SCMA codebook design method and communication method for non-ground network

Abstract

The invention discloses a non-ground network-oriented SCMA codebook design method and a communication method, and belongs to the field of Internet of Things communication, the codebook design method comprises the following steps: sorting users in an ascending order according to the distance between the users and a nadir point of an air base station, allocating df power factors to each orthogonal resource, and calculating # imgabs0 # 1 < = i < = df; generating a set P composed of K-dimensional column vectors including N 1 and K-N 0, dividing the set P into df orthogonal groups, and alternately arranging the remaining column vectors and the df orthogonal groups to form a set P '; after qi is sorted in an ascending order according to rho i, the ith element 1 in each row of P'is replaced by qi, and column vectors in the P 'are sorted in an ascending order according to the column vector power sum and then serve as a signature matrix FK * J; and constructing an SCMA codebook of each user based on the FK * J, calculating the worst bit error rate so as to enable the worst bit error rate to be minimum as an optimization target, solving corresponding parameters, and obtaining the SCMA codebook of each user. According to the invention, the problem of unfairness of users caused by channel difference in a non-ground network can be solved, and offline optimization of the SCMA codebook is realized.

Description

Technical Field

[0001] The present invention belongs to the field of Internet of Things communications, and more specifically, to an SCMA codebook design method and a communication method for non-terrestrial networks. Background Art

[0002] With the rapid development of IoT technology, the next generation of communication systems is committed to building a communication environment that can widely connect various IoT devices, carry massive user data, and achieve efficient spectrum resource utilization. In order to realize the vision of 6G Internet of Everything, the current IoT system is actively expanding its service scope and further using non-terrestrial network (NTN) infrastructure as a supplementary means to expand global coverage and supplement the shortcomings of terrestrial networks.

[0003] However, the traditional communication method based on orthogonal multiple access has difficulties in achieving large-scale connections because the number of orthogonal resources limits the total number of users that the system can accommodate. In order to solve the problem of short-term high-concurrency communication in the Internet of Things system, SCMA (Sparse Code Multiple Access) has emerged as a new code domain non-orthogonal multiple access technology. SCMA greatly improves the resource reuse rate and the number of concurrent access users by directly mapping the instantaneous input information of each user into overlapping multi-dimensional codewords in the sparse codebook. In addition, the performance of the SCMA system mainly depends on the sparse codebook design scheme. Specific optimization schemes for applicable scenarios can effectively reduce multi-user interference and improve the system bit error rate performance.

[0004] Although SCMA is regarded as an effective solution for achieving high-concurrency access to the terrestrial Internet of Things, unlike the fixed positions of base stations in the terrestrial Internet of Things, aerial base stations in non-terrestrial networks have high-speed mobility, and the corresponding channel scenarios also change. The existing SCMA codebook design method for the terrestrial Internet of Things will have obvious channel differences and unfair user services in non-terrestrial network scenarios. In addition, the hardware cost and power consumption in non-terrestrial networks are limited, and the existing SCMA codebook design method involves complex online optimization calculations and cannot be effectively applied to non-terrestrial networks.

[0005] In general, it is of great significance to propose an SCMA codebook design method that can effectively solve the channel difference and user fairness problems in non-terrestrial networks and achieve offline optimization. Summary of the invention

[0006] In view of the defects of the prior art and the need for improvement, the present invention provides an SCMA codebook design method and a communication method for non-terrestrial networks. The purpose is to propose an SCMA codebook design method that can effectively solve the channel difference and user fairness problems in non-terrestrial networks and can achieve offline optimization.

[0007] To achieve the above object, according to one aspect of the present invention, a SCMA codebook design method for a non-terrestrial network is provided, comprising:

[0008] Step S1: Construct a mother constellation consisting of N-dimensional dense constellation points The signature matrix F is constructed by using the orthogonal cascade power allocation method. K×J ;

[0009] K and J represent the number of orthogonal resources and the number of users in the non-terrestrial network cell, respectively. The J users are sorted in ascending order according to the distance between the user and the nadir point of the aerial base station, and the signature matrix F K×J The j-th column vector in represents the connection relationship between the j-th user after sorting and each orthogonal resource, 1≤j≤J; the orthogonal cascade power allocation method includes the following steps:

