Large-scale MIMO satellite mobile communication channel estimation method and system

Abstract

The present invention discloses a method and system for estimating a large-scale MIMO satellite mobile communication channel. A satellite or a gateway uses a sequence of basic pilots modulated by different phase modulation factors as a pilot set, and uses statistical channel information including spatial angle information and average channel energy to implement pilot allocation and determine the pilot sequence used by each user. The user terminal uses the pilot sequence allocated by the satellite or the gateway to send a pilot signal. The satellite or the gateway implements two-stage channel estimation based on the received pilot signal, and the channel estimation value is obtained by first performing spatial domain processing on the satellite side received signal on a subcarrier-by-subcarrier basis and then performing frequency domain processing on a user-by-user basis. The spatial domain processing is the same for all subcarriers, and the frequency domain processing on a user-by-user basis can be quickly implemented using a Toeplitz system solver. The two-stage channel estimation method in the present invention can achieve near-optimal performance with greatly reduced implementation complexity.

Description

Technical Field

[0001] The present invention relates to a satellite mobile communication channel estimation method and system configured with an antenna array, and in particular to a satellite mobile communication channel estimation method and system using a large-scale MIMO technology. Background Art

[0002] Satellite communication is considered to be one of the key technologies for achieving seamless global network coverage. Most existing satellite communications use multi-beam transmission schemes, in which the beamformer on the satellite side is usually fixed, which limits the transmission capacity of satellite communications to a certain extent. Massive multiple-input multiple-output (MIMO) is an important technology for terrestrial 5G. By using a large-scale antenna array on the base station side to generate a large number of dynamic beams, it can support dozens of users to communicate with the base station on the same time-frequency resources. Extending the application of massive MIMO technology to satellite mobile communication systems can significantly improve the spectrum efficiency and power efficiency of satellite mobile communication systems.

[0003] Orthogonal frequency division duplexing (OFDM) has become one of the key technologies for achieving high-speed communication in broadband transmission. It can not only combat frequency selective fading, but also has an efficient implementation method. Combining massive MIMO and OFDM to form a massive MIMO OFDM system is the core enabling technology of 5G mobile communication system, and will continue to evolve and develop in 5G evolution and 6G system. Most previous works have focused on massive MIMO OFDM technology in terrestrial wireless networks, among which massive MIMO OFDM channel estimation technology has also been widely studied and applied, but the particularity of satellite channels has not been considered. Therefore, it is of great significance to conduct research on channel estimation problems in satellite massive MIMO OFDM systems. As far as we know, there is no literature on this subject. Summary of the invention

[0004] Purpose of the invention: The purpose of the present invention is to overcome the deficiencies of the prior art, improve the channel estimation performance of satellite mobile communication systems, and reduce the implementation complexity.

[0005] Technical solution: To achieve the above invention purpose, the present invention adopts the following technical solution:

[0006] A massive MIMO OFDM satellite mobile communication channel estimation method is applied to a satellite mobile communication system and is characterized in that: a satellite is equipped with an antenna array to communicate with a user terminal equipped with multiple antennas or a single antenna in its coverage area; a sequence of basic pilots modulated by different phase modulation factors is used as a pilot set; the satellite or a gateway station implements pilot allocation using statistical channel information including spatial angle information and average channel energy, determines the pilot sequence of each user, and sends its sequence number to each user; the user sends a pilot signal, and the satellite or the gateway station implements channel estimation based on the received pilot signal, and adopts a space-frequency two-level channel estimation method to obtain an estimated value of the channel parameter.

[0007] In a preferred embodiment, the statistical channel information is obtained by an uplink detection process or through feedback information from each user terminal; during the uplink detection process, each user periodically sends a detection signal, and the satellite estimates the spatial angle information and average channel energy of each user based on the received detection signal; the feedback information from each user terminal is the user's geographic location information, spatial angle information, and average channel energy.

[0008] In a preferred embodiment, the satellite or the gateway uses a graph-based pilot allocation algorithm to perform pilot allocation. The graph-based pilot allocation algorithm divides the user set into S groups, where S represents the number of available pilots, so that users in the same group use the same pilot and users in different groups use different pilots. In the graph-based pilot allocation algorithm, the vertex set of the graph represents the user set, and the weights between the vertices of the graph represent the interference between users. The weights between the vertices are calculated from the spatial angle information and the average channel energy. The graph-based pilot allocation algorithm includes the following steps:

[0009] Initialize a user set to which pilots have been allocated and a user set to which pilots have not been allocated;

[0010] Allocate a pilot that minimizes the sum of weights between any two vertices in the group to each user to which a pilot is not allocated, and update the user set to which the pilot is allocated and the user set to which the pilot is not allocated;

[0011] In a preferred embodiment, in the space-frequency two-stage channel estimation method, the channel estimation value is obtained by first performing subcarrier-by-subcarrier space-domain processing on the satellite-side received signal and then performing user-by-user frequency-domain processing. The space-domain processing is the same for all subcarriers. The user-by-user frequency-domain processing can be quickly implemented using a Toeplitz system solver.

