CN107689825A - A kind of spaceborne AIS collision signals separation method based on compressed sensing - Google Patents
A kind of spaceborne AIS collision signals separation method based on compressed sensing Download PDFInfo
- Publication number
- CN107689825A CN107689825A CN201610629623.4A CN201610629623A CN107689825A CN 107689825 A CN107689825 A CN 107689825A CN 201610629623 A CN201610629623 A CN 201610629623A CN 107689825 A CN107689825 A CN 107689825A
- Authority
- CN
- China
- Prior art keywords
- signal
- separation
- compressed sensing
- satellite
- ais
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/14—Relay systems
- H04B7/15—Active relay systems
- H04B7/185—Space-based or airborne stations; Stations for satellite systems
- H04B7/1851—Systems using a satellite or space-based relay
- H04B7/18515—Transmission equipment in satellites or space-based relays
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/0202—Channel estimation
Abstract
The present invention proposes a kind of spaceborne AIS collision signals separation method based on compressed sensing, by build based on compressed sensing it is deficient determine blind source signal separation model, isolate source signal in the reception signal of spaceborne AIS receivers using match tracing method；Signal is separated to N roads non-coherent demodulation is carried out by the method for 2 bit difference, the AIS data for then obtaining N number of ship by adjudicating decoding and sending.Compressed sensing technology is applied in spaceborne ais signal separation by the present invention first, and relies on compressed sensing characteristic, it is possible to reduce receiver antenna quantity, is reduced satellite load, can be separated signal parallel with promotion signal separative efficiency.
Description
Technical Field
The invention belongs to the technical field of space and navigation wireless communication, and particularly relates to a collision signal separation method of an automatic identification system of a satellite-borne ship based on a compressed sensing technology.
Background
In order to ensure safe sea navigation, improve shipping traffic efficiency, and realize real-time data exchange and target information identification between ships and between banks, the International Maritime Organization (IMO) in 2002 proposes an Automatic Identification System (AIS) for ships, and forces the ships to install AIS electronic communication equipment, wherein the AIS equipment performs broadcast type automatic reporting by adopting self-organizing time division multiple access (sodma) technology. The report content comprises dynamic information such as longitude and latitude, navigational speed, course and the like and static information such as a Marine Mobile Service Identification (MMSI) code, a call sign, a ship name and the like. Other ships can receive or forward the messages, and the base station can also receive the messages, so that a novel maritime communication navigation aid system of a shore ship, a ship bank and a ship is formed.
With the prosperous and prosperous sailing in the open sea, the satellite-borne AIS has come into play and is receiving more attention in the field of shipping. The satellite-borne AIS has a large coverage area and a wide visual field, high-quality and long-distance AIS signal transmission can be realized through transfer of the satellite to the transmitted and received signals, and technical support is provided for global ship navigation information acquisition and ship safety supervision. Satellite-borne AIS, by covering multiple independent AIS subnetworks, is likely to receive multiple conflicting AIS signals by the satellite, since there is no coordination mechanism between the subnetworks. The AIS signal separation and decoding in such interference background becomes a technical difficulty of the satellite-borne AIS receiver.
Compressed Sensing (CS) is a new sampling theory applicable to sparse signals, and the method acquires discrete samples of signals at a sampling frequency far less than Nyquist sampling frequency in a linear measurement mode, and then reconstructs the signals according to a certain algorithm by utilizing the sparse characteristics of the signals, thereby being widely applied to the field of image recovery and the field of voice signal separation.
The invention has the application number of 201410608054.6 and the invention name of 'a satellite borne AIS signal serial separation method based on parameter estimation', mainly aims at a mixed satellite borne AIS baseband signal, estimates parameters such as time delay, frequency offset and phase of the signal, performs signal reconstruction by using the information and a non-coherent detected code element, realizes serial separation of the rest signals by repeating the process after one path of signal is separated, and mainly analyzes the separation effect of a main signal, so that the method has the advantages of complex equipment and poor separation timeliness. The invention patent with the application number of 201410502475.0 and the name of 'satellite-borne AIS co-channel interference suppression method based on receiving blind beam forming' utilizes the constant modulus characteristic of AIS signals to realize blind beam forming on user signals through a constant modulus algorithm, and detects the desired signals after separating the desired signals from interference signals, however, the technology is difficult to separate when the arrival direction angles of collision signals are relatively different. The invention is a patent with application number of 201310492497.9 and title of invention is a signal processing method for solving time slot collision of AIS signals, which mainly aims at the background that the number of AIS signals with time slot collision is the same as the number of receiving antennas, separates signals received by multiple antennas (two or more) by performing phase processing, and only recovers the AIS signals needing separation. The invention has the application number of 201410608054.6, and the invention name of the invention is a collision signal processing method for a satellite-borne AIS system.
