap between blocks; Step 2: Frequency offset estimation and compensation module: For each sample data block obtained in step 1, perform a frequency offset estimation to obtain the frequency offset estimation value f cw , and perform frequency offset compensation on the entire data block to obtain the compensated sampling signal Step 3, matched filter module: Using N s / 2 point square wave is used as a matched filter to compensate the sampled signal in step 2 Perform filtering to obtain a matched filter output signal u[n]; Step 4: Matching data cache module: The matched filter output signal u[n] of step 3 is stored in the matching data cache, and the address T num =0; Step 5, Phase Synchronization Module: From the address T in step 4 num Start reading 120*N from the matching data cache continuously s Sampling points, segmented phase estimation and compensation, and then output the phase-synchronized unipolar signal v[n]; Step 6, amplitude estimation module: Utilize phase synchronization to output the end of δ*N s The sampled data is averaged to obtain the signal amplitude estimate Where δ represents N s Multiples of; N s Indicates the number of sampling points for each ADS-B symbol; Step 7, bipolar conversion module: Using the amplitude estimate obtained in step 6 Then the unipolar signal v[n] output in step 5 is converted into a bipolar signal w[n], Step 8, related operation modules: N s / 2 is the interval to extract 26 sampling points, and then compare them with the local frame header v in the bipolar signal w[n] obtained in step 7 uw Perform correlation operation and obtain the correlation value x[T num ]; Step 9, peak detection module: If the correlation value x[T num ], if the address T num =α*N s -1, then search for the peak value from the correlation value x[n] obtained in step 8, where 0≤n≤(α*N s -1)) to get the position ρ corresponding to the peak; then, (ρ+8*N s ) is the starting address, N s / 2 is the interval, 224 sampling points are read out from the matching data cache as the detected ADS-B burst frame y[n], where 0≤n≤223; otherwise, address T num =t num +1, from address T num Start reading 120*N from the matching data cache continuously s sampling points, jump to step 4; Step 10, bit decision module: For each ADS-B symbol, bit decision and confidence calculation are performed based on the first half symbol and the second half symbol values ​​to obtain the decision bit z[n] and confidence p[n], where 0≤n≤111; Step 11, confidence ranking module: The confidence levels p[n] obtained in step 10 are sorted in monotonically increasing order to obtain the confidence level sequence q[n] and its position sequence π[n] in the original sequence p[n]; Step 12, CRC brute force decoding module: Select the N bits with the lowest confidence, and select all combinations of 0 to 5 bits from them, calculate the CRC value after bit flipping, and use the bit sequence with CRC zero as the decoded ADS-B message d[n], where 0≤n≤111.

3. The low signal-to-noise ratio receiving method for satellite-borne ADS-B signals according to claim 2, characterized in that: The method comprises the following steps: In step 2, the frequency offset estimate f cw for: Where: d represents the delay value; k represents the sampling point number used for parameter estimation; L represents the number of sampling points used for parameter estimation; r[] represents a complex signal; arg[] represents the complex number argument operation function; conj() represents the conjugate operation function of complex numbers; start indicates the starting position.

4. The low signal-to-noise ratio receiving method for satellite-borne ADS-B signals according to claim 2, characterized in that: In step 5, when the phase estimation is performed segment by segment, the phase estimation of the kth segment is: Where: Δθ k represents the phase deviation of the kth segment; n represents the serial number of the sampling point; block indicates the number of sampling points; k represents the sequence number of the segment; u[] represents the matched filter output signal.

5. The low signal-to-noise ratio receiving method for satellite-borne ADS-B signals according to claim 2, characterized in that: In step 6, for the T num The unipolar signal v[n] of samples is averaged by the last 320 samples to get the signal amplitude estimate 6. The low signal-to-noise ratio receiving method for satellite-borne ADS-B signals according to claim 2, characterized in that: In step 8, the relevant value Where: n represents the sampling point number; w[] represents a bipolar signal; v uw ＝[1,0,1,0,0,0,0,1,0,1,0,0,0,0,0,0,1,0,0,1,0,1,0,1,1,0]。

Images