Underwater coherent superposition keying method and system combining Reed Solomon code and orbital angular momentum

Abstract

The present invention discloses an underwater coherent superposition keying method and system combining Reedsor code with orbital angular momentum. The present invention uses a semiconductor laser as a light source, uses a beam splitter to generate four parallel light beams, and controls the optical switch in parallel with the encoded data to control the parallel light beams to generate vortex light beams with different orbital angular momentum states through a spatial light modulator. Based on the coherent superposition, a composite vortex light beam is generated. After being disturbed by the ocean turbulence channel, adaptive demodulation based on deep learning is realized at the receiving end, and then the original data is restored through a decoder. The modulation rate of the digital information of the present invention is no longer limited by the refresh rate of the spatial light modulator. The introduction of encoding and decoding can effectively improve the bit error rate performance of the system and increase the anti-interference ability of the system. The demodulation method based on the pre-trained neural network of the present invention improves the demodulation speed and accuracy of the system. The demodulation structure is very simple and reduces the system cost.

Description

Technical Field

[0001] The present invention relates to the field of underwater wireless optical communication, and in particular to an underwater coherent superposition keying method and system combining Reedsor code with orbital angular momentum. Background Art

[0002] Electromagnetic waves have not only momentum but also energy during transmission. Angular momentum consists of two parts: spin angular momentum and orbital angular momentum (OAM). Orbital angular momentum is due to the spatial phase structure of the electromagnetic field. Electromagnetic waves carrying different orbital angular momentum are orthogonal to each other. The value of orbital angular momentum can be integers and fractions, and the theoretical range of values ​​is infinite. Therefore, orbital angular momentum, as a new physical dimension, is independent of wavelength / frequency, time, amplitude, phase and other dimensions, and can be used as a new degree of freedom. Combining it with current mature communication technology can greatly improve the performance of the communication system. Therefore, how to better apply orbital angular momentum in communication systems has become a research hotspot.

[0003] Common digital signal encoding methods currently include amplitude keying, quadrature amplitude modulation, frequency shift keying, phase shift keying, etc. These encoding methods respectively utilize the amplitude, frequency and phase of electromagnetic waves. The orbital angular momentum dimension of the vortex beam can also be used for signal encoding. This encoding method is called orbital angular momentum keying. In this encoding method, N vortex beams with different orbital angular momentum states can be used to represent an N-ary digital information, so that the vortex beam emitted by the system transmitter carries data information. At the receiving end of the system, the transmitted information is obtained by detecting and decoding the vortex beam.

[0004] In OAM coding technology, N different OAM states (l1, l2, l3…l N ) can represent an N-ary number (0,1,2,…N-1), and then one encoding has Compared with traditional binary coding, the coding efficiency is improved by In addition, the composite vortex beams of multiple orbital angular momentum states can be used to represent different digital information, and the coding efficiency can be increased by up to N times. Since the topological charge can be any integer or fraction between -∞ and +∞, the channel capacity of the orbital angular momentum coding system based on orbital angular momentum can be expanded to infinity in theory.

[0005] In orbital angular momentum coding technology, although the orthogonal characteristics between orbital angular momentum states are used to improve the coding efficiency exponentially, a complex demodulation structure is required at the receiving end, and the refresh rate of the spatial light modulator is required to be high, which has caused problems in the practical application of orbital angular momentum coding. Therefore, new technologies are needed to improve the demodulation speed and accuracy of the receiving end, while reducing the requirements for the spatial light modulator. In addition, the huge interference of the ocean turbulence channel on the orbital angular momentum state also causes problems for the identification of the receiving end and the performance of the system. New methods are needed to reduce the interference of ocean turbulence and improve system performance. Summary of the invention

[0006] The purpose of the present invention is to address the defects of the current orbital angular momentum coding system, such as the coding efficiency is limited by the refresh frequency of the spatial light modulator, the demodulation structure of the composite vortex light beam is complex, the efficiency is low, and the ocean turbulence has a huge interference on the propagation of the orbital angular momentum state, and propose an underwater coherent superposition keying method and system combining Reed code and orbital angular momentum. The present invention generates a composite vortex light beam for propagation by coherent superposition of vortex light beams with different orbital angular momentum states, and realizes the demodulation of the received composite vortex light beam based on a pre-trained neural network, which greatly improves the demodulation speed and accuracy of the system, and introduces a coding and decoding part into the system, which greatly improves the system performance, while reducing the performance requirements for the spatial light modulator, and the system structure is simple and easy to implement.