Disclosure of Invention
The invention provides a compression sensing-based satellite-borne AIS collision signal separation method, which applies a compression sensing technology to satellite-borne AIS signal separation for the first time.
In order to solve the technical problem, the invention provides a compression sensing-based satellite-borne AIS collision signal separation method, which comprises the steps of constructing a compression sensing-based underdetermined blind source signal separation model, and separating a source signal from a received signal of a satellite-borne AIS receiver by using a matching tracking method; the underdetermined blind source signal separation model is as follows:
s.t.y(t)＝AS(t)＝AΨα(t),t＝1,2,…,T
wherein T is the total sampling point number of an observation signal y (T) obtained by a collision signal received by the satellite-borne AIS receiver antenna through an analog-to-digital converter; t is a discrete time; a is a mixing matrix of size M × N; α (t) is a sparse coefficient; s (t) is a split source signal, i.e., a split signal, S (t) = Ψ α (t), M is an antenna array of the number of antennas of the satellite-borne AIS receiver, and N is the number of vessels.
The specific separation process is as follows:
step one, randomly generating an MxN mixing matrix according to the known observation signal y (t)As an estimate of the mixing matrix a;
step two, using a mixing matrixThe sparse coefficient α (t) is obtained by a matching pursuit method, and the separation signal is obtained by y (t) = AS (t) using the observation signal y (t)
Step three, calculating a separation signal according to the evaluation functionThe evaluation function is as follows:
wherein, the first and the second end of the pipe are connected with each other,is a fitness value;
median value Is a mixing matrixThe (j) th column vector of (a),is the jth row vector of the split signal;
intermediate value Is composed ofAverage value of (a);
step four, updating the mixing matrix by using a particle swarm algorithmThen returning to the step two until the updating times of the particle swarm algorithm reach the set maximum iteration times; and after the maximum iteration times are reached, the separation signal corresponding to the minimum fitness value is used as the separation signal which is finally output.
Furthermore, the N paths of separation signals are subjected to incoherent demodulation by a 2-bit difference method, and then AIS data sent by N ships are obtained through judgment decoding
Compared with the prior art, the invention has the obvious advantages that (1) the invention adopts a compression sensing-based satellite-borne AIS collision signal separation method as an underdetermined blind separation technology, works under the condition that the number of the antennas is less than or equal to the number of the source signals, reduces the number of the required antennas, and is beneficial to the reduction of load and the simplification of the receiver structure: (2) According to the method, the mixed matrix is estimated through the matching pursuit algorithm and the intelligent optimization algorithm, the recovery signals of N paths of source signals can be obtained in parallel at a time, and compared with a serial separation method which can only separate one path of signal at a time, the separation efficiency of the collision signals is improved; (3) By using a compressed sensing algorithm, frequency offset, phase shift and amplitude estimation are not needed, so that a modulation device for parameter estimation and local signal reconstruction is omitted, and the structure of the satellite-borne AIS receiver is simplified; (4) According to the invention, through the correction of the learning factor and the iteration number which are adjustable according to the busy degree of the marine transportation in the intelligent optimization algorithm used in the compressed sensing algorithm, the estimation precision of the mixed matrix can be obviously improved, the signal separation effect is improved, and the bit error performance of the detection is improved.
Drawings
Fig. 1 is an AIS signal transceiving system model according to an embodiment of the present invention.
Fig. 2 is a flow chart of source signal recovery by observing a signal matrix Y using a compressed sensing method.
Fig. 3 is a partially enlarged comparison graph of a source signal and a corresponding split signal when N =3 paths.
Fig. 4 is a graph of the bit error rate of an N =3 split main signal.
Detailed Description
It is easily understood that according to the technical solution of the present invention, without changing the essential spirit of the present invention, a person having ordinary skill in the art can imagine various embodiments of the present invention based on the compressed sensing on-board AIS collision signal separation method. Therefore, the following detailed description and the accompanying drawings are merely illustrative of the technical aspects of the present invention, and should not be construed as all of the present invention or as limitations or limitations on the technical aspects of the present invention.