[0010] S11: Allocate d for each orthogonal resource f Power factor ρ i , and calculate the constellation operator corresponding to each power factor d f represents the number of users sharing the same orthogonal resource, θ i Indicates the power factor ρ i The corresponding phase rotation angle;

[0011] S12: Generate a set P consisting of all K-dimensional column vectors containing N 1s and KN 0s randomly arranged, and divide d f orthogonal groups, and the remaining column vectors are recorded as set R; each orthogonal group consists of It is composed of two orthogonal column vectors;

[0012] S13: Compare the column vectors in set R with d f Orthogonal groups are arranged alternately to form a new set P';

[0013] S14: Q i According to ρ i After sorting in ascending order, replace the i-th element 1 in each row of the set P' with the constellation operator q i , so that the column vectors in the set P' are sorted in ascending order according to the column vector power and the sorted set P' is used as the signature matrix F K×J ;

[0014] Step S2: Follow Construct a SCMA codebook set consisting of each user’s SCMA codebook M represents the number of codewords in each SCMA codebook, E represents the dimensional energy of each SCMA codebook, Denotes the signature matrix F K×J The non-zero rows in The Kronecker product of the corresponding rows in ;

[0015] Step S3: Calculate the upper bound of the bit error probability of each user based on the SCMA codebook of each user, and take the maximum value as the worst bit error rate;

[0016] Step S4: Taking the worst bit error rate as the optimization goal, the mother constellation The amplitude parameters δ, d f The power factor and d f The phase rotation angle is the optimization variable to be solved, and the optimization model is established. The optimization model is solved while ensuring that the dimensional energy of each SCMA codebook remains unchanged. The obtained optimization variable is later inserted into Get the SCMA codebook of each user.

[0017] Further, in step S3, the upper bound of the bit error probability of each user is calculated based on the SCMA codebook of each user, including:

[0018] For the jth user, the corresponding pairwise error probability is calculated based on its SCMA codebook

[0019] The upper bound of the bit error probability of the jth user is calculated as:

[0020]

[0021] Among them, P e,j represents the upper bound of the bit error probability of the jth user; X = [x1, x2, ..., x J ] represents the transmission codeword set of J users, x j represents the transmission codeword of the jth user, represents the set of transmission codewords obtained after the transmission signal received by the jth user is decoded incorrectly, Represents the set of transmitted codewords The transmission codeword of the jth user in; represents the number of error bits of the jth user.

[0022] Furthermore, the channel model between the user and the aerial base station is modeled as a Rice channel model;

[0023] And, the pairwise error probability of the jth user is The expression is:

[0024]

[0025] Where κ represents the Rice factor; Φ k is the set of users that collide on the kth orthogonal resource, x l [k] and Denote the user set Φ k The transmission codeword of the user in and the transmission codeword after decoding error; pl j is the path loss experienced by the jth user; N0 represents the noise variance.

[0026] Furthermore, the path loss pl of the jth user is j for:

[0027]

[0028] Where H represents the height of the aerial base station, represents the mean distance between the jth user and the nadir point, and α≥1 is the path loss coefficient.

[0029] Furthermore, the mean distance of the jth user for:

[0030]

[0031] Wherein, a=j, b=J-j+1, Γ(·) represents the gamma function, and R represents the radius of the non-terrestrial network cell.

[0032] Furthermore, the mother constellation Based on pulse amplitude modulation, it is constructed in combination with suboptimal interleaving and permutation rules.

[0033] Furthermore, when N is an even number, the mother constellation The N-dimensional dense constellation points in are:

[0034] A N =[-a M / 2 ,-a M / 2-1 …,-a1,a1,…,a M / 2 ]

[0035] When N is an even number, the mother constellation The N-dimensional dense constellation points in are:

[0036] A N =[-a1,a M / 2 ,-a2,a M / 2-1 ,…,-a M / 2 ,a1]

[0037] Among them, am =(m(δ-1)+(2-δ)), m=1,2,…,M / 2; δ>1 indicates the mother constellation Amplitude parameter.

[0038] According to another aspect of the present invention, a computer program product is provided, including a computer program; when the computer program is executed by a processor, the above-mentioned SCMA code base design method provided by the present invention is implemented.

