CN117857280B - A non-uniform quantized digital analog fiber optic radio method - Google Patents

Abstract

The invention provides a non-uniform quantitative digital-analog optical fiber radio method, and relates to the technical field of radio. The non-uniform quantization digital-analog fiber radio method comprises the following steps: s1: off-line generating an analog RoF signal; s2: carrying out non-uniform quantization on the analog RoF signal to generate a digital PS- (2M+1) ² -QAM/PS- (2M+1) -PAM symbol, wherein the quantization factor is M; s3: calculating a residual analog quantization error signal after the non-uniform quantization of the analog RoF signal; s4: performing time domain interleaving on a digital quantized signal of an analog RoF signal and a residual analog quantized error signal to generate a NUDA-RoF signal of Time Division Multiplexing (TDM); s5: NUDA-RoF signals are processed by a sending end DSP, sent into a communication experiment system for transmission, and sampled by a receiving end oscilloscope to obtain receiving signals. The non-uniform quantization digital analog optical fiber radio method provided by the invention improves the anti-noise performance of the OFDM signal, improves the SNR of the recovered QAM signal, and realizes the transmission of ultra-high order QAM.

Description

A non-uniform quantized digital analog fiber optic radio method

Technical Field

The present invention relates to the field of radio technology and is a solution in a mobile fronthaul scenario, which involves non-uniform quantization technology of orthogonal frequency division multiplexing (OFDM)/discrete multi-carrier (DMT) signals, quantization error extraction technology, time division multiplexing technology, and a digital signal processing (DSP) algorithm at the transceiver end of a communication system, and in particular, to a non-uniform quantization digital analog fiber radio method.

Background technique

The rapid growth of mobile data traffic has posed a major challenge to network operators, with traffic expected to surge from 14 exabytes/month in 2017 to a staggering 110 exabytes/month in 2023. The widespread use of smart devices has put considerable pressure on network capacity, with operators struggling to meet the growing demand. To address this exceptionally high data traffic, there is an urgent need to significantly enhance network densification. A commonly suggested solution to reduce the cost of densification is the adoption of cloud radio access networks (C-RAN), which brings various benefits such as increased capacity and reduced capital and operating expenditures. In the C-RAN architecture, the baseband unit (BBU) is centralized in the office, while the remote radio units (RRU) are strategically deployed at each access site. The fronthaul link primarily utilizes optical fiber to establish a connection between the BBU and the RRU. Currently, the dominant standard protocol for commercial fronthaul solutions is the digital Common Public Radio Interface (CPRI). However, it is worth noting that the CPRI protocol introduces significant bandwidth inefficiencies and requires a large amount of capacity, which is approximately 16 times the user data rate. As an alternative approach, the analog radio over fiber (A-RoF) solution offers the advantage of lower RRU cost and achieves higher levels of spectral efficiency (SE) by directly transmitting the unchanged signal waveform. However, this approach is more susceptible to noise and nonlinear impairments during transmission, resulting in a degradation of the recovered signal-to-noise ratio (SNR) and error vector magnitude (EVM).