Fig. 1 is an AIS signal transceiving system model according to an embodiment of the present invention, in which one ship transmits one AIS signal through one antenna, and N ships transmit N source signals S through respective antennas 0 (t) and then received by a satellite-borne AIS receiver having M receive antennas, M being less than N, which forms an underdetermined blind source separation model having the mathematical expression:
X 0 (t)＝A 0 S 0 (t)+n 0 (t),t＝1,2,…T
wherein T is the total sampling point number of observation signals obtained by a collision signal received by a receiving antenna of the satellite-borne AIS receiver through an analog-to-digital converter (ADC), T is discrete time, and X is 0 (t) is the M-dimensional observation signal of the known satellite borne AIS receiver, A 0 Is a column full rank mixing matrix formed by M multiplied by N unknown constants, S 0 (t) is the unknown N-dimensional source signal, N 0 (t) is additive noise.
The method aims to apply a compressed sensing method to an observation signal X 0 (t) performing separation to obtain a separated signal, i.e. an estimate of the source signal.
For a signal x that can be sparsely represented on an N × N orthogonal basis Ψ 0 Discrete signal x is measured by non-correlation 0 Projected onto a set of low-dimensional measurement vectors y 0 The upper part, namely:
y 0 ＝Φx 0 ＝ΦΨα 0
can be solved by 0 Optimization problem under norm:
s.t.y 0 ＝ΦΨα 0
completing observation of signal y from low dimensions 0 Reconstructing a signal x 0 。
The invention corresponds the underdetermined blind source separation model with the compressed sensing model, because X 0 (t) is dimensionally sparse so will observe informationNumber X 0 (t) and the low-dimensional observation signal y 0 Correspondingly, the model is expressed by y (T) in the compressed sensing-based underdetermined blind source signal separation model and represents an M multiplied by T observation signal. Source signal S 0 (t) can be represented by an orthogonal basis Ψ, and the source signal S 0 (t) and the signal x 0 Correspondingly, the under-determined blind source signal separation model based on compressed sensing is denoted by S (T) and represents the source signal of N multiplied by T. Further, S (T) can be expressed as a product of Ψ α (T), α (T) is an N × T sparse coefficient based on the orthogonal basis Ψ, and is a representation of the sparse coefficient in the compressive sensing model in the underdetermined blind source signal separation model based on compressive sensing, and sparsity is K. Column full rank mixing matrix A in the underdetermined blind source separation model 0 A is used for representing an M multiplied by N mixed matrix formed by the comprehensive action of M antennas and channels on signals in an underdetermined blind source signal separation model based on compressed sensing, and N is assumed that noise introduced by a sensor is negligible 0 (t) =0. Therefore, the underdetermined blind source signal separation model based on compressed sensing is constructed as follows:
s.t.y(t)＝AS(t)＝AΨα(t),t＝1,2,…,T
the method comprises the steps that T is the total sampling point number of observation signals obtained by an analog-to-digital converter (ADC) when a satellite-borne AIS receiver antenna receives collision signals, T is discrete time, separation of the collision signals can be completed by utilizing sparsity of the satellite-borne AIS signals in dimensionality through a matching tracking algorithm, N paths of separated signals are subjected to incoherent demodulation through a 2-bit difference method, and N pieces of original frame data sent by a ship are obtained through judgment decoding.
FIG. 2 is a flowchart of separating conflicting signals using a compressive sensing method, which is described in detail as follows:
step one, randomly generating an MXN mixing matrix according to the known observation signal y (t)As a pair of mixingEstimating a matrix A;
step two, using a mixing matrixThe sparse coefficient α (t) is determined by a matching pursuit Method (MP), and the separation signal is determined by y (t) = AS (t) using the observation signal y (t)
Step three, calculating the separation signal obtained in the step two according to the evaluation functionThe obtained fitness value and the corresponding sparse coefficient alpha (t) and the separation signal are recorded(after entering iteration, the minimum one of the fitness values corresponding to all particles in each iteration is recorded, and the corresponding sparse coefficient alpha (t) and the separation signal thereof
The merit function is as follows:
wherein, the first and the second end of the pipe are connected with each other,is a fitness value;
wherein the middle value
In the formulaIs a mixing matrixThe (j) th column vector of (a),to separate the jth row vector of the signal (i.e., the estimate of the jth source signal), since the source signals are sparse, only one source signal plays a major role at most sampling instants, asThe situation of stronger signal and weaker signal comes from, when the source signal is sufficiently sparse,for signals that are approximately sparse, the above equation may also be used as a constraint.
Wherein the middle value
In the formula (I), the compound is shown in the specification,is composed ofIs the correlation coefficient between the separate signals whenWhen the temperature of the water is higher than the set temperature,the minimum value of 0 is obtained. In the invention, the larger the signal-to-interference ratio among N paths of source signals and the larger the time delay among the N paths of source signals, the better the sparsity of the signals,the smaller. The AIS source signals transmitted by the ship are independent of each other, so thatThe smaller the resulting split signal proves to be the more likely to be closer to the source signal. So that the evaluation functionThe smaller the value, the better the separation.