[0012] A method for estimating a massive MIMO OFDM satellite mobile communication channel is applied to a user terminal and is characterized by:

[0013] The user terminal periodically sends a detection signal to the satellite, or feeds back the user's geographical location information, spatial angle information, and average channel energy to the satellite for the satellite or the gateway to implement pilot allocation and channel estimation;

[0014] The user terminal uses the Doppler frequency shift caused by satellite movement and the minimum propagation delay of long-distance propagation to perform frequency and time compensation on the uplink transmission signal;

[0015] Each user terminal sends a pilot signal using a pilot sequence assigned by a satellite or a gateway.

[0016] In a preferred embodiment, the Doppler frequency shift and the minimum propagation delay of long-distance transmission caused by the satellite movement are estimated by the user terminal based on the received synchronization signal, or calculated from the position information of the user terminal and the satellite; as the satellite or user terminal moves, the Doppler frequency shift and minimum propagation delay information are dynamically updated, and the frequency and time compensation amounts change adaptively accordingly.

[0017] A satellite-side device for estimating a massive MIMO OFDM satellite mobile communication channel comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the computer program implements the massive MIMO OFDM satellite mobile communication channel estimation method when loaded into the processor.

[0018] A massive MIMO OFDM satellite mobile communication channel estimation user terminal device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the computer program implements the massive MIMO OFDM satellite mobile communication channel estimation method when loaded into the processor.

[0019] A massive MIMO OFDM satellite mobile communication channel estimation system, comprising a satellite and a user terminal, characterized in that: the satellite is equipped with an antenna array to communicate with a user terminal equipped with multiple antennas or a single antenna in its coverage area; the satellite or a gateway station associated with the satellite is used to:

[0020] The sequence of basic pilots modulated by different phase modulation factors is used as a pilot set; the satellite or the gateway station implements pilot allocation using statistical channel information including spatial angle information and average channel energy, and feeds back the pilot allocation results to each user; the user sends a pilot signal, and the satellite or the gateway station performs channel estimation based on the received pilot signal;

[0021] The satellite or gateway uses a graph-based pilot allocation algorithm to allocate pilots. The graph-based pilot allocation algorithm divides the user set into S groups, where S represents the number of available pilots, so that users in the same group use the same pilot and users in different groups use different pilots. In the graph-based pilot allocation algorithm, the vertex set of the graph represents the user set, and the weights between the vertices of the graph represent the interference between users. The weights between the vertices are calculated by spatial angle information and average channel energy;

[0022] The satellite or the gateway implements channel estimation using a space-frequency two-stage channel estimation method based on the received pilot signal. In the space-frequency two-stage channel estimation method, the channel estimation value is obtained by first performing subcarrier-by-subcarrier spatial domain processing on the satellite-side received signal and then performing user-by-user frequency domain processing. The spatial domain processing is the same for all subcarriers. The user-by-user frequency domain processing can be quickly implemented using a Toeplitz system solver;

[0023] During the movement of the satellite or each user terminal, the channel estimation result is dynamically updated as the statistical channel information changes;

[0024] The user terminal is used to: periodically send a detection signal to the satellite, or feed back the user's geographical location information, spatial angle information, and average channel energy to the satellite, so that the satellite or the gateway station can implement pilot allocation and channel estimation; use the Doppler frequency shift caused by satellite movement and the minimum propagation delay of long-distance propagation to perform frequency and time compensation on the uplink transmission signal; and use the pilot sequence allocated by the satellite or the gateway station to send the pilot signal.

[0025] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0026] (1) The application of large-scale MIMO technology to satellite mobile communication systems has improved the spectrum efficiency and power efficiency of satellite mobile communication systems.

[0027] (2) By making full use of the characteristics of satellite mobile communication channels, a complete channel estimation method is designed, which greatly reduces the computational complexity.

[0028] (3) The channel estimation method is applicable to time division duplex and frequency division duplex satellite mobile communication systems.

[0029] (4) Each user terminal performs frequency and time compensation on its transmitted signal, which simplifies the system design. The communication method is applicable to satellite mobile communications using high-orbit satellites, medium-orbit satellites, and low-orbit satellites. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description only show some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0031] Figure 1 Schematic diagram of the channel estimation method for massive MIMO satellite mobile communications.

[0032] Figure 2 Schematic diagram of a massive MIMO satellite mobile communication system.

[0033] Figure 3 This is a comparison chart of channel estimation performance for massive MIMO satellite mobile communications.