Recently, hybrid digital-analog radio-over-fiber (DA-RoF) schemes have made significant progress, providing a balanced compromise between SNR and SE. In a previous study [X.Liu, "Hybrid digital-analog radio-over-fiber (DA-RoF) modulation and demodulation achieving a SNR gain over analog RoF of> 10dB at halved spectral efficiency," in Optical Fiber Communication Conference (2021), paper Tu5D.4.], researchers successfully transmitted 8-Gbaud DA-RoF signals using an intensity modulation and direct detection (IM-DD) system. This achievement brought a significant SNR gain of 12.8dB compared to the A-RoF method, although the SE was halved. Based on this achievement, the DA-RoF scheme was further verified in standard single-mode fiber (SSMF) [Y.Xu et al., "Coherent digital-analog radio-over-fiber (DA-RoF) system with a CPRI-equivalent data rate beyond 1Tb/s for fronthaul," Opt Express 30, 29409-29420 (2022).] and uncoupled 7-core fiber [Y.Zhu, C.Zhang, X.Zeng, H.Jiang, Y.Xu, X.Xie, Q.Zhuge, and W.Hu, "1λ10.5Tb/s CPRI-equivalent rate1024-QAM transmission via self-homodyne digital-analogradio-over-fiber architecture," in European Conference on Optical Communication (2022), paper Th3A.5.] dual-polarization coherent systems. References [Y.Zhu, C.Zhang, J.Lin, Y.Xu, Q.Zhuge, W.Hu, Z.Chen, W.Hu, and X.Xie,"203.6Tb/sCPRI-equivalent rate 1024-QAM DA-RoF fronthaul with comb-based WDM and SDMsuperchannel," in Optical Fiber Communication Conference (2023), paper Th4C.6.] demonstrate the successful transmission of the DA-RoF scheme within a comb-based wavelength division multiplexing and space division multiplexing (WDM/SDM) superchannel, achieving a total CPRI equivalent data rate (CPRI-EDR) of 203.6Tb/s. Thanks to the excellent noise resistance of the digital part, the DA-RoF scheme effectively retains most of the characteristics of the original OFDM/DMT. However, the analog quantization error in the DA-RoF scheme is still susceptible to noise and nonlinearity. The reduced fidelity of the analog part hinders accurate OFDM/DMT reconstruction, thereby reducing the recovered SNR. To improve the fidelity of analog quantization errors, researchers used delta-sigma modulation (DSM) [M.Wang, J.Yu, X.Zhao, W.Li, Y.Wei, X.Yang, J.Shi, C.Bian, T.Xie, F.Zhao, J.Yu, W.Zhou, and K.Wang, "SNR improved digital-delta-sigma-modulation radio-over-fiber scheme for D-band 4.6-km photonics-aided wireless fronthaul," Opt.Lett.48, 3997-4000(2023).] and pulse code modulation (PCM) [M.Wang et al., "SNR improved digital-cascaded-pulse-code-modulation radio-over-fiber scheme supporting 16,777,216QAM for mobile fronthaul," J. Opt. Commun. Netw. 15, 948-957 (2023).] and other methods have studied the re-digitization of residual analog errors. However, re-quantization of the analog part will lead to an increase in the required bandwidth, thereby reducing CPRI-EDR.

In the above RoF scheme, the quantization of analog OFDM/DMT signals adopts a uniform quantization scheme. However, for non-uniformly distributed signals (such as OFDM, DMT), the uniform quantization scheme provides a lower signal-to-noise power ratio (SQNR) limit.

Therefore, it is necessary to propose a novel non-uniform quantization digital analog fiber radio method for mobile fronthaul scenarios to solve the above technical problems.

Summary of the invention

Based on the technical problems raised above, the present invention proposes a novel non-uniform quantization digital analog fiber radio method, which adopts a variable quantization interval, that is, by allocating a smaller interval to the amplitude range with high probability density, to reduce the quantization noise and improve the SQNR. Applying the non-uniform quantization scheme to the DA-RoF scheme can reduce the impact of quantization error distortion on OFDM/DMT reconstruction, thereby improving the recovered SNR, and solves the problems of large quantization noise, low SQNR, and severe impact of quantization noise distortion on the original OFDM/DMT reconstruction in the DA-RoF scheme studied by previous researchers. Compared with the A-RoF and DA-RoF schemes, the SNR is improved; at the same time, the problems of low spectrum efficiency and low CPRI-EDR of the digital fiber radio (D-RoF) scheme are solved.

The non-uniform quantization digital analog optical fiber radio method proposed by the present invention comprises the following steps:

S1: Generate simulated RoF signal offline;

S2: Perform non-uniform quantization on the analog RoF signal to generate digital PS-(2M+1) ² -QAM/PS-(2M+1)-PAM symbols, with a quantization factor of M;

S3: Calculate the residual analog quantization error signal after non-uniform quantization of the analog RoF signal;

S4: interleave the digital quantization signal of the analog RoF signal and the residual analog quantization error signal in the time domain to generate a time division multiplexed (TDM) NUDA-RoF signal;

S5: The NUDA-RoF signal is processed by the DSP at the transmitting end, sent to the communication experiment system for transmission, and sampled by the oscilloscope at the receiving end to obtain the received signal;

S6: The received signal passes through the receiving end DSP, time division multiplexing, NUDA-RoF signal demodulation, and OFDM/DMT demodulation to obtain the ultra-high-order QAM signal carried by OFDM/DMT;

S7: Calculate the SNR and EVM of the ultra-high-order QAM signal to evaluate the performance of the solution;

The specific workflow is as follows: the analog RoF signal is modulated by NUDA-RoF to generate a non-uniform digital quantization signal D _1U and a residual analog quantization error signal A _1U ; these two parts are interleaved in the time domain through time division multiplexing (TDM) technology to generate NUDA-RoF symbols, which are processed by the digital signal at the transmitter and then generated by the arbitrary waveform generator (AWG) to be sent into the communication transmission system for transmission; the receiving end samples through a digital oscilloscope (OSC), and then performs DSP processing and NUDA-RoF signal demodulation to recover the OFDM/DMT signal, and then demodulates the OFDM/DMT to recover the original high-order QAM signal carried by the OFDM/DMT, and calculates the SNR and EVM to evaluate the performance of the scheme.

