CN117233462A - Photon-assisted distributed compressive sampling system and implementation method thereof - Google Patents

Abstract

The invention discloses a photon-assisted distributed compression sampling system, which comprises a multi-wavelength light source, a first wavelength division multiplexer in a far-end node, a first electro-optic modulator, a second wavelength division multiplexer, an erbium-doped fiber amplifier (EDFA), a wavelength division demultiplexer in a central site, a second electro-optic modulator, a photoelectric detector, a low-pass filter, a sampler, a joint reconstruction calculation module, an antenna connected with the first electro-optic modulator in the far-end node and a pseudo-random sequence generator connected with the second electro-optic modulator in the central site. Aiming at the monitoring requirement of the distributed broadband signal spectrum, the invention uses the optical carrier radio frequency mode to pull back a plurality of remote node signals to the central station for processing, and simultaneously adopts the photon compression sampling technology and the combined reconstruction algorithm to realize the efficient compression sampling of the remote node signal spectrum at the central station.

Description

A photon-assisted distributed compressed sampling system and its implementation method

Technical field

The invention belongs to the technical field of microwave photon signal processing, and specifically relates to a photon-assisted distributed compression sampling system and its implementation method.

Background technique

Digital signal processing technology has many advantages such as flexibility, high speed, high precision and strong anti-interference ability, and has become a mainstream technology in the field of signal processing. Analog-to-digital converter, ADC, builds a bridge to represent and process natural signals with digital signals. In recent years, analog-to-digital conversion technology has continued to develop, and sampling rates have also continued to increase. However, for some systems with instantaneous bandwidth greater than 10 GHz, such as ultra-wideband communications and radar countermeasures systems, existing ADCs cannot meet the requirements. The compressed sampling theory was proposed by D.L. Donoho et al. in 2006. This theory believes that if the signal is sparse or can be sparsely represented in a certain domain, then it can be reconstructed at a sampling rate much lower than the Nyquist rate. Sparse signals. In 2009, J.A. Tropp and others proposed the use of a random demodulator model to achieve compressed sampling. The frequency sparse signal can be reconstructed through the structure of random mixing, filtering and low-speed sampling. However, compressed sampling systems based on stochastic demodulator models are limited by the bandwidth and performance of electrical random signals and electrical mixing devices. Photonics technology and devices have the characteristics of low loss, large bandwidth, strong anti-interference ability, and parallel processing, which have attracted widespread research interest. In 2011, J.M.Nichols and others used photonic links to implement the mixing function in compressed sampling, which improved the bandwidth of the compressed sampling system.

On the other hand, in order to make full use of the correlation between signals and within signals, Baron et al. proposed the distributed compressed sampling theory in the compressed sampling theory. In this theory, the system has multiple distributed nodes. The signals received by these nodes satisfy the joint sparse model. The sending end performs independent compression sampling on each node, while the receiving end uses the correlation of the signal through the distributed compression sampling algorithm. For joint reconstruction, distributed compression sampling requires fewer measurements and has higher reconstruction accuracy than the scenario where each sparse signal is reconstructed individually at the receiving end. Baron et al. also proposed three joint sparse models suitable for different scenarios, called JSM-1, JSM-2 and JSM-3 respectively. The common part of JSM-1, JSM-2 and JSM-3 requires that the frequency point position and amplitude of the spectrum sparse signal are the same. Later, Sundman et al. proposed a hybrid support set model. In the hybrid support set, the common parts only need to have the same frequency point position, but the amplitude can be different, which is more universal and applicable to practical scenarios. The distributed compression sampling system has unique advantages, and has been continuously studied since. However, only electronic devices are used for simulations and experiments, and its working bandwidth and frequency monitoring range are constrained by the performance of the electrical domain itself.

In view of the above technical problems, it needs to be improved. The following content describes the application situation where the signal conforms to the mixed support set model.

Contents of the invention

Based on the shortcomings in the existing technology, the present invention provides a photon-assisted distributed compressed sampling system and its implementation method. In response to the needs of distributed broadband signal spectrum monitoring, the optical carrier radio frequency method is used to pull the remote node signal back to the central station. processing, while using photon compression sampling technology and joint reconstruction algorithm at the central station to achieve efficient compression sampling of the remote node signal spectrum. It has the advantages of ultra-high bandwidth of photon compression sampling technology and optical fiber distributed long-distance transmission, and is expected to achieve long-distance transmission. Distance, multi-node, wide coverage space electromagnetic spectrum monitoring.