[0034] Figure 4 This is a schematic diagram of the structure of the satellite-side equipment for massive MIMO satellite mobile communication channel estimation.

[0035] Figure 5 Schematic diagram of the structure of the user terminal equipment for massive MIMO satellite mobile communication channel estimation.

[0036] Figure 6 Schematic diagram of the structure of the massive MIMO satellite mobile communication channel estimation system. DETAILED DESCRIPTION

[0037] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0038] like Figure 1 As shown, the massive MIMO OFDM satellite mobile communication channel estimation system disclosed in the example of the present invention includes a satellite and a user terminal, characterized in that: the satellite is equipped with an antenna array to communicate with a user terminal equipped with multiple antennas or a single antenna in its coverage area; the satellite or a gateway station associated with it is used to:

[0039] The sequence of basic pilots modulated by different phase modulation factors is used as a pilot set; the satellite or the gateway station implements pilot allocation using statistical channel information including spatial angle information and average channel energy, and feeds back the pilot allocation results to each user; the user sends a pilot signal, and the satellite or the gateway station performs channel estimation based on the received pilot signal;

[0040] The satellite or gateway uses a graph-based pilot allocation algorithm to allocate pilots. The graph-based pilot allocation algorithm divides the user set into S groups, where S represents the number of available pilots, so that users in the same group use the same pilot and users in different groups use different pilots. In the graph-based pilot allocation algorithm, the vertex set of the graph represents the user set, and the weights between the vertices of the graph represent the interference between users. The weights between the vertices are calculated by spatial angle information and average channel energy;

[0041] The satellite or the gateway implements channel estimation using a space-frequency two-stage channel estimation method based on the received pilot signal. In the space-frequency two-stage channel estimation method, the channel estimation value is obtained by first performing subcarrier-by-subcarrier spatial domain processing on the satellite-side received signal and then performing user-by-user frequency domain processing. The spatial domain processing is the same for all subcarriers. The user-by-user frequency domain processing can be quickly implemented using a Toeplitz system solver;

[0042] During the movement of the satellite or each user terminal, the channel estimation result is dynamically updated as the statistical channel information changes;

[0043] The example of the present invention discloses a method for estimating a massive MIMO OFDM satellite mobile communication channel, which is applied to a user terminal and is characterized by:

[0044] The user terminal periodically sends a detection signal to the satellite, or feeds back the user's geographical location information, spatial angle information, and average channel energy to the satellite for the satellite or the gateway to implement pilot allocation and channel estimation;

[0045] The user terminal uses the Doppler frequency shift caused by satellite movement and the minimum propagation delay of long-distance propagation to perform frequency and time compensation on the uplink transmission signal;

[0046] Each user terminal sends a pilot signal using a pilot sequence assigned by a satellite or a gateway;

[0047] The Doppler frequency shift and the minimum propagation delay of long-distance transmission caused by satellite movement are estimated by the user terminal based on the received synchronization signal, or calculated from the position information of the user terminal and the satellite; as the satellite or user terminal moves, the Doppler frequency shift and minimum propagation delay information are dynamically updated, and the frequency and time compensation amounts change adaptively accordingly.

[0048] The method of the embodiment of the present invention is further introduced below in conjunction with a specific implementation scenario. The method of the present invention is not limited to a specific scenario. For other implementations outside the exemplary scenario of the present invention, those skilled in the art can make adaptive adjustments according to the specific scenario based on the technical ideas of the present invention and existing knowledge.

[0049] (1) System Configuration

[0050] Consider the case of a single satellite (which can be a low-orbit satellite, a medium-orbit satellite, or a high-orbit satellite), and the satellite side is equipped with an antenna array (which can be a one-dimensional or two-dimensional array, with dozens to hundreds of antennas). Antenna arrays or large-scale antenna arrays can be arranged in different shapes according to requirements such as quantity and ease of installation. The most basic is the two-dimensional uniform panel antenna array (Uniform Planar Array, UPA), that is, the antenna units are evenly arranged in the horizontal and vertical directions, and the spacing between adjacent antenna units can be λ / 2 or Where α is the carrier wavelength.

[0051] Take a single satellite communicating with multiple users as an example. Figure 2 Assume that the satellite is equipped with UPA, and the number of antenna units in the x-axis and y-axis directions are M and x and M y , then M=M x M y The total number of antennas equipped for the satellite. Assume that each user side is equipped with an omnidirectional antenna and the number of users is K. The user index set is recorded as

[0052] Consider using orthogonal frequency division multiplexing (OFDM) for broadband data transmission, so that the frequency selective fading channel can be converted into multiple parallel flat fading channels. The total number of subcarriers is N c , the interval between adjacent subcarriers is Δf. The system sampling time is T s =1 / (N c Δf). Add a length of N at the beginning of each OFDM symbol. g The protection interval is called the cyclic prefix (CP). The duration of the CP is T g =N g T s The time lengths of OFDM symbols without and with CP are T c =N c T s and T = T g +T c .