Preferably, the analog RoF signal can be either an OFDM signal or a DMT signal.

Preferably, since the OFDM/DMT signal obeys Gaussian distribution, the quantized signal of the simulated RoF signal is PS-QAM/PS-PAM.

Preferably, due to the use of a non-uniform quantization scheme, the distances between the generated digital signal constellation points are different. Where the signal probability density is high, the quantization interval is small and the constellation points are dense. Conversely, where the signal probability density is low, the quantization interval is large and the constellation points are sparse.

Preferably, the non-uniform quantization scheme is any one of A-law, μ-law, and Lloyd algorithms.

Preferably, the scheme is applicable to a variety of communication systems, including but not limited to photon-assisted millimeter-wave terahertz systems, all-electric millimeter-wave terahertz systems, polarization multiplexed coherent transmission systems, direct modulation and direct detection systems, and multi-mode/multi-core optical fiber transmission systems. It has good universality. For example, in a photon-assisted millimeter-wave/terahertz system, the NUDA-RoF electrical signal completes electrical-optical conversion in the I/Q modulator, beats with another optical signal in a photodetector (PD) to generate a millimeter-wave/terahertz electrical signal, and transmits it in free space through an antenna and a lens. After the antenna at the receiving end receives the high-frequency electrical signal, it is down-converted to an intermediate frequency through a low-noise amplifier and a mixer, and then sampled by an oscilloscope. The sampled data is demodulated by the receiving end DSP and NUDA-RoF to restore the original transmitted symbol to evaluate the performance of the system.

Preferably, in the photon-assisted millimeter wave/terahertz system, the NUDA-RoF electrical signal completes electrical-to-optical conversion in the IQ modulator, beats with another optical signal in the photodetector (PD) to generate a millimeter wave/terahertz radio frequency signal, and transmits it in free space through the antenna. After the antenna at the receiving end receives the high-frequency electrical signal, it passes through a low-noise amplifier and a mixer to down-convert the high-frequency signal to an intermediate frequency, and then is sampled by an oscilloscope.

Preferably, the receiving end DSP and NUDA-RoF demodulation, OFDM demodulation is the inverse process of the transmitting end.

Preferably, the modulation parameters of the NUDA-RoF, including the PS-QAM/PS-PAM order, the size of the quantization factor, and the number of quantization times, can be specifically selected according to different systems, different channel characteristics, and different transmission indicators.

In addition, the key part of the present invention is the modulation and demodulation of NUDA-RoF, and the principle is as follows: the analog RoF signal (here taking OFDM signal as an example, the same applies to DMT) W is divided into a digital part and an analog part, the digital part is a digital signal D _1U generated by non-uniform quantization technology, and since the time domain amplitude of the OFDM signal obeys the complex Gaussian distribution, D _1U is naturally a probability-shaped quadrature amplitude modulation (PS-QAM) signal. After non-uniform quantization, the residual analog error A _1U is extracted through (WD _1U ) as the analog segment in the NUDA-RoF scheme;

The selection of the quantization factor (M) in the NUDA-RoF scheme involves a trade-off between the recovered SNR and symbol error. As the quantization factor (M) increases, the high-order PS-(2M+1) ² -QAM signal encapsulates more original OFDM information, resulting in a reduction in OFDM features present in the analog part. Therefore, the impact of A _1U distortion on OFDM recovery is mitigated, which helps to improve the recovered SNR;

However, high-order PS-(2M+1) ² -QAM signals are more susceptible to damage caused by noise and transceiver nonlinearity, and symbol errors in the digital part may cause the recovered SNR to decrease. Therefore, it is necessary to select a suitable quantization factor according to the actual channel conditions to achieve the best demodulation SNR and EVM. After that, the digital quantization part D _1U and the analog quantization error part A _1U of the analog RoF signal are time-division multiplexed to generate a NUDA-RoF signal. The demodulation of NUDA-RoF is the inverse process of modulation.

Compared with the related art, the non-uniform quantization digital analog optical fiber radio method proposed in the present invention has the following beneficial effects:

The present invention proposes a non-uniform quantized digital analog optical fiber radio method:

1. Compared with A-RoF, this solution improves the noise resistance of OFDM signals, improves the SNR of recovered wireless signals, and realizes the transmission of ultra-high-order QAM;

2. Compared with DA-RoF, this solution uses non-uniform quantization technology to compress the analog quantization error and reduce the OFDM information it carries, thereby reducing the impact of analog error distortion on OFDM reconstruction, thereby further improving the demodulation SNR;

3. Compared with D-RoF, this solution saves bandwidth, greatly improves CPRI-EDR, realizes high spectrum efficiency transmission of signals, and provides a good solution for future mobile fronthaul;

4. The modulation parameters of NUDA-RoF in the present invention, such as PS-QAM/PS-PAM order, quantization factor size, quantization times, etc., can be selected according to different systems, different channel characteristics, different transmission indicators and other specific conditions, and have high adjustability and applicability;

5. The present invention has good versatility and is applicable to various application scenarios such as photon-assisted millimeter-wave terahertz systems, all-electric millimeter-wave terahertz systems, polarization multiplexed coherent transmission systems, direct modulation and direct detection systems, and multi-mode/multi-core optical fiber transmission systems;

6. Both OFDM and DMT signals can be used as analog RoF signals in this solution.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of the specific principle and system architecture of the NUDA-RoF solution of the present invention;

Figure 2 is a constellation diagram and an enlarged detail diagram after OFDM non-uniform quantization;

FIG3 is a schematic diagram of a time domain waveform of a NUDA-RoF signal (taking an I-channel signal as an example).

Numbers in the figure:

Figure 1:

W: analog RoF signal (OFDM/DMT);

D _1U : PS-QAM/PS-PAM symbol after non-uniform quantization of simulated RoF signal;

A _1U : residual analog quantization error;

A _1U ': receiving A ₁ after system transmission;

D _1U ': received D _1U after system transmission;

W’: W restored by adding the digital and analog signal parts after the system transmission;

Figure 2:

W: Analog RoF signal (OFDM/DMT)

D _1U : Simulate the signal after non-uniform quantization of RoF signal (taking PS-QAM as an example)

d ₁ , d ₂ , d ₃ : distances between constellation points after non-uniform quantization.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings and implementation modes.

Please refer to Figures 1-3, where Figure 1 is a schematic diagram of the specific principle and system architecture of the NUDA-RoF solution in the present invention; Figure 2 is a constellation diagram and an enlarged detail diagram after OFDM non-uniform quantization; and Figure 3 is a schematic diagram of the time domain waveform of the NUDA-RoF signal (taking the I-channel signal as an example).

The present invention proposes a non-uniform quantized digital analog optical fiber radio method comprising the following steps:

S1: Generate simulated RoF signal offline;

S2: Perform non-uniform quantization on the analog RoF signal to generate digital PS-(2M+1) ² -QAM/PS-(2M+1)-PAM symbols, with a quantization factor of M;

S3: Calculate the residual analog quantization error signal after non-uniform quantization of the analog RoF signal;

S4: interleave the digital quantization signal of the analog RoF signal and the residual analog quantization error signal in the time domain to generate a time division multiplexed (TDM) NUDA-RoF signal;

S5: The NUDA-RoF signal is processed by the DSP at the transmitting end, sent to the communication experiment system for transmission, and sampled by the oscilloscope at the receiving end to obtain the received signal;

S6: The received signal passes through the receiving end DSP, time division multiplexing, NUDA-RoF signal demodulation, and OFDM/DMT demodulation to obtain the ultra-high-order QAM signal carried by OFDM/DMT;

S7: Calculate the SNR and EVM of the ultra-high-order QAM signal to evaluate the performance of the solution;

The specific workflow is as follows: the analog RoF signal is modulated by NUDA-RoF to generate a non-uniform digital quantization signal D _1U and a residual analog quantization error signal A _1U ; these two parts are interleaved in the time domain through time division multiplexing (TDM) technology to generate NUDA-RoF symbols, which are processed by the digital signal at the transmitter and then generated by the arbitrary waveform generator (AWG) to be sent into the communication transmission system for transmission; the receiving end samples through a digital oscilloscope (OSC), and then performs DSP processing and NUDA-RoF signal demodulation to recover the OFDM/DMT signal, and then demodulates the OFDM/DMT to recover the original high-order QAM signal carried by the OFDM/DMT, and calculates the SNR and EVM to evaluate the performance of the scheme.

The analog RoF signal may be either an OFDM signal or a DMT signal.

Since the OFDM/DMT signal obeys Gaussian distribution, the simulated RoF signal is quantized into PS-QAM/PS-PAM.

Due to the use of a non-uniform quantization scheme, the distances between the generated digital signal constellation points are different. Where the signal probability density is high, the quantization interval is small and the constellation points are dense. Conversely, where the signal probability density is low, the quantization interval is large and the constellation points are sparse.

The non-uniform quantization scheme may be A-law, μ-law, or Lloyd algorithm, and is not limited to a certain non-uniform quantization algorithm.

The scheme is applicable to a variety of communication systems, including but not limited to photon-assisted millimeter-wave terahertz systems, all-electric millimeter-wave terahertz systems, polarization-multiplexed coherent transmission systems, direct modulation and direct detection systems, and multi-mode/multi-core optical fiber transmission systems.

In the photon-assisted millimeter wave/terahertz system, the NUDA-RoF electrical signal completes electrical-to-optical conversion in the IQ modulator, beats with another optical signal in the photodetector (PD) to generate a millimeter wave/terahertz RF signal, and transmits it in free space through the antenna. After the antenna at the receiving end receives the high-frequency electrical signal, it passes through a low-noise amplifier and a mixer to down-convert the high-frequency signal to an intermediate frequency, and then is sampled by an oscilloscope.

The receiving end DSP and NUDA-RoF demodulate, and OFDM demodulation is the inverse process of the transmitting end.

The modulation parameters of the NUDA-RoF, such as the PS-QAM/PS-PAM order, the size of the quantization factor, and the number of quantization times, can be selected according to different systems, different channel characteristics, and different transmission indicators.

In the present invention, both OFDM and DMT signals can be used as analog RoF signals for this solution, and OFDM is used as an example to illustrate the solution:

As shown in FIG1 , the input OFDM signal W is non-uniformly quantized to generate a digital portion D _1U , as shown in FIG2 , given that the amplitude of the analog OFDM signal follows a complex Gaussian distribution, the resulting digital segment D _1U essentially behaves as a PS-m-QAM signal, where m is equal to (2M+1) ² , and M is a quantization factor, which determines the quantization format of the quantized signal;

It can be seen from the enlarged diagram of FIG. 2 that the distances between quantization levels are different. The smaller the quantization interval is for the small signal part with denser distribution, and the larger the quantization interval is for the large signal part with low distribution probability. This non-uniform quantization helps to reduce the quantization error. After quantization, the non-uniform quantization error A _1U is extracted from (WD _1U ).

Then, the digital part D _1U and the analog part A _1U are normalized respectively, and finally aggregated through TDM technology to generate a NUDA-RoF signal. FIG3 shows the I channel time domain waveform of the NUDA-RoF signal.

NUDA-RoF demodulation is the reverse process of modulation, as shown in Figure 1. The input signal is demultiplexed by time division multiplexing technology, and the separated digital and analog segments are amplified to the original amplitude level. After a series of decision and addition operations, the OFDM waveform W' can be reconstructed by the recovered digital part D _1U ' and analog A _1U '. The reconstructed OFDM is demodulated into an ultra-high-order QAM signal, and its demodulation SNR and EVM are calculated to evaluate the performance of the scheme.

Compared with the related art, the non-uniform quantization digital analog optical fiber radio method proposed in the present invention has the following beneficial effects:

The present invention proposes a non-uniform quantization digital analog optical fiber radio method. Compared with A-RoF, the method improves the anti-noise performance of OFDM signals, improves the SNR of restored wireless signals, and realizes the transmission of ultra-high-order QAM.

Compared with DA-RoF, this method uses non-uniform quantization technology to compress the analog quantization error and reduce the OFDM information it carries, thereby reducing the impact of analog error distortion on OFDM reconstruction, thereby further improving the demodulation SNR;

Compared with D-RoF, this method saves bandwidth, significantly improves CPRI-EDR, and achieves high spectrum efficiency transmission of signals, providing a good solution for future mobile fronthaul.

The modulation parameters of NUDA-RoF in the present invention, such as PS-QAM/PS-PAM order, size of quantization factor, number of quantization, etc., can be specifically selected according to different systems, different channel characteristics, different transmission indicators and other specific conditions, and have high adjustability and applicability;

The present invention has good versatility and is applicable to various application scenarios such as photon-assisted millimeter-wave terahertz systems, all-electric millimeter-wave terahertz systems, polarization multiplexed coherent transmission systems, direct modulation and direct detection systems, and multi-mode/multi-core optical fiber transmission systems.

Both OFDM and DMT signals can be used as analog RoF signals for this solution.

The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims (9)

1. A non-uniformly quantized digital-to-analog fiber optic radio method, the method comprising the steps of:

s1: off-line generating an analog RoF signal;

S2: carrying out non-uniform quantization on the analog RoF signal to generate a digital PS- (2M+1) ² -QAM/PS- (2M+1) -PAM symbol, wherein the quantization factor is M;

S3: calculating a residual analog quantization error signal after the non-uniform quantization of the analog RoF signal;

S4: performing time domain interleaving on a digital quantized signal of the analog RoF signal and a residual analog quantized error signal to generate a time division multiplexed NUDA-RoF signal;

S5: NUDA-RoF signals are processed by a sending end DSP, sent into a communication experiment system for transmission, and sampled by a receiving end oscilloscope to obtain receiving signals;

S6: the received signal is subjected to time division demultiplexing, NUDA-RoF signal demodulation and OFDM/DMT demodulation by a receiving end DSP to obtain an ultra-high order QAM signal carried by OFDM/DMT;

s7: calculating SNR and EVM of the ultra-high order QAM signal, and evaluating scheme performance;

The specific workflow of the method is as follows: the analog RoF signal is subjected to NUDA-RoF modulation, and a non-uniform digital quantized signal D _1U and a residual analog quantized error signal A _1U are generated; the two parts are interleaved in the time domain through a time division multiplexing technology to generate NUDA-RoF symbols, after being processed by a digital signal at a transmitting end, NUDA-RoF electric signals are generated through an arbitrary waveform generator and are sent into a communication transmission system for transmission; the receiving end samples through a digital oscilloscope, then carries out DSP processing and NUDA-RoF signal demodulation to recover OFDM/DMT signals, then demodulates the OFDM/DMT to recover original high-order QAM signals carried by the OFDM/DMT, and calculates SNR and EVM, thereby evaluating scheme performance.

2. The non-uniform quantized digital-analog radio over fiber method according to claim 1, wherein the analog RoF signal is either an OFDM signal or a DMT signal.

3. The non-uniformly quantized digital analog fiber optic radio method according to claim 1, wherein the analog RoF signal quantized signal is PS-QAM/PS-PAM because the OFDM/DMT signal obeys gaussian distribution.

4. The non-uniform quantization digital analog optical fiber radio method according to claim 1, wherein the distances between constellation points of the generated digital signals are different due to the non-uniform quantization scheme, the quantization interval is small where the signal probability density is high, the constellation points are dense, and conversely, the quantization interval is large where the signal probability density is low, and the constellation points are sparse.

5. The non-uniform quantization digital analog radio over fiber method according to claim 1, wherein the non-uniform quantization scheme is any one of a-law, μ -law, lloyd algorithm.

6. The non-uniform quantized digital analog fiber optic radio method according to claim 1, wherein the method is applicable to a variety of communication systems including, but not limited to, photon assisted millimeter wave terahertz systems, all-electric millimeter wave terahertz systems, polarization multiplexed coherent transmission systems, direct alignment detection systems, multimode/multicore fiber optic transmission systems.

7. The method according to claim 6, wherein in the photon-assisted millimeter wave/terahertz system, NUDA-RoF electrical signals are subjected to electro-optical conversion in an IQ modulator, and are subjected to beat frequency with another optical signal in a photodetector to generate millimeter wave/terahertz radio frequency signals, and free space transmission is performed through an antenna, and after receiving the high frequency electrical signals, the antenna at the receiving end performs down-conversion to an intermediate frequency through a low noise amplifier and a mixer, and then sampling is performed by an oscilloscope.

8. The method of claim 1, wherein the receiving DSP demodulates NUDA-RoF, and OFDM demodulation is the inverse of the transmitting DSP.

9. The method according to claim 1, wherein the modulation parameters of NUDA-RoF include PS-QAM/PS-PAM order, quantization factor size, quantization times, and the like, which can be specifically selected according to different systems, different channel characteristics, and different transmission indexes.