[0039] According to another aspect of the present invention, there is provided a communication method for a non-terrestrial network, comprising:

[0040] The air base station maps the input information bits of each user into SCMA codewords based on the SCMA codebook of each user, superimposes them, and transmits them to each user terminal;

[0041] The SCMA codebook of each user is designed by the SCMA codebook design method provided by the present invention.

[0042] According to another aspect of the present invention, there is provided a non-terrestrial network system, comprising: an air base station and a user terminal;

[0043] The transmission codeword transmitted by the air base station to the user terminal is obtained by superimposing the SCMA codewords of each user, and the SCMA codeword of the user is obtained by mapping the user's information bit based on the SCMA codebook;

[0044] The SCMA codebook of each user is designed by the SCMA codebook design method provided by the present invention.

[0045] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects:

[0046] (1) The present invention allocates the same power factor to each orthogonal resource, realizes power balance among different orthogonal resources, and improves the bit error rate performance of the system; when generating the signature matrix for constructing the SCMA codebook, the users are sorted in ascending order of distance, that is, the users are sorted in ascending order of path loss, and the column vectors in the signature matrix are sorted in ascending order of column vector power, so that users with larger path loss will be allocated larger power, thus realizing user fairness; by introducing an appropriate constellation operator, the parameter quantity to be optimized is changed from J+d f +1 reduced to d f +d f +1, and d f The amplitude of the change with the scale of the non-terrestrial network is much smaller than the number of users J, which effectively reduces the computational complexity. In addition, the optimization calculation only involves system configuration parameter information and known user distance ranking information, does not rely on online information, and can be optimized offline.

[0047] (2) The present invention combines large-scale and small-scale fading to model the differences of different users in the NTN channel model. Specifically, based on the Rice distribution, the channel between the user and the orthogonal resource is established to reflect the small-scale fading between the user and the orthogonal resource, and the order statistics are combined to generalize the variation law of the distance between the user and the nadir point of the aerial base station to reflect the path loss of different users, and the channel model of the user is accurately established; based on the established NTN model, the multi-user codebook optimization is completed, and the system gain of the SCMA codebook is effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A schematic diagram of a scenario of an existing non-terrestrial network;

[0049] Figure 2 A schematic diagram of a SCMA codebook design method for non-terrestrial networks provided in an embodiment of the present invention;

[0050] Figure 3 The bit error rate performance simulation diagram of different methods under 150% overload rate provided by the embodiment of the present invention; wherein (a) is the simulation result corresponding to the Rice factor κ=10, and (b) is the simulation result corresponding to the Rice factor κ=2;

[0051] Figure 4 This is a simulation diagram of the bit error rate performance of different methods under a 200% overload rate provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0053] In the present invention, the terms "first", "second", etc. (if any) in the present invention and the accompanying drawings are used to distinguish similar objects but not necessarily to describe a specific order or sequence.

[0054] Before explaining the technical solution of the present invention in detail, basic concepts related to non-terrestrial networks and SCMA codebook design are briefly introduced as follows.

[0055] In order to solve the problem of short-term high-concurrency communication in the IoT system, sparse code multiple access (SCMA) has emerged as a new code domain non-orthogonal multiple access technology. SCMA greatly improves the resource reuse rate and the number of concurrent access users by directly mapping the instantaneous input information of each user into multi-dimensional codewords in the sparse codebook for overlapping transmission. In addition, the performance of the SCMA system mainly depends on the sparse codebook design scheme. Specific optimization schemes for applicable scenarios can effectively reduce multi-user interference and improve the system bit error rate performance.

[0056] The communication model of the non-terrestrial network downlink communication system based on SCMA is as follows: Figure 1 As shown in Figure 1, in this NTN cell, the aerial base station is located at the center of the cell and transmits information to J users through K orthogonal resource nodes, where the overload factor is defined as λ = J / K > 1. Assume that the cell is a circular area with a radius of R and users are randomly distributed in the cell. Each user is assigned a unique SCMA codebook, expressed as Contains M codewords of dimension K, then the process of mapping the transmission information bit of the jth user into the codeword is f j : ie,x j =f j (b j ),in is the input bit information of user j, and is the transmitted codeword. Each codeword contains only N non-zero elements, and N < K. The arrangement of the user codebook depends on the signature matrix F K×J The element f in k,j represents the connection relationship between the kth orthogonal resource and the jth user, k∈{1,2,…,K}, j∈{1,2,…,J}. If and only if f k,j =1, there is a connection relationship between the kth orthogonal resource and the jth user.