Step four, updating the mixing matrix by using a particle swarm algorithmThen returning to the step two until the updating times of the particle swarm algorithm reach the set maximum iteration times; output minimum after reaching maximum iteration timesValue-corresponding sparse coefficient alpha (t) and separation signalThey are the optimal sparse coefficients α best (t) and optimal separation signalThe split signal is the source signal.
The practical application process of the method comprises the following steps:
step 1, setting an antenna array with M mutually independent antennas, wherein M is usually more than or equal to 2, namely more than or equal to 2 times of sparsity in source signal dimension.
And 2, after the antenna receives the conflict signal, sampling by an analog-to-digital converter (ADC) to form M channels of received data, sampling T points by each channel of data, and forming an M multiplied by T observation signal matrix Y.
And 3, performing blind source signal separation on the observation signal matrix Y by utilizing a compressed sensing technology through the estimation of the mixed matrix A to obtain estimation signals (namely separation signals) of N paths of source signals.
And 4, respectively carrying out intermediate frequency filtering on the obtained N paths of separation signals, and demodulating by adopting a 2-bit difference method.
And 5, respectively carrying out sampling judgment on the obtained N paths of demodulation signals to obtain N AIS data frames.
Fig. 3 shows that under the simulation condition of N =3, the amplitude ratio of the source signal arriving at the satellite-borne AIS receiver antenna is 1:0.5, the time delay of the two paths of signals after 0.3 is 6.25ms and 6.25ms relative to the previous path of signal, when the signal-to-noise ratio is 20dB, the waveforms of the two paths of signals are very similar to the local comparison graph of the separated signal with the maximum amplitude and the corresponding source signal, the distortion is small, the separated waveforms can be better demodulated and decoded, and the AIS collision signal can be seen to be better separated from the main signal through a satellite-borne AIS collision signal separation method based on compressed sensing.
Fig. 4 is N =3, when the time delay of the two subsequent signals in the signals reaching the satellite-borne AIS receiver antenna is set to 6.25ms and 6.25ms relative to the previous signal, the separated signals are subjected to 2-bit differential demodulation and decoding, and then an error rate curve is drawn according to the error rate condition of the main signal, where A1/A2 represents the amplitude ratio of the maximum interference signal to the main signal, the amplitude A3 of the third signal is a random number greater than 0 and smaller than A2, and the abscissa is the normalized signal-to-noise ratio. As can be seen from the figure, when A1/A2 is close to 0.5, satisfactory error code performance in the satellite borne AIS collision resolution field can be obtained, and when A1/A2 is close to 0.5&After 0.5, the bit error rate of 10 can be obtained under the condition of more than or equal to 8dB signal-to-noise ratio -4 The following error code performance is excellent.
According to the method, the AIS collision signal can be separated well and the error code performance is good.
Claims (3)
1. A satellite-borne AIS collision signal separation method based on compressed sensing is characterized in that an underdetermined blind source signal separation model based on compressed sensing is constructed, and a source signal is separated from a received signal of a satellite-borne AIS receiver by using a matching tracking method; the underdetermined blind source signal separation model is as follows:
s.t.y(t)＝AS(t)＝AΨα(t),t＝1,2,…,T
wherein T is the total sampling point number of an observation signal y (T) obtained by a collision signal received by the satellite-borne AIS receiver antenna through an analog-to-digital converter; t is a discrete time; a is a mixing matrix of size M × N; α (t) is a sparse coefficient; s (t) is a split source signal, i.e., a split signal, S (t) = Ψ α (t), M is an antenna array of the number of antennas of the satellite-borne AIS receiver, and N is the number of vessels.
2. The compressed sensing-based satellite-borne AIS collision signal separation method according to claim 1, characterized in that,
step one, randomly generating an MxN mixing matrix according to the known observation signal y (t)As an estimate of the mixing matrix a;
step two, using a mixing matrixThe sparse coefficient α (t) is obtained by a matching pursuit method, and the separated signal is obtained by y (t) = AS (t) using the observation signal y (t)
Step three, calculating a separation signal according to the evaluation functionThe evaluation function is as follows:
wherein the content of the first and second substances,is a fitness value;
intermediate value As a mixing matrixThe (j) th column vector of (a),is the jth row vector of the split signal;
intermediate value Is composed ofAverage value of (a);
step four, updating the mixing matrix by using a particle swarm algorithmThen returning to the step two until the updating times of the particle swarm algorithm reach the set maximum iteration times; and after the maximum iteration times are reached, the separation signal corresponding to the minimum fitness value is used as the separation signal which is finally output.