[0007] The scheme adopted by the present invention to achieve the above-mentioned purpose is as follows:

[0008] An underwater coherent superposition keying system combining Reedsor code and orbital angular momentum includes a semiconductor laser 1, a first optical beam splitter, a second optical beam splitter, a third optical beam splitter, a fourth optical beam splitter, a fifth optical beam splitter, a sixth optical beam splitter, a first optical switch, a second optical switch, a third optical switch, a fourth optical switch, a first spatial light modulator, a second spatial light modulator, a third spatial light modulator, a fourth spatial light modulator, an underwater turbulent channel model, a CCD camera, a data link and a host, an encoder, and a decoder.

[0009] Another object of the present invention is to provide an underwater coherent superposition keying scheme combining the Reed code and the orbital angular momentum of the light beam of the system, the steps are as follows: the light beam emitted by the semiconductor laser is divided into two parallel light beams through the first optical beam splitter, and then passes through the second optical beam splitter and the third optical beam splitter to generate four parallel light beams. The encoder parallel output controls an optical switch, a second optical switch, a third optical switch, and a fourth optical switch respectively. The four parallel light beams pass through the first optical switch, the second optical switch, the third optical switch, and the fourth optical switch respectively, so that the parallel light beams carry the information of the digital signal, and then are respectively input into the first spatial light modulator, the second spatial light modulator, the third spatial light modulator, and the fourth spatial light modulator for modulation. The four parallel vortex light beams output by the first spatial light modulator, the second spatial light modulator, the third spatial light modulator, and the fourth spatial light modulator pass through the fourth optical beam splitter, the fifth optical beam splitter, and the sixth optical beam splitter to generate a coherent superposition composite vortex light beam. The composite vortex light beam is incident on the CCD camera through an underwater turbulent channel, and a light intensity distribution diagram of the composite vortex light beam changing with time is obtained. The light intensity distribution diagram is transmitted to the host via a digital link and demodulated with the help of a pre-trained neural network. The demodulated information is then used to obtain the transmitted digital information through a decoder.

[0010] The beneficial effects of the present invention are as follows:

[0011] The present invention proposes an underwater coherent superposition keying method and system combining Reedsor code and orbital angular momentum. Compared with the existing method relying on spatial light modulator for modulation, the present invention combines optical switch and spatial light modulator for modulation, which is no longer limited by the refresh rate of the spatial light modulator, thereby greatly improving the modulation rate. The introduction of codec can effectively improve the bit error rate performance of the system and increase the anti-interference ability of the system. Compared with the existing digital signal demodulation method and system, the demodulation method based on pre-trained neural network of the present invention improves the demodulation speed and accuracy of the system, the demodulation structure is very simple, and the system cost is reduced. It has great application prospects in underwater wireless optical communication systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a schematic diagram of an underwater coherent superposition keying method and system structure combining Reedsor code and orbital angular momentum proposed by the present invention;

[0013] Figure 2 The vortex beam intensity distribution diagram is generated after four parallel light beams are modulated by the first optical switch, the first spatial light modulator; the second optical switch, the second spatial light modulator; the third optical switch, the third spatial light modulator; the fourth optical switch, and the fourth spatial light modulator;

[0014] Figure 3The intensity distribution diagram of the coherently superimposed composite vortex light beam generated by the fourth optical beam splitter, the fifth optical beam splitter, and the sixth optical beam splitter, and the code pattern sequences corresponding to the different intensity distribution diagrams;

[0015] Figure 4 It is a schematic diagram of the structure of the encoder;

[0016] Figure 5 It is a schematic diagram of the structure of the decoder;

[0017] Figure 6 It is the distribution graph of the accuracy and loss function of the convolutional neural network with the number of iterations;

[0018] Figure 7 It is the improvement of system performance after the introduction of codec.

[0019] Among them, 1. semiconductor laser; 2. first optical beam splitter; 3. second optical beam splitter; 4. third optical beam splitter; 13. fourth optical beam splitter; 14. fifth optical beam splitter; 15. sixth optical beam splitter; 5. first optical switch; 6. second optical switch; 7. third optical switch; 8. fourth optical switch; 9. first spatial light modulator; 10. second spatial light modulator; 11. third spatial light modulator; 12. fourth spatial light modulator; 16. underwater turbulence channel model; 17. CCD camera; 18. data link and host; 19. encoder; 20. decoder. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0021] like Figure 1 As shown, the present invention proposes an underwater coherent superposition keying method and system combining Reedsor code and orbital angular momentum of light beam, including a semiconductor laser 1, a first optical beam splitter 2, a second optical beam splitter 3, a third optical beam splitter 4, a fourth optical beam splitter 13, a fifth optical beam splitter 14, a sixth optical beam splitter 15, a first optical switch 5, a second optical switch 6, a third optical switch 7, a fourth optical switch 8, a first spatial light modulator 9, a second spatial light modulator 10, a third spatial light modulator 11, a fourth spatial light modulator 12, an underwater turbulent channel model 16, a CCD camera 17, a data link and a host 18, an encoder 19, and a decoder 20.

