CN116800576A - A 3D-PD-NOMA optical access method based on multiple power distribution - Google Patents

Abstract

The invention discloses a 3D-PD-NOMA optical access method based on multi-power distribution, which comprises the following steps: (1) Dividing input bit data into 1×m groups, where M represents different users; (2) Converting serial data into parallel data through S/P conversion; (3) Performing three-dimensional mapping on the converted data to form a multi-carrier signal; (4) Transforming the multi-carrier signal into a time domain signal by using two-dimensional inverse fast fourier transform; (5) converting the time domain signal into serial data through P/S; (6) Obtaining a transmitted 3D-NOMA signal through PD-NOMA superposition; (7) channel equalizing the received signal; (8) Obtaining a frequency domain signal by utilizing two-dimensional fast Fourier transform; (9) Different signal points are converted into bits according to the rule of three-dimensional mapping of the transmitting end, so that demodulation of the three-dimensional constellation is realized; (10) recovering the output bit data by P/S conversion; the invention reduces the peak-to-average power ratio of the system under the same transmitting power; the system error rate is reduced at the same transmitting power.

Description

A 3D-PD-NOMA optical access method based on multiple power distribution

Technical field

The present invention relates to the field of optical transmission communication technology, and in particular to a 3D-PD-NOMA optical access method based on multiple power distribution.

Background technique

With the advent of the 5G era, global digital traffic demand has exploded. The huge capacity of information in wireless communications will eventually flow into the access network and be transmitted through the backbone network. This puts new requirements on the access capacity of the optical access system and the number of access users. The optical fiber access system is divided into There are two types of active optical networks and passive optical networks. Passive optical network (PON) has been widely used due to its advantages of large bandwidth, low cost, and low power consumption. Passive optical networks have undergone many changes. The most original time division multiplexing passive optical network (time division multiplexing TDM-PON) multiplexes in the time dimension, and the wavelength division multiplexing passive optical network (wavelength division multiplexing WDM-PON) uses Multiplexing in the wavelength dimension. Up to the current Orthogonal Frequency Division Multiplexing (OFDM) technology in the 4G communication system, the above PON technologies require that each resource block (time slot, wavelength, frequency) meet strict requirements. Orthogonality, this orthogonal multiple access method is difficult to meet 5G’s needs for large capacity and multiple access. Non-orthogonal multiple access technology (non-orthogonal multiple access, NOMA) emerged as the times require.

NOMA is regarded as a key technology for the next generation wireless communication system and has received widespread attention from scientific researchers. Compared with other orthogonal multiple access technologies such as traditional OFDM, NOMA signals can be overlapped in the time domain, frequency domain, and power, thus providing greater transmission capacity and higher spectral efficiency. NOMA is mainly divided into sparse coding multiple access in the code domain and power division multiplexing technology in the power domain (Power division NOMA, PD-NOMA). Since non-orthogonal access technologies will overlap different dimensions, additional digital signal processing algorithms are required at the receiving end for demodulation. The sparse coding method in the code domain uses a message propagation algorithm at the receiving end to demodulate based on the message propagation between user nodes and resource nodes. The complexity is extremely high. For 5G large-scale user access, the complexity will double. The non-orthogonal multiple access technology based on power multiplexing superimposes signals of different powers and can dynamically allocate power according to the user channel quality. Higher power is allocated to users who are far away from the OLT, and higher power is allocated to users who are far away from the OLT. Lower optical power, thus ensuring fairness among users. There are already relatively many PD-NOMAs, and experiments have proven that under the same bandwidth conditions, PD-NOMA can achieve twice the transmission capacity compared to traditional OFDM access. The current PD-NOMA is mainly based on OFDM for power superposition. The constellations used are traditional two-dimensional QAM constellations. Under the same transmit power, the minimum Euclidean distance of the constellation is smaller than the minimum Euclidean distance of the three-dimensional constellation, resulting in its error. Code performance is poor.

Contents of the invention

Purpose of the invention: The purpose of the invention is to provide a 3D-PD-NOMA optical access method based on multi-power distribution, using the three-dimensional constellation space for power superposition, and increasing the number of constellation points as much as possible under the condition of limited transmission power. The minimum Euclidean distance between them to improve the transmission performance of the system.

Technical solution: The present invention provides a 3D-PD-NOMA optical access method based on multiple power distribution, which includes the following steps:

On the transmitter side:

(1) Divide the input bit data into 1×M groups, where M represents different users;;

(2) Convert serial data into parallel data through S/P conversion;

(3) Three-dimensional mapping of the converted data to form a multi-carrier signal;

(4) Convert multi-carrier signals into time domain signals using two-dimensional inverse fast Fourier transform;

(5) Convert time domain signals into serial data through P/S;

(6) Obtain the emitted 3D-NOMA signal through PD-NOMA superposition;

On the receiving end:

(7) Channel equalize the received signal;

(8) The equalized signal is a frequency domain signal obtained by using two-dimensional fast Fourier transform;

(9) Convert different signal points into bits according to the three-dimensional mapping rules of the transmitter to achieve demodulation of the three-dimensional constellation;

(10) Restore the output bit data through P/S conversion.

Further, the step (2) is specifically: converting the 1×M groups of original data into an M×N matrix.

Further, the step (3) is specifically: taking two bits in two rows and one column as a group, and using three-dimensional constellation space mapping for each group, and its coordinates can be expressed as:

Four regular tetrahedrons are obtained. The four vertices of each regular tetrahedron correspond to four constellation points respectively. The minimum Euclidean distance between the four points is set to 2. Through the superposition of two three-dimensional constellation points, 16 constellation points are formed. The three-dimensional constellation point distribution.

Further, the step (4) is specifically expressed as:

Among them, 0≤n1≤2,0≤n2≤C-1, C represents the number of carriers; k1 and k2 represent the columns and rows of the OFDM matrix respectively; n1 and n2 are the column sums of the time domain signal matrix after two-dimensional IFFT OK.

Further, the step (5) is specifically: converting the modulated M×N parallel time domain signal into a 1×W matrix for transmission.

Further, the step (6) is specifically expressed as follows:

S(t)＝P1*S ₁ (t)+P2*S ₂ (t)

Among them, S1(t) and S2(t) are two 3D-NOMA signals respectively; P1 and P2 represent high power and low power respectively.

Further, the step (8) is specifically expressed as:

Among them, S' _3D (k1,k2) is the equalized signal, s' ₁ (n2,n1) is the frequency domain signal; the frequency domain signal at this time is the superposition of the high power signal and the low power signal; for the low power signal Demodulation: treat the low-power signal as noise and directly perform QAM symbol demodulation; for high-power signal demodulation: use the SIC algorithm to subtract the high-power signal from the signal at the receiving end to obtain the low-power signal. Demodulate.

Further, the step (9) is specifically expressed as: Assume that the obtained three-dimensional constellation point coordinates are:

According to the three-dimensional mapping rules of the transmitter, different signal points are converted into bits to realize the demodulation of the three-dimensional constellation.

Beneficial effects: Compared with the existing technology, the present invention has the following significant advantages: reducing the peak-to-average power ratio of the system under the same transmission power; reducing the system bit error rate under the same transmission power.

Description of the drawings

Figure 1 is an overall principle block diagram of the present invention;

Figure 2 is a schematic diagram of the constellation superposition of the present invention;

Figure 3 is a constellation diagram of the receiving end of the present invention;

Figure 4 is a comparison of PAPR of the present invention;

Figure 5 is a comparison of the bit error rates of the two-dimensional signal and the three-dimensional signal of the present invention;

Figure 6 shows the different power bit error performance of the present invention.

Detailed ways

The technical solution of the present invention will be further described below with reference to the accompanying drawings.

As shown in Figure 1, an embodiment of the present invention provides a 3D-PD-NOMA optical access method based on multiple power distribution, which includes the following steps:

On the transmitter side:

(1) Divide the input bit data into 1×102400 groups, where M represents different users;

(2) Convert serial data into parallel data through S/P conversion; specifically: convert 1×102400 groups of original data into a 200×512 matrix;

(3) Perform three-dimensional mapping on the converted data to form a multi-carrier signal; specifically: the principle of three-dimensional mapping is shown in Figure 2. In the present invention, the 200×512 bit matrix after serial-to-parallel transformation is used for constellation mapping, with two bits in two rows and one column as a group, and the two bits correspond to the four situations of the constellation points in Figure 2; Figure Figure 2(a) shows the way in which two QPSK signals are superimposed to form 16QAM in traditional 2D-PD-NOMA; Figure 2(b) shows 3D-PD-NOMA. Each power uses a three-dimensional constellation space, and its coordinates can be Expressed as:

The four vertices of the regular tetrahedron correspond to four constellation points respectively, and the minimum Euclidean distance between the four points is set to 2. In this way, through the superposition of two three-dimensional constellation points, a three-dimensional constellation point distribution of 16 constellation points can be formed. Compared with the traditional two-dimensional constellation space, the peak-to-average power ratio of this method is relatively lower, so it will have better Transmission performance. Since the two bits in each column of each two rows correspond to three-dimensional coordinates, the original 200×512 bits are converted into 300×512 three-dimensional coordinates.

(4) Convert multi-carrier signals into time domain signals using two-dimensional inverse fast Fourier transform; specifically expressed as:

Among them, 0≤n1≤2,0≤n2≤C-1, C represents the number of carriers; k1 and k2 represent the columns and rows of the OFDM matrix respectively; n1 and n2 are the column sums of the time domain signal matrix after two-dimensional IFFT OK.

(5) Convert the time domain signal into serial data through P/S; specifically: convert the modulated 300×512 parallel time domain signal into a 1×153600 matrix for transmission.

(6) The emitted 3D-NOMA signal is obtained through PD-NOMA superposition; the specific expression is as follows:

S(t)＝P1*S ₁ (t)+P2*S ₂ (t)

Among them, S1(t) and S2(t) are two 3D-NOMA signals respectively; P1 and P2 represent high power and low power respectively. As shown in Figure 2(b), the superimposed constellation diagram of the two channels just makes up a 16QAM.

On the receiving end:

(7) Channel equalize the received signal;

(8) The equalized signal is a frequency domain signal obtained by using two-dimensional fast Fourier transform; as shown in Figure 3, the specific expression is:

Among them, S' _3D (k1,k2) is the equalized signal, s' ₁ (n2,n1) is the frequency domain signal; the frequency domain signal at this time is the superposition of the high power signal and the low power signal; for the low power signal Demodulation: treat the low-power signal as noise and directly perform QAM symbol demodulation; for high-power signal demodulation: use the SIC algorithm to subtract the high-power signal from the signal at the receiving end to obtain the low-power signal. Demodulate.

(9) Convert different signal points into bits according to the three-dimensional mapping rules of the transmitter to realize the demodulation of the three-dimensional constellation; the specific expression is: Assume that the coordinates of the obtained three-dimensional constellation points are:

According to the three-dimensional mapping rules of the transmitter, different signal points are converted into bits to realize the demodulation of the three-dimensional constellation.

(10) Restore the output bit data through P/S conversion.

As shown in Figure 4, in order to test the superiority of the 3D-PD-NOMA solution proposed in the present invention, the PAPR and bit error rate performance of two-dimensional and three-dimensional signals were compared. It can be clearly found from Figure 4 that the PAPR of the three-dimensional signal is significantly lower than the PAPR of the two-dimensional signal. This is because the power difference between the three-dimensional constellation points is smaller than that of the two-dimensional constellation points, which leads to a reduction in PAPR.

As shown in Figure 5, in order to compare the signal transmission quality of the two signals under signal-to-noise ratio conditions, simulated channels with signal-to-noise ratios of 1-20 were simulated. It can be found that the 3D-PD-NOMA proposed by the present invention The signal has better transmission performance than traditional 2D-NOMA. This is because under the same transmit power, the three-dimensional constellation points have a larger minimum Euclidean distance than the two-dimensional constellation, which is more conducive to the receiving end's decision-making and therefore has better bit error performance. When the bit error rate is 10 ^-3 , the transmission performance of the three-dimensional constellation is 4dB better than that of the two-dimensional constellation.

As shown in Figure 6, the present invention compares the bit error rate curves of two channels with different powers. It can be found that because the high-power signal is allocated higher power, it has better transmission performance, while the low-power signal has better transmission performance due to the allocated power. The power is relatively low and is more affected by noise, so its transmission effect is relatively poor. However, for the NOMA system, high-power signals are allocated to users farther away from the OLT, and low-power signals are allocated to users closer to the OLT, thereby ensuring fairness among different users. Based on the good transmission performance of 3D-PD-NOMA and ensuring fairness between different users, the 3D-PD-NOMA proposed in the present invention has very good application prospects in future optical access systems.

Claims (8)

1. A 3D-PD-NOMA optical access method based on multiple power distribution, including a transmitter and a receiver, characterized in that it includes the following steps: On the transmitter side: (1) Divide the input bit data into 1×M groups, where M represents different users; (2) Convert serial data into parallel data through S/P conversion; (3) Three-dimensional mapping of the converted data to form a multi-carrier signal; (4) Convert multi-carrier signals into time domain signals using two-dimensional inverse fast Fourier transform; (5) Convert time domain signals into serial data through P/S; (6) Obtain the emitted 3D-NOMA signal through PD-NOMA superposition; On the receiving end: (7) Channel equalize the received signal; (8) The equalized signal is a frequency domain signal obtained by using two-dimensional fast Fourier transform; (9) Convert different signal points into bits according to the three-dimensional mapping rules of the transmitter to achieve demodulation of the three-dimensional constellation; (10) Restore the output bit data through P/S conversion. 2. A 3D-PD-NOMA optical access method based on multiple power distribution according to claim 1, characterized in that the step (2) is specifically: converting 1×M groups of original data into M× N matrix. 3. A 3D-PD-NOMA optical access method based on multiple power distribution according to claim 1, characterized in that the step (3) is specifically: taking two bits in two rows and one column as a group, for each Each group adopts three-dimensional constellation space mapping, and its coordinates can be expressed as: Four regular tetrahedrons are obtained. The four vertices of each regular tetrahedron correspond to four constellation points respectively. The minimum Euclidean distance between the four points is set to 2. Through the superposition of two three-dimensional constellation points, 16 constellation points are formed. The three-dimensional constellation point distribution. 4. A 3D-PD-NOMA optical access method based on multiple power distribution according to claim 1, characterized in that the step (4) is specifically expressed as: Among them, 0≤n1≤2,0≤n2≤C-1, C represents the number of carriers; k1 and k2 represent the columns and rows of the OFDM matrix respectively; n1 and n2 are the column sums of the time domain signal matrix after two-dimensional IFFT OK. 5. A 3D-PD-NOMA optical access method based on multiple power distribution according to claim 1, characterized in that the step (5) is specifically: converting the modulated M×N parallel time domain signal into 1×W matrix for transmission. 6. A 3D-PD-NOMA optical access method based on multiple power distribution according to claim 1, characterized in that the step (6) is specifically expressed as follows: S(t)＝P1*S ₁ (t)+P2*S ₂ (t) Among them, S1(t) and S2(t) are two 3D-NOMA signals respectively; P1 and P2 represent high power and low power respectively. 7. A 3D-PD-NOMA optical access method based on multiple power distribution according to claim 1, characterized in that the step (8) is specifically expressed as: Among them, S' _3D (k1,k2) is the equalized signal, s' ₁ (n2,n1) is the frequency domain signal; the frequency domain signal at this time is the superposition of the high power signal and the low power signal; for the low power signal Demodulation: treat the low-power signal as noise and directly perform QAM symbol demodulation; for high-power signal demodulation: use the SIC algorithm to subtract the high-power signal from the signal at the receiving end to obtain the low-power signal. Demodulate. 8. A 3D-PD-NOMA optical access method based on multiple power distribution according to claim 1, characterized in that the step (9) is specifically expressed as: Assume that the obtained three-dimensional constellation point coordinates are: According to the three-dimensional mapping rules of the transmitter, different signal points are converted into bits to realize the demodulation of the three-dimensional constellation.