In order to solve the above technical problems, the technical solution of the present invention is:

A photon-assisted distributed compression sampling system, including a first wavelength division multiplexer in a remote node, a first electro-optical modulator, an antenna, a second wavelength division multiplexer, an erbium-doped fiber amplifier EDFA, and a central site Multi-wavelength light source, wavelength division multiplexer, second electro-optical modulator, pseudo-random sequence generator, photodetector, low-pass filter, sampler, and joint reconstruction calculation module.

An implementation method of a photon-assisted distributed compressed sampling system, including the following steps:

The multi-wavelength light source provides J wavelengths of continuous light, with wavelengths λ ₁ , λ ₂ ,..., until λ _J . The multi-wavelength optical carrier is transmitted clockwise through the single-mode fiber to the first remote node, where node, the optical carrier with wavelength λ ₁ is extracted through the first wavelength division multiplexer, while other optical carriers pass directly. The sparse signal of the frequency to be measured received by the antenna is modulated onto the optical carrier λ ₁ through the first electro-optical modulator, and then the modulated light signal is combined with other optical wavelengths by the second wavelength division multiplexer and continues to be transmitted on the optical fiber. In the optical fiber transmission line, an erbium-doped fiber amplifier EDFA is added to compensate for the loss of line optical power. Similarly, at the jth (j represents the wavelength sequence number, j ≤ J remote nodes), the sparse signal of the frequency to be measured is modulated onto the optical carrier λ _j through the first electro-optical modulator until the Jth remote node. The multi-wavelength optical carrier carrying J-channel radio frequency information is transmitted back to the central station via optical fiber. First, it is decomposed into J optical carriers through a wavelength division multiplexer. Each optical carrier carrying radio frequency information enters the second electro-optical modulator and is mixed with J different pseudo-random sequences generated by the pseudo-random sequence generator. The mixed signal of each channel is converted into an electrical signal through a photodetector, and then passes through a low-pass filter and a sampler to obtain J sets of compressed sampling measurement value sequences y _j [n]. Send y _j [n] to the signal joint reconstruction calculation module, and use the distributed compression sampling reconstruction algorithm to jointly restore the sparse signals of the measured frequencies of J remote nodes.

As a preferred solution of the present invention, in the implementation method of a photon-assisted distributed compression sampling system, the J frequency sparse signals received by J remote nodes are frequency domain sparse signals and satisfy the mixed support set model, To represent the correlation between signals and the signal itself.

As a preferred solution of the present invention, the repetition rate of the pseudo-random binary sequence emitted by the pseudo-random sequence generator in the implementation method of the photon-assisted distributed compressed sampling system must be greater than or equal to the Nyquanian frequency of the J-channel frequency sparse signal. ster frequency, which ensures that the signal after mixing the frequency sparse signal and the pseudo-random sequence retains all the information of the original sparse signal.

The invention has the following characteristics and beneficial effects:

1 In the present invention, the sparse signals of the frequency to be measured received by each node satisfy the hybrid support set model, and the compressed sampling results of the radio frequency signals carried by each wavelength are jointly restored by the joint reconstruction calculation module. Compared with single-node photon compression sampling that recovers signals individually, photon distributed compression sampling can achieve a higher sampling rate compression ratio.

2. The radio frequency signal to be measured in the present invention is transmitted back to the central station in the form of optical radio frequency, so that the system has the advantages of long distance and wide range coverage.

3. The present invention has the advantage of large bandwidth of photonics compressed sampling technology.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without exerting any creative effort.

Figure 1 is a schematic structural diagram of the photon-assisted distributed compressed sampling system provided by the present invention.

Figure 2 shows the original spectrum diagrams of two sets of signals that meet the conditions of the mixed support set and the signal spectrum diagrams recovered through single-node photon-assisted compression sampling.

Figure 3 shows two sets of original signal spectrum diagrams that meet the conditions of the hybrid support set and the signal spectrum diagram obtained through joint recovery by a photon-assisted distributed compression sampling system provided by the present invention.