[0057] In the signature matrix basis F K×J On the top, construct a mother constellation constructed by N-dimensional dense constellation points Afterwards, you can follow Construct a SCMA codebook set consisting of each user’s SCMA codebook M represents the number of codewords in each SCMA codebook, and E represents the dimensional energy of each SCMA codebook; represents the energy normalization parameter; A special operation is defined, that is, the signature matrix F K×J The nth non-zero element of each column vector in is related to the mother constellation The nth row in performs the Crocker inner product operation.

[0058] In the communication system, each orthogonal resource is d f For this reason, there must be at least d f Different phases are used to distinguish these users, and finally J+d f +1 parameter, namely d f phase rotation angles, J power parameters, and the mother constellation However, as the scale of non-terrestrial networks increases, the number of parameters that need to be designed will also increase, resulting in an increase in the optimization complexity of codebook design and a possible degradation in the performance of the codebook derived using heuristic algorithms.

[0059] In non-terrestrial network scenarios, due to limited hardware costs and power consumption, it is necessary to reduce the number of optimization parameters. In addition, the power balance at different orthogonal resources needs to be considered, because if the power of a certain carrier is small, the codeword distance occupying the resource block will become smaller, which will lead to a significant reduction in decoding accuracy. In addition, the cells of non-terrestrial networks are much larger than those of terrestrial networks, so the differences in user locations in non-terrestrial network cells cannot be ignored. This position difference will lead to differences in path loss for different users; in addition, the movement of air base stations will cause frequent changes in user locations, which will cause the specific value of path loss to change all the time, and it is impossible to complete codebook optimization for fixed path loss. When designing SCMA codebooks for non-terrestrial networks, it is also necessary to optimize the allocated power to achieve user fairness.

[0060] The following are examples.

[0061] Embodiment 1:

[0062] A SCMA codebook design method for non-terrestrial networks, such as Figure 2 As shown, including:

[0063] Step S1: Construct a mother constellation consisting of N-dimensional dense constellation points The signature matrix F is constructed by using the orthogonal cascade power allocation method. K×J ;

[0064] Step S2: Follow Construct a SCMA codebook set consisting of each user’s SCMA codebook M represents the number of codewords in each SCMA codebook, E represents the dimensional energy of each SCMA codebook, represents the Kronecker product;

[0065] Step S3: Calculate the upper bound of the bit error probability of each user based on the SCMA codebook of each user, and take the maximum value as the worst bit error rate;

[0066] Step S4: Taking the worst bit error rate as the optimization goal, the mother constellation The amplitude parameters δ, d f The power factor and d f The phase rotation angle is the optimization variable to be solved, and the optimization model is established. The optimization model is solved while ensuring that the dimensional energy of each SCMA codebook remains unchanged. The obtained optimization variable is later inserted into Get the SCMA codebook of each user.

[0067] Each step is explained in detail below.

[0068] Considering that pulse amplitude modulation has the characteristic of symbol energy diversity, optionally, in step S1 of this embodiment, the mother constellation Based on pulse amplitude modulation, it is constructed in combination with suboptimal interleaving and permutation rules;

[0069] When N is an even number, the mother constellation The N-dimensional dense constellation points in are:

[0070] A N =[-a M / 2 ,-a M / 2-1 …,-a1,a1,…,a M / 2 ]

[0071] When N is an even number, the mother constellation The N-dimensional dense constellation points in are:

[0072] A N =[-a1,a M / 2 ,-a2,a M / 2-1 ,...,-a M / 2 ,a1]

[0073] Among them, a m =(m(δ-1)+(2-δ)), m=1,2,...,M / 2; δ>1 indicates the mother constellation to be optimized Amplitude parameter.