[0022] like Figure 1As shown, the light beam emitted by the semiconductor laser 1 is divided into two parallel light beams by the first light splitter 2, and then respectively passes through the second light splitter 3 and the third light splitter 4 to generate four parallel light beams; the four parallel light beams respectively pass through the first optical switch 5, the second optical switch 6, the third optical switch 7, and the fourth optical switch 8 controlled by the parallel output of the encoder, so that the parallel light beams carry the information of the digital signal; then they are respectively input into the first spatial light modulator 9, the second spatial light modulator 10, the third spatial light modulator 11, and the fourth spatial light modulator 12 for modulation; then the vortex light beams output by the first spatial light modulator 9 and the second spatial light modulator 10 pass through the fourth The vortex light beams output by the optical beam splitter 13, the third spatial light modulator 11 and the fourth spatial light modulator 12 pass through the fifth optical beam splitter 14, and the light beams output by the fourth optical beam splitter 13 and the fifth optical beam splitter 15 generate a coherently superimposed composite vortex light beam in the sixth optical beam splitter 15; the composite vortex light beam is incident on the CCD camera 17 through the underwater turbulent channel 16, and a light intensity distribution diagram of the composite vortex light beam varying with time is obtained, and the light intensity distribution diagram is transmitted to the data link and the host 18, and is demodulated with the help of the pre-trained neural network in the data link and the host 18, and the output result is transmitted to the decoder 20, and then the transmitted digital information is obtained through the decoder 20.

[0023] Furthermore, the four parallel vortex light beams output by the first spatial light modulator 9, the second spatial light modulator 10, the third spatial light modulator 11, and the fourth spatial light modulator 12 can be expressed as:

[0024]

[0025]

[0026]

[0027]

[0028] Among them, A1, A2, A3, A4 are the complex amplitudes of the light field, and l1, l2, l3, l4 represent different orbital angular momentum states. The light intensity distribution diagram is as follows: Figure 2 Then, the four parallel vortex beams pass through the fourth beam splitter 13, the fifth beam splitter 14, and the sixth beam splitter 15 to generate a coherent superposition composite vortex beam, which can be expressed as:

[0029] E=E1+E2+E3+E4

[0030] The intensity distribution of the composite vortex beam generated by different digital information is as follows: Figure 3As shown. The composite vortex beam is incident on the CCD camera 17 through the underwater turbulent channel 16, and a light intensity distribution diagram of the composite vortex beam changing with time is obtained. The light intensity distribution diagram is transmitted to the data link and the host 18, and is demodulated with the help of the pre-trained neural network in the data link and the host 18, wherein the convolutional neural network on the host is based on the ResNet architecture. After the image is input, it passes through a 7*7 convolution layer, a batch normalization layer, a rectified linear unit, and a maximum pooling layer, and then is recognized by the ResNet framework. When outputting, it passes through a rectified linear unit, a global average pooling layer, a fully connected layer, and a maximum pooling layer, and the output result is transmitted to the decoder 20, and then the transmitted digital information is obtained through the decoder 20. Figure 6 A graph showing the distribution of the accuracy and loss function of the convolutional neural network with the number of iterations.

[0031] like Figure 4 As shown, the encoder is implemented as follows:

[0032] The encoder includes a system converter, a block, an interleaver, a mapper and a polynomial generation module. After the information sequence passes through the encoder 19, it is connected to the first optical switch 5 and the fourth optical switch 8 at the same time. The information sequence is first converted into an n-base number by a system converter in the encoder 19, where n is the same as the base of the system keying modulation. The digital sequence is then divided into blocks, multiplied with the generation matrix for encoding, and then interleaved and converted, and finally the encoded information is output through the mapper.

[0033] like Figure 5 As shown, the decoder is implemented as follows:

[0034] The decoder includes a demapper, a base conversion, a deinterleaver, a block and a polynomial generation module. The information output by the neural network is demapped, base converted, and deinterleaved, and then the digital sequence is calculated in blocks. The error position polynomial is calculated by solving the error position polynomial to obtain the error position. The error pattern is obtained according to the error position, and then calculated with the original block data to obtain the decoded data, which is then output after base conversion.

[0035] The present invention proposes an underwater coherent superposition keying method and system combining Reed code and orbital angular momentum. Compared with the existing method of modulation by spatial light modulator, taking the 1920×1152XY phase series spatial light modulator provided by Meadowlark Optics of the United States as an example, its refresh rate is only 31Hz, which greatly limits the modulation rate of the system. The present invention combines optical switch and spatial light modulator for modulation, which is no longer limited by the refresh rate of the spatial light modulator. The optical switch is controlled by the frequency of the digital signal, and the refresh frequency can reach the Ghz level. The modulation is performed by switching the optical switch, thereby greatly improving the modulation rate of the system. The introduction of codec can effectively improve the bit error rate performance of the system and increase the anti-interference ability of the system. Compared with the existing demodulation method and system of digital signals, the demodulation method based on pre-trained neural network of the present invention improves the demodulation speed and accuracy of the system, and the demodulation structure is very simple, which reduces the system cost. It has great application prospects in underwater wireless optical communication systems.

Claims

1. An underwater coherent superposition keying system combining Reed Solomon code and orbital angular momentum, characterized in that The invention comprises a semiconductor laser (1), a first optical beam splitter (2), a second optical beam splitter (3), a third optical beam splitter (4), a fourth optical beam splitter (13), a fifth optical beam splitter (14), a sixth optical beam splitter (15), a first optical switch (5), a second optical switch (6), a third optical switch (7), a fourth optical switch (8), a first spatial light modulator (9), a second spatial light modulator (10), a third spatial light modulator (11), a fourth spatial light modulator (12), an underwater turbulence channel model (16), a CCD camera (17), a data link and a host (18), an encoder (19), and a decoder (20); A light beam emitted by a semiconductor laser (1) is split into two parallel light beams through a first light beam splitter (2), and then respectively passes through a second light beam splitter (3) and a third light beam splitter (4) to generate four parallel light beams; the four parallel light beams respectively pass through a first optical switch (5), a second optical switch (6), a third optical switch (7), and a fourth optical switch (8) controlled by a parallel output of an encoder, so that the parallel light beams carry information of digital signals; and then are respectively input into a first spatial light modulator (9), a second spatial light modulator (10), a third spatial light modulator (11), and a fourth spatial light modulator (12) for modulation; then the first spatial light modulator (9), the second spatial light modulator (10) The output vortex light beam passes through a fourth optical beam splitter (13), and the vortex light beams output by the third spatial light modulator (11) and the fourth spatial light modulator (12) pass through a fifth optical beam splitter (14), and the light beam output by the sixth optical beam splitter (15) generates a coherently superimposed composite vortex light beam; the composite vortex light beam is incident on a CCD camera (17) through an underwater turbulent channel (16), and a light intensity distribution diagram of the composite vortex light beam varying with time is obtained, and the light intensity distribution diagram is transmitted to a data link and a host (18), and demodulated with the help of a pre-trained neural network in the data link and the host (18), and the output result is transmitted to a decoder (20), and then the transmitted digital information is obtained through the decoder (20).

2. The underwater coherent superposition keying system combining Reed Solomon code and orbital angular momentum according to claim 1 is characterized in that The four parallel vortex light beams output by the first spatial light modulator (9), the second spatial light modulator (10), the third spatial light modulator (11), and the fourth spatial light modulator (12) can be expressed as follows: Among them, A1, A2, A3, A4 are the complex amplitudes of the light field, and l1, l2, l3, l4 represent different orbital angular momentum states.

3. The underwater coherent superposition keying system combining Reed Solomon code and orbital angular momentum according to claim 2 is characterized in that The sixth beam splitter (15) generates a coherently superimposed composite vortex beam expressed as: E=E1+E2+E3+E4.

4. The underwater coherent superposition keying system combining Reed Solomon code and orbital angular momentum according to claim 2 or 3, characterized in that The pre-trained neural network in the data link and the host (18) is a convolutional neural network based on the ResNet architecture. After the image input, it passes through a 7*7 convolution layer, a batch normalization layer, a rectified linear unit, and a maximum pooling layer, and then is recognized by the ResNet framework. At the output, it passes through a rectified linear unit, a global average pooling layer, a fully connected layer, and a maximum pooling layer, and the output result is transmitted to the decoder (20).

5. The underwater coherent superposition keying system combining Reed Solomon code and orbital angular momentum according to claim 1, 2 or 3, characterized in that The encoder is implemented as follows: The encoder includes a base converter, a block, an interleaver, a mapper and a polynomial generation module. The information sequence is first converted into an n-base number by the base converter, where n is the same as the base of the system keying modulation; the digital sequence is then divided into blocks and multiplied with the generation matrix for encoding, and then interleaved coding and base conversion are performed, and finally the encoded information is output through the mapper.

6. The underwater coherent superposition keying system combining Reed Solomon code and orbital angular momentum according to claim 5, characterized in that The decoder is implemented as follows: The decoder includes a demapper, a base conversion, a deinterleaver, a block and a polynomial generation module. The information output by the neural network is demapped, base converted, and deinterleaved, and then the digital sequence is calculated in blocks. The error position polynomial is calculated by solving the error position polynomial to obtain the error position. The error pattern is obtained according to the error position, and then calculated with the original block data to obtain the decoded data, which is then output after base conversion.

Images