Reference numbers in the figure: 1. Multi-wavelength light source, 2. Remote node, 3. First wavelength division multiplexer, 4. First electro-optical modulator, 5. Second wavelength division multiplexer, 6. Antenna, 7. Erbium-doped fiber amplifier EDFA, 8. Central site, 9. Wavelength division multiplexer, 10. Second electro-optical modulator, 11. Pseudo-random sequence generator, 12. Photodetector, 13. Low-pass filter , 14. Sampler, 15. Joint reconstruction calculation module.

Detailed ways

It should be noted that, as long as there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "upper", "lower", "front", "rear", "left", "right", " The orientations or positional relationships indicated by "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and The simplified description is not intended to indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore should not be construed as a limitation of the present invention. Furthermore, the terms “first”, “second”, etc. are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of indicated technical features. Thus, features defined by "first," "second," etc. may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise specified, "plurality" means two or more.

In the description of the present invention, it should be noted that, unless otherwise clearly stated and limited, the terms "installation", "connection" and "connection" should be understood in a broad sense. For example, it can be a fixed connection or a detachable connection. Connection, or integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be an internal connection between two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood through specific situations.

As shown in Figure 1, the present invention provides a photon-assisted distributed compression sampling system. As shown in Figure 1, it includes a first wavelength division multiplexer 3, a first electro-optical modulator 4, and a first wavelength division multiplexer 3 in a remote node 2. Two-wavelength division multiplexer 5, antenna 6, erbium-doped fiber amplifier EDFA7, multi-wavelength light source 1 in the central site 8, demultiplexer 9, second electro-optical modulator 10, pseudo-random sequence generator 11, photoelectric Detector 12, low-pass filter 13, sampler 14, joint reconstruction calculation module 15.

This embodiment also provides a method for implementing a photon-assisted distributed compressed sampling system, which includes the following steps:

Multi-wavelength light source 1 provides J wavelengths of continuous light with wavelengths λ ₁ , λ ₂ ,..., until λ _J . The multi-wavelength optical carrier is transmitted clockwise through single-mode optical fiber to the first remote node 2. At this node, the wavelength of λ ₁ is extracted through the first wavelength division multiplexer 3, and other optical carriers pass directly. The sparse signal of the frequency to be measured is received by the antenna 6 and modulated onto the optical carrier λ ₁ through the first electro-optical modulator 4. Then the modulated light signal is combined with other optical wavelengths by the second wavelength division multiplexer 5 and continues. Transmitted over optical fiber. In the optical fiber transmission line, an erbium-doped fiber amplifier 7 is added to compensate for the loss of line optical power. Similarly, at the jth (j represents the wavelength number, j≤J) remote node 2, the frequency sparse signal to be measured is modulated onto the optical carrier λ _j through the first electro-optical modulator 4, until the jth remote node 2 .

In this embodiment, each remote node is connected by optical fiber and distributed at different locations, and the signal received by each remote node is transmitted to the central station via optical fiber at different distances, so the signal received by each node satisfies the hybrid support set model. . A spectrally sparse signal satisfying a mixed support set can be modeled as:

in Represents the public part of the signal, /> Represents the innovative part of the signal, /> and/> can be sparsely expressed on a certain sparse basis, that is:

Among them, W is the Fourier orthonormal basis; and/> Sparse spectral vectors representing common and innovative parts. The common part of each signal in the mixed support set/> Not exactly the same, only the frequency point position is the same, the amplitude is different, for each signal, the innovative part/> is completely independent. Assume that the sparsity of the common part of each signal is K _c and the sparsity of the innovative part of each signal is K _j . Therefore, the sparsity of each signal is K=K _c +K _j . For the signal set (x ₁ , x ₂ ,..., x _j ), the total sparsity of the signal set is/>

Among them, the J frequency sparse signals to be measured received by the J remote nodes 2 are sparse signals in the frequency domain and satisfy the mixed support set model to represent the correlation between signals and the signal itself.

The multi-wavelength optical carrier carrying J-channel radio frequency information is transmitted back to the central station 8 through optical fiber. First, it is decomposed into J optical carriers by the demultiplexer 9. Each optical carrier carrying radio frequency information enters the second electro-optical modulator 10 and J different pseudo-random sequences generated by the pseudo-random sequence generator 11 for mixing. frequency.