[0074] Based on the designed mother constellation The energy of each codebook dimension is:

[0075]

[0076] K and J represent the number of orthogonal resources and the number of users in the non-terrestrial network cell, respectively. In order to achieve fairness in user services, in this embodiment, the signature matrix F is constructed by using the orthogonal cascade power allocation method. K×J Before, the J users in the non-ground network cell are sorted in ascending order according to the distance between the user and the nadir point of the aerial base station, and the constructed signature matrix F K×JThe column vectors in are sorted in ascending order according to the column vector power. It is easy to understand that the greater the distance between the user and the nadir point of the aerial base station, the greater the path loss. Its signature matrix F K×J The j-th column vector in represents the connection relationship between the j-th user after sorting and each orthogonal resource. Therefore, this embodiment can allocate greater power to users with greater path loss, thereby achieving user fairness.

[0077] Due to the high-speed mobility of aerial base stations, the relative positions between base stations and users in non-terrestrial IoT are constantly changing, that is, the position of the jth user relative to the nadir point of the aerial base station is unknown. However, the distance relationship between users can be known from the power of the uplink received signal, so the distance ranking of users is known information for the downlink system.

[0078] In order to achieve power balance between different orthogonal resources and effectively reduce the computational complexity, in this implementation, the signature matrix F is constructed. K×J The orthogonal cascade power allocation method comprises the following steps:

[0079] S11: Allocate d for each orthogonal resource f Power factor ρ i , and calculate the constellation operator corresponding to each power factor d f represents the number of users sharing the same orthogonal resource, θ i Indicates the power factor ρ i The corresponding phase rotation angle;

[0080] The constellation operator indicates that the power and phase of the generated constellation points are changed to distinguish signals of different users. In this embodiment, the constellation operator q is introduced. i On the basis of , it is necessary to allocate orthogonal resources and constellation operators to each user to complete the construction of the user codebook. The allocation follows the following principles: higher power is allocated to users with larger propagation loss first to achieve user fairness; each resource block has the same power factor to achieve power balance between different resource blocks;

[0081] Specifically, some users in SCMA are divided into d f Orthogonal groups and all users in the group ensure that the occupied resources are orthogonal to each other, and the remaining Considering the power balance between different resource nodes, each group should be assigned the same q i ;

[0082] S12: Generate a set P consisting of K-dimensional column vectors containing N 1s and KN 0s. At this time, the total number of column vectors in the set P is J. Divide d forthogonal groups, and the remaining column vectors are recorded as set R; each orthogonal group consists of It is composed of two orthogonal column vectors;

[0083] S13: Compare the column vectors in set R with d f Orthogonal groups are arranged alternately to form a new set P';

[0084] S14: Q i According to ρ i After sorting in ascending order, replace the i-th element 1 in each row of the set P' with the constellation operator q i , so that the column vectors in the set P' are sorted in ascending order according to the column vector power and the sorted set P' is used as the signature matrix F K×J ;

[0085] For the jth column in the set P', the column vector power sum is calculated as:

[0086] In step S3, the calculation of the upper bound of the bit error probability of each user depends on the specific channel model.

[0087] The traditional SCMA codebook for ground IoT mainly considers Gaussian channels or Rayleigh channels, ignoring the problem of large-scale fading (path damage) and the coexistence of direct and indirect paths in small-scale fading. Unlike the ground IoT, there is a large direct path component between the user terminal of the non-ground network cell and the air base station. Therefore, the traditional design method cannot accurately characterize the channel model between the user terminal of the non-ground network cell and the air base station. To address this problem, this embodiment uses Rice distribution for channel modeling. Specifically, define is the channel between the air base station and the jth user, each component h j,k represents the channel between the kth orthogonal resource and the jth user, further expressed as where g j,k It is modeled as a Rician fading channel, following a Rician distribution with a Rician factor of κ, d j is the distance between the air base station and the jth user, α≥1 is the path loss coefficient, and the channel h established based on the Rice distribution j,k It reflects the small-scale fading of the channel. In the downlink, the base station sends the superimposed signal to user j as follows: Among them, p l represents the power allocated to the lth user, n j is the noise of the jth user, n j It follows a Gaussian distribution with a mean of 0 and a variance of N0.

[0088] By using the known information of user distance ranking, we can sort the user distances and measure the relative distances of J users with the expected value of the order statistic, and obtain the general law of large-scale fading changes in non-terrestrial networks.

[0089] In the polar coordinate system constructed with the nadir point as the origin, the position of user j can be obtained by the tensor angle and the distance r from the user to the center. j After sorting the users from the nearest to the farthest, the distances from the center to J users form an independent and identically distributed random sequence r1≤r2≤,…,≤r J , where the mean distance of the jth user is Γ(·) represents the gamma function, a=j, b=J-j+1, then the path loss of user j is d j is the distance between the aerial base station and the jth user, and α≥1 is the path loss coefficient.

[0090] Define the transmission codeword of J users as X = [x1, x2, ..., x J ], if the jth user's transmission signal X is incorrectly decoded into another set of codewords Then the pairwise error probability is:

[0091]

[0092] Where X=[x1,x2,...,x J ] represents the transmission codeword set of J users, x j represents the transmission codeword of the jth user, represents the set of transmission codewords obtained after the transmission signal received by the jth user is decoded incorrectly, Represents the set of transmitted codewords The transmission codeword of the jth user in; κ represents the Rice factor; Φ k is the set of users that collide on the kth orthogonal resource, x l [k] and Denote the user set Φ k The transmission codeword of the user in and the transmission codeword after decoding error; pl j is the path loss experienced by the jth user; N0 represents the noise variance.

[0093] It should be emphasized that the theoretical bit error rate of a single-user codebook is taken into account only when the information transmitted by the user is decoded incorrectly. That is, the upper bound of the bit error probability of the jth user is in represents the number of error bits of user j.

[0094] Based on the calculation, the average bit error rate and the worst bit error rate of J users are solved as follows:

[0095] P e,wor =max 1≤j≤J P e,j .

[0096] The channel states experienced by different users are different, resulting in different service qualities provided by the base station to each user in the IoT system. In addition, according to the short board effect, the average performance of the IoT system is most affected by the worst user codebook performance. Therefore, by taking the worst codebook performance among all users as the optimization target, the uneven distribution of power is used to balance the differential path losses experienced by different users, and the overall system performance is improved while ensuring user fairness. Accordingly, in step S4 of this embodiment, when establishing an optimization model to solve relevant parameters, the calculated worst bit error rate is used as the optimization target.

[0097] In step S4 of this embodiment, the optimization model can be expressed as:

[0098]

[0099] It is easy to understand that this problem has the following constraints:

[0100]

[0101] ρ i >0,1≤i≤d f

[0102] 0<θ i <π,1≤i≤d f

[0103] In this embodiment, the optimization variable of the final demand solution is d f The power factor and d f phase parameters and the amplitude parameter δ in the mother constellation A, a total of d f +d f +1 thanks to d f Less than J, and d f The growth rate of is much lower than the growth rate of the number of users in the non-terrestrial network cells. Therefore, this implementation effectively reduces the computational complexity of the SCMA codebook design.

[0104] In addition, in this embodiment, the optimization calculation only involves system configuration parameter information and known user distance ranking information, and does not rely on online information. Therefore, the SCMA codebook design method provided in this embodiment can be executed offline to generate the SCMA codebook and directly deployed at the IoT system origin of the non-ground network.

[0105] Optionally, in this embodiment, a genetic algorithm is used to solve the optimization problem, the population size is set to 50, and the maximum number of iterations is limited to 20 times to obtain a final user codebook.

[0106] Although the specific values ​​of the codebook finally generated in this embodiment may change, they have similar theoretical pairwise error probability values ​​and show excellent bit error rate performance in the simulation of system performance and user fairness.

[0107] In general, this embodiment can generate codebooks offline and directly deploy them at the origin of the Internet of Things system in a non-ground network. It can solve the problems of high complexity of online optimization of existing codebooks and unfair user services, and can effectively alleviate the channel difference problem caused by high-speed movement of air base stations. It is especially suitable for narrowband non-ground Internet of Things downlink communication systems.

[0108] Embodiment 2:

[0109] A computer program product includes a computer program; when the computer program is executed by a processor, the SCMA code base design method provided in the above embodiment 1 is implemented.

[0110] Embodiment 3:

[0111] A communication method for a non-terrestrial network, comprising:

[0112] The air base station maps the input information bits of each user into SCMA codewords based on the SCMA codebook of each user, superimposes them, and transmits them to each user terminal;

[0113] The SCMA codebook of each user is designed by the SCMA codebook design method provided in the above-mentioned embodiment 1.

[0114] Embodiment 4:

[0115] A non-ground network system, comprising: an air base station and a user terminal;

[0116] The transmission codeword transmitted by the air base station to the user terminal is obtained by superimposing the SCMA codewords of each user, and the SCMA codeword of the user is obtained by mapping the user's information bit based on the SCMA codebook;

[0117] The SCMA codebook of each user is designed by the SCMA codebook design method provided in the above-mentioned embodiment 1.

[0118] The following is a comparison of the downlink SCMA system performance using the existing golden angle codebook (GAM), Star-QAM codebook, Chen codebook, Huang codebook, and Li codebook to further verify the beneficial effects that the SCMA codebook design method provided in the embodiment of the present invention can achieve over the prior art.

[0119] Figure 3 The figure shows the error performance simulation diagram of SCMA codebooks designed by six different methods when the overload rate is 150%. Figure 3 In the figure, the horizontal axis represents the signal-to-noise ratio in dB, and the vertical axis represents the worst bit error rate of each user in the system. Figure 3 In the SCMA codebook design method provided by the present invention, the codebook designed is abbreviated as "Prop.". Figure 3 In the example, the number of transmitting antennas and receiving antennas are both 1, the ratio of the base station height to the cell radius is 1, the codebook modulation order is 4, the path loss factor α=3, the codebook overload rate is 150%, and the user positions in the cell are uniformly distributed. Figure 3 In (a), the Rice factor is κ = 10, Figure 3 In (b), the Rice factor is κ = 2. Figure 3 It can be seen from the results that the worst bit error rate of the present invention is always better than other methods, which shows that the codebook designed by the SCMA codebook design method for non-terrestrial networks provided by the present invention has better system performance.

[0120] Figure 4 The figure shows the error performance simulation diagram of the SCMA codebook designed by different methods when the overload rate is 200%. Because the Huang codebook and Chen codebook do not include the error performance under the 200% overload rate in the public data, Figure 4 In the figure, only the error performance simulation results of the remaining three design methods (i.e., golden angle codebook, Star-QAM codebook, Li codebook) and the codebook design method provided by the present invention are shown. Figure 4 In the figure, the horizontal axis represents the signal-to-noise ratio in dB, and the vertical axis represents the worst bit error rate or average bit error rate of each user in the system. Figure 4 In the example, the number of transmitting antennas and receiving antennas are both 1, the ratio of base station height to cell radius is 1, the codebook modulation order is 4, the path loss factor α = 3, the Rice factor κ = 10, the codebook overload rate is 200%, and the user positions in the cell are uniformly distributed. Figure 4 From the results shown, it can be seen that the codebook designed by the SCMA codebook design method for non-terrestrial networks provided by the present invention has better worst bit error rate and average bit error rate than other methods, and the difference between the worst bit error rate and the average bit error rate is the smallest, which shows that the present invention has better user fairness and system performance.

[0121] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A SCMA codebook design method for non-terrestrial networks, characterized in that: include: Step S1: Construct a mother constellation consisting of N-dimensional dense constellation points The signature matrix F is constructed by using the orthogonal cascade power allocation method. K×J ; K and J represent the number of orthogonal resources and the number of users in the non-terrestrial network cell, respectively. The J users are sorted in ascending order according to the distance between the user and the nadir point of the aerial base station, and the signature matrix F K×J The j-th column vector in represents the connection relationship between the j-th user and each orthogonal resource after sorting, 1≤j≤J; the orthogonal cascade power allocation method comprises the following steps: S11: Allocate d for each orthogonal resource f Power factor ρ i , and calculate the constellation operator corresponding to each power factor 1≤i≤d f , d f represents the number of users sharing the same orthogonal resource, θ i Indicates the power factor ρ i The corresponding phase rotation angle; S12: Generate a set P consisting of all K-dimensional column vectors containing N 1s and KN 0s randomly arranged, and divide d f orthogonal groups, and the remaining column vectors are recorded as set R; each orthogonal group consists of It is composed of two orthogonal column vectors; S13: Compare the column vectors in set R with d f Orthogonal groups are arranged alternately to form a new set P'; S14: Q i According to ρ i After sorting in ascending order, replace the i-th element 1 in each row of the set P' with the constellation operator q i , so that the column vectors in the set P' are sorted in ascending order according to the column vector power, and the sorted set P' is used as the signature matrix F K×J ; Step S2: Follow Construct a SCMA codebook set consisting of each user’s SCMA codebook M represents the number of codewords in each SCMA codebook, E represents the dimensional energy of each SCMA codebook, and “ο” represents the signature matrix F K×J The non-zero rows in The Kronecker product of the corresponding rows in ; Step S3: Calculate the upper bound of the bit error probability of each user based on the SCMA codebook of each user, and take the maximum value as the worst bit error rate; Step S4: Taking the worst bit error rate as the optimization goal, the mother constellation The amplitude parameters δ, d f The power factor and d f The phase rotation angle is the optimization variable to be solved, and the optimization model is established. The optimization model is solved while ensuring that the dimensional energy of each SCMA codebook remains unchanged. The obtained optimization variable is later inserted into Get the SCMA codebook of each user.

2. The SCMA codebook design method for non-terrestrial networks according to claim 1, characterized in that: In step S3, the upper bound of the bit error probability of each user is calculated based on the SCMA codebook of each user, including: For the jth user, the corresponding pairwise error probability is calculated based on its SCMA codebook The upper bound of the bit error probability of the jth user is calculated as: Among them, P e,j represents the upper bound of the bit error probability of the jth user; X = [x1, x2, ..., x J ] represents the transmission codeword set of J users, x j represents the transmission codeword of the jth user, represents the set of transmission codewords obtained after the transmission signal received by the jth user is decoded incorrectly, Represents the set of transmitted codewords The transmission codeword of the jth user in; represents the number of error bits of the jth user.

3. The SCMA codebook design method for non-terrestrial networks according to claim 2, characterized in that: The channel model between the user and the aerial base station is modeled as the Rice channel model; And, the pairwise error probability of the jth user is The expression is: Where κ represents the Rice factor; Φ k is the set of users that collide on the kth orthogonal resource, x l [k] and Denote the user set Φ k The transmission codeword of the user in and the transmission codeword after decoding error; pl j is the path loss experienced by the jth user; N0 represents the noise variance.

4. The SCMA codebook design method for non-terrestrial networks according to claim 3, characterized in that: The path loss pl of the jth user j for: Where H represents the height of the aerial base station, represents the mean distance between the jth user and the nadir point, and α≥1 is the path loss coefficient.

5. The SCMA codebook design method for non-terrestrial networks according to claim 4, characterized in that: The mean distance of the jth user for: Wherein, a=j, b=J-j+1, Γ(·) represents the gamma function, and R represents the radius of the non-terrestrial network cell.

6. The SCMA codebook design method for non-terrestrial networks according to any one of claims 1 to 5, characterized in that: The mother constellation Based on pulse amplitude modulation, it is constructed in combination with suboptimal interleaving and permutation rules.

7. The SCMA codebook design method for non-terrestrial networks according to claim 6, characterized in that: When N is an even number, the mother constellation The N-dimensional dense constellation points in are: A N =[-a M / 2 ,-a M / 2-1 …,-a1,a1,…,a M / 2 ] When N is an even number, the mother constellation The N-dimensional dense constellation points in are: A N =[-a1,a M / 2 ,-a2,a M / 2-1 ,...,-a M / 2 ,a1] Among them, a m =(m(δ-1)+(2-δ)), m=1,2,...,M / 2; δ>1 indicates the mother constellation Amplitude parameter.

8. A computer program product, characterized in that It includes a computer program; when the computer program is executed by a processor, it implements the SCMA code base design method described in any one of claims 1 to 7.

9. A communication method for non-terrestrial networks, characterized in that: include: The air base station maps the input information bits of each user into SCMA codewords based on the SCMA codebook of each user, superimposes them, and transmits them to each user terminal; The SCMA codebook of each user is designed by the SCMA codebook design method according to any one of claims 1 to 7.

10. A non-terrestrial network system, characterized in that: Including: aerial base station and user terminal; The transmission codeword transmitted by the air base station to the user terminal is obtained by superimposing the SCMA codewords of each user, and the SCMA codeword of the user is obtained by mapping the information bit of the user based on the SCMA codebook; The SCMA codebook of each user is designed by the SCMA codebook design method according to any one of claims 1 to 7.

Images