3. The compressed sensing-based satellite-borne AIS collision signal separation method according to claim 2, wherein the N paths of separation signals are subjected to incoherent demodulation by a 2-bit difference method, and then AIS data sent by N ships are obtained through decision decoding.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610629623.4A CN107689825B (en) | 2016-08-03 | 2016-08-03 | Satellite-borne AIS (automatic identification System) collision signal separation method based on compressed sensing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610629623.4A CN107689825B (en) | 2016-08-03 | 2016-08-03 | Satellite-borne AIS (automatic identification System) collision signal separation method based on compressed sensing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107689825A true CN107689825A (en) | 2018-02-13 |
| CN107689825B CN107689825B (en) | 2020-05-22 |
Family
ID=61151331
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610629623.4A Active CN107689825B (en) | 2016-08-03 | 2016-08-03 | Satellite-borne AIS (automatic identification System) collision signal separation method based on compressed sensing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107689825B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111521968A (en) * | 2020-05-22 | 2020-08-11 | 南京理工大学 | Underdetermined DOA estimation method based on target space diversity |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100208941A1 (en) * | 2009-02-13 | 2010-08-19 | Broaddus Christopher P | Active coordinated tracking for multi-camera systems |
| CN103226196A (en) * | 2013-05-17 | 2013-07-31 | 重庆大学 | Radar target recognition method based on sparse feature |
-
2016
- 2016-08-03 CN CN201610629623.4A patent/CN107689825B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100208941A1 (en) * | 2009-02-13 | 2010-08-19 | Broaddus Christopher P | Active coordinated tracking for multi-camera systems |
| CN103226196A (en) * | 2013-05-17 | 2013-07-31 | 重庆大学 | Radar target recognition method based on sparse feature |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111521968A (en) * | 2020-05-22 | 2020-08-11 | 南京理工大学 | Underdetermined DOA estimation method based on target space diversity |
| CN111521968B (en) * | 2020-05-22 | 2022-05-20 | 南京理工大学 | Underdetermined DOA estimation method based on target space diversity |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107689825B (en) | 2020-05-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sharma et al. | Application of compressive sensing in cognitive radio communications: A survey | |
| KR100663525B1 (en) | Interference power measurement apparatus and method required space-time beam forming | |
| CN108574525B (en) | Satellite-borne AIS collision signal separation method based on adaptive moment estimation | |
| US8681902B2 (en) | Method and apparatus for low complexity robust reconstruction of noisy signals | |
| CN101960757B (en) | Channel information prediction system and channel information prediction method | |
| Zhou et al. | Multi-user LEO-satellite receiver for robust space detection of AIS messages | |
| CN104283825A (en) | Channel estimation method based on dynamic compression sensing | |
| CN111030952A (en) | Beam space channel estimation method and system of millimeter wave system | |
| CN106899316A (en) | Multiple antennas space-based AIS receivers | |
| Guo et al. | Deep learning for joint channel estimation and feedback in massive MIMO systems | |
| CN113595941A (en) | Deep learning compressed sensing large-scale MIMO channel estimation method and system | |
| US7706768B2 (en) | Diversity switching | |
| CN107689825B (en) | Satellite-borne AIS (automatic identification System) collision signal separation method based on compressed sensing | |
| CN110060510B (en) | Satellite-borne AIS collision signal separation method based on improved artificial bee colony algorithm | |
| CN104994041B (en) | Three-dimensional extensive Bo Dabo digression combined estimation method of the antenna system based on pilot tone | |
| US11005507B2 (en) | Targeted ratio of signal power to interference plus noise power for enhancement of a multi-user detection receiver | |
| CN111865844B (en) | Channel estimation method and device for large-scale MIMO full-duplex relay system | |
| CN110113281B (en) | Method for realizing space division multiplexing by multi-system FSK noncoherent detection in MIMO communication | |
| CN114978382A (en) | Data processing method and communication device based on space division | |
| Arunkumar et al. | Soft symbol decoding in sweep-spread-carrier underwater acoustic communications: a novel variational Bayesian algorithm and its analysis | |
| Sandra et al. | Measurement-Based Wideband Maritime Channel Characterization | |
| CN115022134A (en) | Millimeter wave large-scale MIMO system channel estimation method and system based on non-iterative reconstruction network | |
| Zhang et al. | Data-Driven Co-Channel Signal Interference Elimination Algorithm for Terrestrial-Satellite Communications and Broadcasting | |
| Kaplan et al. | A message transmission scheme for linear time-varying multipath channels | |
| CN105974413A (en) | Self-focusing method of multi-base outer radiation source radar imaging system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |