Power allocation method and system based on maximizing minimum Euclidean distance in NOMA non-terrestrial network

Abstract

The present invention relates to the field of communication technology, and discloses a power allocation method and system based on maximizing the minimum Euclidean distance in a NOMA non-terrestrial network. The method includes: selecting a modulation constellation of a user; establishing a transmission signal model formed by superimposing users p and q; establishing an optimal power allocation coefficient objective function by maximizing the minimum Euclidean distance of the superimposed constellation, and solving the optimal power allocation problem; calculating the minimum Euclidean distance of users p and q in the superimposed constellation, and the maximum real part value from the origin to the user and q modulation constellation points in the superimposed constellation; when the minimum Euclidean distance d between satellites is inter Equal to the minimum Euclidean distance d within the star intra When the minimum Euclidean distance between the superimposed constellation points is maximized, the optimal power allocation coefficient is calculated, and fixed power is pre-allocated to the user. 2n ‑QAM or non-square 2 2n+1 The scheme of superimposing ‑QAM modulation constellations solves the optimal power allocation coefficient with the goal of maximizing the minimum Euclidean distance of the superimposed constellations, reducing the system complexity.

Description

Technical Field

[0001] The present invention relates to the field of non-orthogonal multiple access non-terrestrial network communication technology, and in particular to a power allocation method and system based on maximizing the minimum Euclidean distance in a NOMA non-terrestrial network. Background Art

[0002] As more and more wireless mobile devices need to access wireless communication networks, the fifth generation (5G) network provides communication services for a large number of users. However, for users in remote areas without ground base station coverage, wireless communication services are not easy to obtain. In the next generation communication system, satellite communication systems as non-terrestrial networks (NTN) are considered to be an important supplement to the ground 5G network. They can provide wireless communication services to users in remote areas, alleviate the congestion of ground networks, and make "anytime, anywhere" communication feasible and achievable. However, in the face of the demand for massive connections, traditional orthogonal multiple access (OMA) limits the number of users that can be served due to the orthogonality of resources. In this context, non-orthogonal multiple access (NOMA) technology is considered to be a promising multiple access technology in 5G because of its simultaneous sharing of time / frequency resources. Its advantages are: higher spectrum efficiency, higher throughput and lower latency.

[0003] NOMA can be mainly divided into power domain NOMA (PD-NOMA) and code domain sparse code multiple access (SCMA). The decoding performance of the PD-NOMA scheme depends largely on the power allocation scheme when users share the channel. Therefore, power allocation is a key issue that NOMA technology needs to solve. At present, most of the research on the power allocation coefficient of NOMA networks is aimed at maximizing the sum rate, which needs to be updated in real time according to the constellation constraint capacity and channel state information (CSI). This method is highly complex and is undoubtedly a big challenge for satellite networks.

[0004] Maximizing mutual information can be approximated as maximizing the minimum Euclidean distance of any input constellation. However, in the absence of CSI, there are few studies on NOMA power allocation algorithms using QAM constellation modulation. In the existing literature, the NOMA power allocation algorithm with the goal of maximizing the minimum Euclidean distance is only applicable to 2 2n -QAM constellation, not applicable to 2 2n+1 Regardless of the channel conditions, to achieve the same bit error rate (BER), 2 2(n+1) -QAM requires a signal-to-noise ratio (SNR) of 2 2n -QAM needs 6dB more; when SNR is just at 2 2n -QAM and 2 2(n+1)-QAM, odd-order QAM modulation is required to maximize the channel capacity while ensuring BER performance. 2(n +1) Low-complexity power allocation schemes for -QAM modulation and without CSI are still worth studying. Summary of the invention

[0005] In order to solve the problems existing in the prior art, the present invention provides a power allocation method and system based on maximizing the minimum Euclidean distance in a NOMA non-terrestrial network, which solves the optimal power allocation coefficient with the goal of maximizing the minimum Euclidean distance of the superimposed constellation, and then performs fixed power pre-allocation, thereby reducing the system complexity. The method is suitable for square and non-square modulation constellation superposition schemes and has a wide range of applications.

[0006] The technical solution adopted in the embodiment of the present invention is: a power allocation method based on maximizing the minimum Euclidean distance in a NOMA non-terrestrial network, comprising the following steps:

[0007] S1. Select the modulation constellation S of user p p and the modulation constellation S of user q q ;

[0008] S2. Establish a transmission signal model formed by superposition of user p and user q;

[0009] S3, by maximizing the minimum Euclidean distance of the superimposed constellations, an optimal power allocation coefficient objective function is established to solve the optimal power allocation problem;

[0010] S4, calculating the minimum Euclidean distance between user p and user q in the superposition constellation;

[0011] S5, calculating the maximum real part value from the origin to the modulation constellation points of user p and user q in the superposition constellation;

[0012] S6. Calculate the minimum Euclidean distance d between stars inter , the minimum Euclidean distance within the star d intra ;

[0013] S7, when d inter =d intra When , the minimum Euclidean distance between the superimposed constellation points reaches the maximum value, and the optimal power allocation coefficient is calculated;

[0014] S8. Output the optimal power allocation coefficient result, which is used in the transmission signal model established in step S2 to perform fixed power pre-allocation for users.

[0015] Preferably, user p uses an asymmetric modulation constellation, and the bit data of user p uses a size of N at the transmitting end. pThe constellation is modulated and mapped; user q uses a symmetrical modulation constellation, and the bit data of user q uses a size of N at the transmitting end. q The constellation is modulated and mapped.

[0016] Preferably, the transmission signal model established in step S2 is:

[0017]

[0018] Among them, α p , α q are the power allocation coefficients for user p and user q, respectively, p +α q =1,α p , α q ∈1,P s is the satellite transmission power; x p is the modulated constellation signal after phase rotation, x p ∈S p ;x q is the constellation signal modulated by user q, x q ∈S q ; and:

[0019]

[0020] in is the constellation signal modulated by user p, with the phase rotation

[0021] Preferably, the optimal power allocation problem m in step S3 opt Described as:

[0022]

[0023] Among them, α q is the power allocation coefficient of user q, α p is the power allocation coefficient of user p; d min is the minimum Euclidean distance between the superimposed constellation points, specifically:

[0024] d min =min{d inter ,d intra}

[0025] d inter The minimum Euclidean distance between constellation points in different quadrants of the superimposed constellation is called the minimum Euclidean distance between constellations; d intra It is the minimum Euclidean distance between different constellation points in the same quadrant, called the minimum intra-constellation Euclidean distance.

[0026] The technical solution adopted in another embodiment of the present invention is: a power allocation system based on maximizing the minimum Euclidean distance in a NOMA non-terrestrial network, comprising the following modules:

[0027] Modulation constellation selection module, used to select the modulation constellation S of user p p and the modulation constellation S of user q q ;

[0028] A signal model building module, used to build a transmission signal model formed by superposition of user p and user q;

[0029] The power allocation module is used to solve the optimal power allocation problem by maximizing the minimum Euclidean distance of the superimposed constellation to establish the optimal power allocation coefficient objective function; calculate the minimum Euclidean distance between user p and user q in the superimposed constellation; calculate the maximum real part value from the origin to the modulation constellation point of user p and user q in the superimposed constellation; calculate the minimum Euclidean distance d between satellites inter 、The minimum Euclidean distance within the star d intra When d inter =d intra When , the minimum Euclidean distance between the superimposed constellation points reaches the maximum value, and the optimal power allocation coefficient is calculated;

[0030] The output module outputs the optimal power allocation coefficient result, which is used in the established transmission signal model to pre-allocate fixed power to users.

[0031] Compared with the prior art, the technical effects achieved by the present invention include:

[0032] 1. The present invention improves the power allocation algorithm used in NOMA non-terrestrial network systems. In the absence of channel state information (CSI) feedback, the optimal power allocation coefficient is solved with the goal of maximizing the minimum Euclidean distance of the superimposed constellation, and then fixed power pre-allocation is performed; compared with the maximum constellation constrained capacity (MCCC) power allocation method that relies on CSI feedback, the complexity of the system is reduced and a similar MI curve can be obtained.

[0033] 2. The power distribution method of the present invention is not only applicable to square 2 2n -QAM modulation constellation diagram superposition scheme, also applicable to non-square 2 2n+1 -QAM modulation constellation diagram superposition scheme. Taking superposition 32QAM as an example, the new 8QAM+QPSK superposition 32QAM scheme has a lower peak-to-average power ratio (PAPR) than square 8QAM+QPSK, traditional 8QAM+QPSK and 8PSK+QPSK superposition 32QAM. Under the optimal power allocation coefficient, the new 8QAM+QPSK superposition scheme has better bit error rate (BER) performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 Schematic diagram of satellite NOMA downlink in an embodiment of the present invention;

[0035] Figure 2 Flow chart of the MMED optimal power allocation method according to an embodiment of the present invention;

[0036] Figure 3 Schematic diagrams of different superimposed 32QAM constellation scheme designs, wherein (a) is a schematic diagram of a new 8QAM+QPSK modulation constellation point scheme proposed in an embodiment of the present invention, (b) is a schematic diagram of an existing square 8QAM+QPSK modulation constellation point scheme, (c) is a schematic diagram of an existing 8PSK+QPSK modulation constellation point scheme, and (d) is a schematic diagram of a traditional 8QAM+QPSK modulation constellation point scheme;

[0037] Figure 4 A complementary cumulative distribution function curve diagram of PAPR of the new 8QAM+QPSK superposition scheme and other superposition 32QAM schemes in an embodiment of the present invention;

[0038] Figure 5 A user p bit error rate curve diagram of the new 8QAM+QPSK superposition scheme and other superposition 32QAM schemes at optimal power in an embodiment of the present invention;

[0039] Figure 6 A user q bit error rate curve diagram of the new 8QAM+QPSK superposition scheme and other superposition 32QAM schemes at optimal power in an embodiment of the present invention;

[0040] Figure 7 It is a MI curve diagram of the power allocation method based on the new 8QAM+QPSK superposition scheme according to the embodiment of the present invention and other existing different methods. DETAILED DESCRIPTION

[0041] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings and embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] Example 1

[0043] This embodiment provides a power allocation method based on maximizing the minimum Euclidean distance (MMED) in a NOMA non-terrestrial network. In this embodiment, the satellite NOMA downlink scenario is as follows Figure 1As shown in Figure 1, it is assumed that both the ground user and the satellite are equipped with a single antenna, user p is close to the beam center, and user q is located at the beam edge. NOMA of the two users is achieved through different power allocations.

[0044] like Figure 2 As shown, this embodiment is based on a power allocation method for maximizing the minimum Euclidean distance, which is used for a NOMA satellite downlink and includes the following steps:

[0045] S1. Select the modulation constellation S of user p p and the modulation constellation S of user q q ;

[0046] In this step, the modulation constellation S of user k (k = p, q) is selected k , the bit data of user k is transmitted in the form of N k The constellation is modulated and mapped, and the modulation constellation used is expressed as:

[0047] s k ={x k (n)|n=1,2,…,N k}

[0048] Among them, x k (n) represents the nth modulation constellation point of user k, N k is the number of modulation constellation points.

[0049] In a preferred embodiment, user p uses asymmetric 8QAM (N p =8) modulation constellation, user q uses symmetrical QPSK (N q =4) modulation constellation; and compare different superposition 32QAM schemes, such as Figure 3 As shown, the modulation constellation points of the asymmetric 8QAM constellation and the symmetric QPSK are specifically shown in Table 1 below, and the comparison of different parameters of different superimposed 32QAM schemes is shown in Table 2 below.

[0050] Table 1

[0051]

[0052] Table 2

[0053]

[0054] Depend on Figure 3 As can be seen from the table above, the power allocation method proposed in this embodiment is applicable to 2 (2n+1) The order superposition constellation reduces the complexity and can obtain the mutual information similar to the MCCC power allocation method.

[0055] S2, establish a transmission signal model;

[0056] This step establishes a satellite transmission signal model, which is formed by superimposing user p and user q. Specifically, it is:

[0057]

[0058] Among them, α p , α q are the power allocation coefficients for user p and user q, respectively, p +α q =1,α p , α q ∈1,P s is the satellite transmission power, x p ∈S p , x q ∈S q . The user's signal power E[x p ]=1,E[x q ]=1, satellite transmission power E[X]=P s .

[0059] It is worth noting that due to Figure 3 The new 8QAM signal used in the neutron diagram (a) is asymmetric. In order to achieve the target superposition of the 32QAM constellation diagram, in this embodiment, the new 8QAM signal and the QPSK signal need to be jointly encoded, and the rotation angle of the new 8QAM signal is determined by the QPSK signal. Therefore, in this embodiment, the 8QAM signal after phase rotation is specifically in is the 8QAM signal modulated by user p; x p is the 8QAM signal after phase rotation; the rotation phase x q It is the QPSK signal modulated by user q.

[0060] S3, establishing an optimal power allocation coefficient objective function;

[0061] Generally speaking, the power allocation method of NOMA network that relies on real-time channel state information feedback is highly complex. In order to reduce the complexity of power allocation in NOMA network so that it can be better used in satellite communication systems, this step performs optimal power allocation for different users with the optimization goal of maximizing the minimum Euclidean distance of the superimposed constellation.

[0062] This step maximizes the minimum Euclidean distance of the superimposed constellation to establish the optimal power allocation coefficient objective function to solve the optimal power allocation problem. opt It can be described as:

[0063]

[0064] Among them, α q is the power allocation coefficient of user q, α p is the power allocation coefficient of user p; d min is the minimum Euclidean distance between the superimposed constellation points, specifically:

[0065] d min =min{d inter ,d intra}

[0066] d inter The minimum Euclidean distance between constellation points in different quadrants of the superimposed constellation is called the minimum Euclidean distance between constellations; d intra It is the minimum Euclidean distance between different constellation points in the same quadrant, called the minimum intra-constellation Euclidean distance.

[0067] S4, calculating the minimum Euclidean distance of user k in the superposition constellation;

[0068] Through the optimal power allocation problem opt After the optimal power allocation coefficient objective function is established, the power allocation method between different users is related to the minimum Euclidean distance of the superposition constellation. This step calculates the minimum Euclidean distance d of user k (k = p, q) in the superposition constellation. k Specifically:

[0069]

[0070]

[0071] Among them, α k is the power allocation coefficient of user k (k = p, q), d 0,k is the minimum Euclidean distance after normalization of the modulation constellation of user k, x k (i), x k (j) represent the i-th and j-th modulation constellation points of user k, respectively, β k is the normalized power coefficient of the modulation constellation.

[0072] S5, calculating the maximum real part value from the origin to the modulation constellation point of user k in the superposition constellation;

[0073] This step calculates the maximum real value of the modulation constellation point from the origin to user k (k = p, q) in the superposition constellation. Specifically:

[0074]

[0075] Among them, x k (n)(n=1,2,…,N k) represents the nth modulation constellation point of user k, is the maximum real value from the origin to the modulation constellation point of user k in the superposition constellation, represents the maximum real value of the modulation constellation point of user k, β k is the modulation constellation S k The normalized power coefficient of Represents the real part operation.

[0076] S6. Calculate the minimum Euclidean distance d between stars inter 、The minimum Euclidean distance within the star d intra ;

[0077] This step calculates the minimum Euclidean distance d between the stars of the superimposed constellation inter and the minimum Euclidean distance within the star d intra , specifically:

[0078]

[0079] S7, when d inter =d intra Calculate the optimal power allocation coefficient when ;

[0080] In this step, let d inter =d intra When min Get the maximum value, specifically:

[0081]

[0082] From this we can get the optimal power allocation problem m opt The solution is:

[0083]

[0084] S8. Output the optimal power allocation coefficient result, specifically:

[0085]

[0086] After obtaining the optimal power allocation coefficient result, it is used in the satellite transmission signal model established by S2 to pre-allocate fixed power to users, reduce the complexity of power allocation, and then demodulate the received signal at the receiving end to restore the original bit data.

[0087] The power allocation coefficient result is allocated to different users, and the transmission signal model after superposition in step S2 can be obtained, and then it reaches the receiving end through channel transmission. The receiving end signal model is specifically:

[0088] Y k =H k X+Z k

[0089] Among them, Y k represents the signal received by user k, H k represents the channel vector from satellite to ground user k, Z k Indicates that the mean is 0 and the variance is σ 2 Additive white Gaussian noise.

[0090] H k Specifically:

[0091]

[0092] Among them, G s is the satellite antenna gain; h k is the attenuation channel gain from satellite to user k; is the free space loss coefficient, c is the speed of light, f c is the optical frequency, D k is the distance from the satellite to user k; is the angle between user k and the beam center relative to the satellite; is the beam gain of user k, specifically:

[0093]

[0094] Among them, G k is the antenna gain of user k, J1(·) is the first-order Bessel function, J3(·) is the third-order Bessel function, is the 3dB angle between the satellite and the ground user.

[0095] The attenuation channel adopts the SR channel model, and the attenuation channel gain |h k | 2 The probability density function of is:

[0096]

[0097] Where k represents user k (k = p, q), Ω k and 2b k are the average power of the visible component and the multipath component, respectively, m k is the Nakagami-m parameter, 1F1(a; b; x) is the confluent hypergeometric function; as an example, h q is the attenuation channel gain from satellite to user q, h p is the attenuation channel gain from the satellite to user p, assuming that |h q | 2 <|h p | 2, the satellite transmission signal experiences light shadow (LS) and average shadow (AS) to reach user p and user q respectively, and the light shadow LS has a b k 、m k ,Ω k They are 0.158, 19.4, and 1.29 respectively, and the average shadow AS b k 、m k ,Ω k They are 0.126, 10.1 and 0.835 respectively.

[0098] Preferably, G s =24.3dBi, the antenna gain of user p and user q is G p =G q =3.5dBi, The angle between user p and the beam center The angle between user q and the beam center The distance from the satellite to user p and user q is D p =D q =600km.

[0099] At the receiving end, the QPSK signal is directly demodulated at the high-power user q end, and the demodulated signal is specifically:

[0100]

[0101] H q represents the channel vector from satellite to ground user q, Y q Represents the signal received by user q.

[0102] The high-power signal is eliminated by using the interference cancellation method (SIC) at the low-power user p end. The SIC method is divided into two steps. The first step is to detect the high-power user signal. The high-power signal detected is specifically:

[0103]

[0104] H p represents the channel vector from satellite to ground user p, Y p represents the signal received by user p.

[0105] The second step is to eliminate the interference of high-power signals, specifically:

[0106]

[0107] It is worth noting that in Figure 3 In the superposition scheme shown in sub-figure (a), it is necessary to detect the high-power signal Estimate the rotation phase of the new 8QAM signal. The rotation phase is specifically Then the new 8QAM signal can be expressed as

[0108] The detected QAM signal is then demodulated to restore the user's original bit data.

[0109] like Figure 4 is a complementary cumulative distribution function curve of the peak-to-average ratio (PAPR) of different superimposed 32QAM schemes. According to the calculation method in steps S4 and S5, different superimposed 32QAM schemes d are calculated. 0,p ,d 0,q , m opt , maximum MED and PAPR, as shown in Table 1; according to m in Table 1 opt Perform power pre-allocation, such as Figure 5 and Figure 6 The bit error rate performance of different users under different superimposed 32QAM schemes is compared under the optimal power allocation coefficient. Figure 7 The mutual information curves of the MMED-based power allocation method of this embodiment and the existing MCCC-based power allocation method are compared.

[0110] Example 2

[0111] This embodiment is based on the same inventive concept as Embodiment 1, and proposes a power allocation system based on maximizing the minimum Euclidean distance in a NOMA non-terrestrial network, including the following modules:

[0112] Modulation constellation selection module, used to select the modulation constellation S of user p p and the modulation constellation S of user q q ;

[0113] A signal model building module, used to build a transmission signal model formed by superposition of user p and user q;

[0114] The power allocation module is used to solve the optimal power allocation problem by maximizing the minimum Euclidean distance of the superimposed constellation to establish the optimal power allocation coefficient objective function; calculate the minimum Euclidean distance between user p and user q in the superimposed constellation; calculate the maximum real part value from the origin to the modulation constellation point of user p and user q in the superimposed constellation; calculate the minimum Euclidean distance d between satellites inter , the minimum Euclidean distance within the star d intra When d inter =d intra When , the minimum Euclidean distance between the superimposed constellation points reaches the maximum value, and the optimal power allocation coefficient is calculated;

[0115] The output module outputs the optimal power allocation coefficient result, which is used in the established transmission signal model to pre-allocate fixed power to users.

[0116] The above modules of this embodiment are respectively used to implement the corresponding steps of Embodiment 1. For the detailed implementation process, please refer to Embodiment 1.

[0117] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

[0118] Those skilled in the art can understand that all or part of the processes in the above embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer or a readable storage medium. When the program is executed, it can include the processes in the above embodiments. The storage medium can be a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

Claims

1. A power allocation method based on maximizing the minimum Euclidean distance in a NOMA non-terrestrial network, characterized in that: The following steps are involved: S1. Select the modulation constellation S of user p. p and the modulation constellation S of user q q ; S2. Establish a transmission signal model formed by superposition of user p and user q; S3, by maximizing the minimum Euclidean distance of the superimposed constellations, an optimal power allocation coefficient objective function is established to solve the optimal power allocation problem; S4, calculating the minimum Euclidean distance between user p and user q in the superposition constellation; S5, calculating the maximum real part value from the origin to the modulation constellation points of user p and user q in the superposition constellation; S6. Calculate the minimum Euclidean distance d between stars inter , the minimum Euclidean distance within the star d intra ; S7, when d inter =d intra When , the minimum Euclidean distance between the superimposed constellation points reaches the maximum value, and the optimal power allocation coefficient is calculated; S8, outputting the optimal power allocation coefficient result, which is used in the transmission signal model established in step S2 to pre-allocate fixed power to the user; The transmission signal model established in step S2 is: Among them, α p , α q are the power allocation coefficients for user p and user q, respectively, p +α q =1,P s is the satellite transmission power; x p is the modulated constellation signal after phase rotation, x p ∈S p ;x q is the constellation signal modulated by user q, x q ∈S q ; and: in is the constellation signal modulated by user p, with the phase rotation The optimal power allocation problem in step S3 is opt Described as: Among them, α q is the power allocation coefficient of user q, α p is the power allocation coefficient of user p; d min is the minimum Euclidean distance between the superimposed constellation points, specifically: the min =min{d inter ,d intra } d inter The minimum Euclidean distance between constellation points in different quadrants of the superimposed constellation is called the minimum Euclidean distance between constellations; d intra It is the minimum Euclidean distance between different constellation points in the same quadrant, called the minimum intra-constellation Euclidean distance.

2. The power distribution method according to claim 1, characterized in that: User p uses an asymmetric modulation constellation. The bit data of user p is transmitted using a constellation of size N. p The constellation is modulated and mapped; user q uses a symmetrical modulation constellation, and the bit data of user q uses a size of N at the transmitting end. q The constellation is modulated and mapped.

3. The power distribution method according to claim 1, characterized in that: Step S4 calculates the minimum Euclidean distance d of user k in the superposition constellation k Specifically: where k = p, q, α k is the power allocation coefficient of user k, d 0,k is the minimum Euclidean distance after normalization of the modulation constellation of user k, x k (i), x k (j) represent the i-th and j-th modulation constellation points of user k, respectively, β k is the normalized power coefficient of the modulation constellation; Step S5 calculates the maximum real value from the origin to the user k modulation constellation point in the superposition constellation Specifically: Among them, x k (n) represents the nth modulation constellation point of user k, n = 1, 2, ..., N k , is the maximum real value from the origin to the modulation constellation point of user k in the superposition constellation, represents the maximum real value of the modulation constellation point of user k, represents the real part operation; The minimum Euclidean distance between stars d inter , the minimum Euclidean distance within the star d intra for: Let d inter =d intra When min Get the maximum value, specifically: The optimal power allocation problem m is obtained opt The solution is: The optimal power allocation coefficient is:

4. The power distribution method according to claim 1, characterized in that: After allocating the power allocation coefficient results to different users, the transmitted signal model after superposition in step S2 is obtained, and then transmitted through the channel to the receiving end. The receiving end signal model is specifically: Y k =H k X+Z k Among them, Y k represents the signal received by user k, H k represents the channel vector from satellite to ground user k, Z k Indicates that the mean is 0 and the variance is σ 2 Additive Gaussian white noise; H k Specifically: Among them, G s is the satellite antenna gain; h k is the attenuation channel gain from satellite to user k; is the free space loss coefficient, c is the speed of light, f c is the optical frequency, D k is the distance from the satellite to user k; is the angle between user k and the beam center relative to the satellite; is the beam gain of user k, specifically: Among them, G k is the antenna gain of user k, J1(·) is the first-order Bessel function, J3(·) is the third-order Bessel function, is the 3dB angle between the satellite and the ground user.

5. A power allocation system based on maximizing the minimum Euclidean distance in a NOMA non-terrestrial network, characterized in that: Includes the following modules: Modulation constellation selection module, used to select the modulation constellation S of user p p and the modulation constellation S of user q q ; A signal model building module, used to build a transmission signal model formed by superposition of user p and user q; The power allocation module is used to solve the optimal power allocation problem by maximizing the minimum Euclidean distance of the superimposed constellation to establish the optimal power allocation coefficient objective function; calculate the minimum Euclidean distance between user p and user q in the superimposed constellation; calculate the maximum real part value from the origin to the modulation constellation point of user p and user q in the superimposed constellation; calculate the minimum Euclidean distance d between satellites inter , the minimum Euclidean distance within the star d intra When d inter =d intra When , the minimum Euclidean distance between the superimposed constellation points reaches the maximum value, and the optimal power allocation coefficient is calculated; An output module outputs the optimal power allocation coefficient result, which is used in the established transmission signal model to pre-allocate fixed power to users; The established transmission signal model is: Among them, α p , α q are the power allocation coefficients for user p and user q respectively, α p +α q =1,P s is the satellite transmission power; x p is the modulated constellation signal after phase rotation, x p ∈S p ;x q is the constellation signal modulated by user q, x q ∈S q ; and: in is the constellation signal modulated by user p, with the phase rotation Optimal power allocation problem opt Described as: Among them, α q is the power allocation coefficient of user q, α p is the power allocation coefficient of user p; d min is the minimum Euclidean distance between the superimposed constellation points, specifically: the min =min{d inter ,d intra } d inter The minimum Euclidean distance between constellation points in different quadrants of the superimposed constellation is called the minimum Euclidean distance between constellations; d intra It is the minimum Euclidean distance between different constellation points in the same quadrant, called the minimum intra-constellation Euclidean distance.

6. The power distribution system according to claim 5, characterized in that: Calculate the minimum Euclidean distance d of user k in the superposition constellation k Specifically: where k = p, q, α k is the power allocation coefficient of user k, d 0,k is the minimum Euclidean distance after normalization of the modulation constellation of user k, x k (i), x k (j) represent the i-th and j-th modulation constellation points of user k, respectively, β k is the normalized power coefficient of the modulation constellation; Calculate the maximum real value from the origin to the user k modulation constellation point in the superposition constellation Specifically: Among them, x k (n) represents the nth modulation constellation point of user k, n = 1, 2, ..., N k , is the maximum real value from the origin to the modulation constellation point of user k in the superposition constellation, represents the maximum real value of the modulation constellation point of user k, represents the real part operation; The minimum Euclidean distance between stars d inter , the minimum Euclidean distance within the star d intra for: Let d inter =d intra When min Get the maximum value, specifically: The optimal power allocation problem m is obtained opt The solution is: The optimal power allocation coefficient is:

7. The power distribution system according to claim 5, characterized in that: User p uses an asymmetric modulation constellation. The bit data of user p is transmitted using a constellation of size N. p The constellation is modulated and mapped; user q uses a symmetrical modulation constellation, and the bit data of user q uses a size of N at the transmitting end. q The constellation is modulated and mapped.

Images