Among them, the pseudo-random sequence generator 11 emits a pseudo-random sequence, and its repetition rate must be greater than or equal to the Nyquist frequency of the J-channel sparse signal, thereby ensuring that the signal after mixing the sparse signal and the pseudo-random sequence retains the original sparse signal. All information.

Further, the mixed frequency signal of each channel is converted into an electrical signal through the photodetector 12, and then passes through the low-pass filter 13 and the sampler 14 to obtain J sets of compressed sampling measurement value sequences y _j [n]. Send y _j [n] to the signal joint reconstruction calculation module 15, and jointly restore the sparse signals of the measured frequencies of J remote nodes 2 through the distributed compressed sampling reconstruction algorithm.

Specifically, in this embodiment, the sparse signal x _j of frequency to be measured received by the jth remote node is subjected to photon distributed compression sampling to obtain the observation value y _j .

Proceed as follows:

Assume that the sparse signal received by the jth remote node is x _j (t). The sparse signal x _j (t) is modulated on the optical carrier with wavelength λ _j through the first electro-optical modulator. The electro-optic modulator completes the mixing of the sparse signal x _j (t) and the pseudo-random binary sequence r _j (t), and then passes through the photodetector to output a photogenerated current of i _j (t)∝[1+αx _j (t) ]·r _j (t)＝x′ _j (t)·r _j (t), where α is the modulation coefficient. After low-pass filtering and downsampling, the compressed sampling system can be expressed by the matrix equation as follows: y _j ＝Φ _j x _j =D _j H _j R _j x _j , where is the measurement result of y _j , Φ _j =D _j H _j R _j is the measurement matrix, R _j represents the pseudo-random binary sequence r _j (t), and H _j represents the low pass The impulse response of the filter, D _j represents the downsampling process of the sampler ADC. Φ _j is a random matrix following an appropriate probability distribution, and the matrix product Φ _j W satisfies the constrained isometry condition.

The joint sparse signal

in is a sparse signal set in the frequency domain that satisfies the mixed support set, Indicates the measurement result of J route,/> is a diagonal matrix composed of the measurement matrix Φ _j of each signal x _j , /> is a diagonal matrix composed of J Fourier orthogonal basis matrices W, Represents the sparse spectrum information of J signals.

The main goal of this embodiment is to simultaneously reconstruct the signal set X from the measured value Y through the joint sparse characteristics in the frequency domain. θ ^* can be completely reconstructed by solving the minimum l ₁ problem.

The joint reconstruction calculation module restores the joint sparse signal X, and the joint reconstruction algorithm is the distributed compressed sampling sparse adaptive matching pursuit (DCS-SAMP) algorithm. Compared with electrical domain distributed compression sampling and single-node photon compression sampling, this solution can achieve a wider range of spectrum detection and higher reconstruction accuracy.

It should be noted that the joint reconstruction algorithm mentioned in this embodiment is the same as the recovery algorithm of distributed compression sampling mentioned in the background art, and is an existing technology.

To further verify the effectiveness of this embodiment, in this embodiment, two nodes are included, and the two sets of signals are set to the common sparsity K _c =3 and the innovative sparsity K ₁ =K ₂ =1. The signal frequency of the first node is set to 0.4GHz, 1.1GHz, 1.5GHz, 1.8GHz, and the signal frequency of the second node is set to 0.4GHz, 1GHz, 1.5GHz, 1.8GHz. The first node is modulated to the optical carrier through the first electro-optical modulator biased at the orthogonal bias point, and then transmitted through a section of optical fiber with a length of 20km into another second electro-optical modulator biased at the orthogonal bias point. The pseudo-random binary sequence generated by the generator and the pseudo-random sequence generator realizes random mixing. Photoelectric conversion is performed by a photodetector with a 3dB bandwidth of 10GHz, and then through a low-pass filter and sampler, the compressed sampling measurement results are finally obtained. The steps for photon distributed compression sampling of the second node are the same as those of the first node, except that the optical fiber between the first electro-optical modulator and the second electro-optical modulator is replaced with 10km. The original spectrum of the two sets of signals and the individually restored spectrum of a single node are given in Figure 2(a) and Figure 2(b), while the original spectrum and jointly restored spectrum are given in Figure 3(a) and Figure 3(b). In the case of a single node being restored alone, the restored frequency point is different from the original signal, indicating that the restoration failed. In the case of joint recovery, the system accurately restored the frequency of the original signal. This example illustrates that joint recovery has better performance than single-node recovery alone.

The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, without departing from the principle and spirit of the invention, various changes, modifications, substitutions and modifications can be made to these embodiments, including components, and still fall within the protection scope of the invention.

Claims (6)

1. The utility model provides a photon auxiliary distributed compression sampling system, its characterized in that includes central website (8) and a plurality of far-end node (2) that connect gradually, far-end node (2) include first wavelength division multiplexer (3), first electro-optic modulator (4), second wavelength division multiplexer (5) and erbium-doped fiber amplifier (7), first electro-optic modulator (4) set up between first wavelength division multiplexer (3) and second wavelength division multiplexer (5), first electro-optic modulator (4) connect antenna (6) are used for receiving the sparse signal of frequency, central website (8) include multiwave wavelength light source (1), joint reconfiguration calculation module (15) and de-multiplexer (9), be equipped with the sampling module between joint reconfiguration calculation module (15) and de-multiplexer (9), tail end amplifier (7) in far-end node (2) output end and erbium-doped fiber amplifier (7) in the far-end node (2) are connected with the far-end multiplexer (5) input end of the adjacent wavelength division multiplexer (9).

2. A photon assisted distributed compressed sampling system according to claim 1, wherein the sampling module comprises a number of sampling paths comprising a second electro-optical modulator (10), a photodetector (12), a low pass filter (13) and a sampler (14) connected in sequence.

3. A photon-assisted distributed compressed sampling system according to claim 2, wherein a pseudo-random sequence generator (11) is connected to the second electro-optic modulator (10).

4. The implementation method of the photon-assisted distributed compressive sampling system is characterized by comprising the following steps of:

s1, providing continuous light with J wavelengths by a multi-wavelength light source, wherein the wavelength is lambda ₁ ，λ ₂ ，···，λ _J And transmitting to a remote node, said remote node comprising a first wavelength division multiplexer, a first electro-optic modulator, a second wavelength division multiplexer and an erbium-doped fiber amplifier;

s2, extracting wavelength lambda from first wavelength division multiplexer in far-end node ₁ Other optical carriers directly pass through;

s3, the frequency sparse signal to be detected is received by the antenna and modulated to an optical carrier lambda through a first electro-optic modulator ₁ Then the modulated light signal and other light wavelengths are output by a second wavelength division multiplexer to be combined and then continuously transmitted on the optical fiber;

s4, outputting a combined signal by the second wavelength division multiplexer, and compensating the loss of the optical power of the line through the erbium-doped optical fiber amplifier;

s5, repeating the steps S1-S4, wherein J represents a wavelength sequence number and is less than or equal to J at a J-th remote node, and modulating a frequency sparse signal to be detected to an optical carrier lambda through a first electro-optic modulator _j Up to the J-th far-end node and outputting a multi-wavelength optical carrier wave carrying J paths of frequency sparse signals to be tested;

s6, decomposing the multi-wavelength optical carrier carrying J paths of frequency sparse signals to be detected into J paths of optical carriers through a wavelength division demultiplexer;

s7, each path of optical carrier wave respectively enters a second electro-optical modulator and is mixed with J different pseudo-random sequences generated by a pseudo-random sequence generator to obtain J mixed signals;

s8, the mixed signals of each path are converted into electric signals through a photoelectric detector, and then the electric signals are filtered through a low-pass filter andthe sampler obtains a J group compressed sampling measurement value sequence y _j [n]；

S9, y _j [n]And sending the signals to a signal joint reconstruction calculation module, and jointly recovering the to-be-detected frequency sparse signals of J remote nodes through a distributed compressive sampling reconstruction algorithm.

5. The implementation method of a photon-assisted distributed compressive sampling system according to claim 4, wherein J frequency sparse signals received by J of said remote nodes (2) are frequency domain sparse signals and satisfy a hybrid support set model to represent correlations between signals and signals themselves.

6. The implementation method of a photon-assisted distributed compressive sampling system according to claim 4, wherein the repetition rate of J pseudo-random sequences sent by the pseudo-random sequence generator (11) is not less than the nyquist frequency of the sparse frequency signal to be measured.