[0053] Consider a massive MIMO satellite communication system operating in frequency division duplexing (FDD) mode, where uplink and downlink occupy different frequency bands. The time resources in the uplink phase are divided into multiple time slots, each of which contains N S OFDM symbols. The first OFDM symbol in each time slot is used for uplink channel estimation, and the remaining N S - 1 OFDM symbol is used for uplink data transmission.

[0054] In training-based uplink channel estimation, multiple ground mobile users simultaneously send pilot sequences to a single satellite. The pilot sequences are known accurately to both the satellite and the user side. After receiving the pilot sequences, the satellite estimates the channel parameters for all users.

[0055] remember Represents the set of all n×m dimensional complex (real) matrices.

[0056] (2) Signal model

[0057] The received signal received by the satellite at time t for

[0058]

[0059] in and are the channel impulse response and the transmitted signal of user k, respectively. is the additive Gaussian noise signal on the satellite side. Channel impulse response It can be expressed as

[0060]

[0061] Where δ(x) is the Dirc delta function, Q k is the multipath number of user k, v k，q , τ k，q and g k，q are the channel gain, Doppler frequency shift, propagation delay and satellite-side array response vector of the qth path of user k respectively.

[0062] The Doppler frequency shift ν of the qth path of user k k，q It can be expressed as in and are the Doppler shifts caused by the movement of satellite and user k respectively. In addition, The different multipaths for user k are approximately the same, that is, The minimum propagation delay of user k channel is recorded as remember are the angle-of-arrival (AoA) of the qth path of user k channel. The satellite side array response vector g k，q It can be expressed as g k，q =g(θ k，q ). The vector g(θ) for any θ=(θ x ,θ y ) is defined as

[0063]

[0064] in represents the Kronecker product of two vectors. In formula (3), It can be expressed as

[0065]

[0066] in λ=c / f c is the uplink carrier wavelength, c = 3 × 10 8 m / s is the speed of light, f c is the uplink carrier frequency, d v is the distance between adjacent antenna elements along the v∈{x,y} axis, (·) T Indicates the transpose of a vector or matrix. If the satellite is equipped with other antenna arrays, just replace Just replace it with its corresponding array response vector.

[0067] In satellite communications, since users are far away from satellites, the departure angles corresponding to different multipath signals of the same user can be considered to be approximately the same, that is, θ k，q =θ k , q=1,...,Q k Therefore, the array response vector g corresponding to the qth path of user k channel is k，q It can be abbreviated as g k，q =g k =g(θ k ),in is called the AoA of user k’s channel. is the spatial angle of user k’s channel, where and The nadir angle of user k is denoted as Its maximum value is Nadir angle and spatial angle ξ k Satisfies the following relationship

[0068]

[0069] Therefore, the spatial angle ξ of user kk Falling in the circular area Because the satellite is far away from the inside, the spatial angle The changes are slow, so it can be assumed that the satellite and the user know the exact spatial angle information.

[0070] remember The frequency domain signal transmitted by user k in the sth OFDM symbol is

[0071]

[0072] remember and By using the Doppler frequency shift and delay compensation method, the time domain transmitted signal of user k after compensation is:

[0073]

[0074] The received signal of the satellite in the sth OFDM symbol is

[0075]

[0076] where z s (t) is the additive Gaussian noise signal. Here, is the equivalent channel impulse response

[0077]

[0078] in and are the equivalent channel gain and propagation delay of the qth path of user k’s channel respectively. In the OFDM system, in order to avoid inter-symbol interference, N g Should satisfy

[0079] In addition, remember h k (t, f) is the equivalent channel frequency response of user k, which can be expressed as

[0080]

[0081] in The received signal of the satellite on the rth subcarrier of the sth OFDM symbol is

[0082]

[0083] in is the channel vector of user k, z s，r is the additive noise on the satellite side. The channel vector h in (11) k，s，r It can be expressed as

[0084] h k，s，r =h k (sT, rΔf) = d k，s，r g k , (12)

[0085] where d k，s，r =d k (sT, rΔf). Since the Doppler shift and propagation delay have been compensated, the frequency and time of the satellite and user sides can be considered to be perfectly synchronized.

[0086] (3) Channel Model

[0087] This section analyzes the channel characteristics of the OFDM symbol where the pilot is located. For convenience, the subscript s of the OFDM symbol is omitted hereafter. The channel vector of user k on the rth subcarrier of the OFDM symbol is

[0088] h k，r =d k，r g k , (13)

[0089] in a k，q is the equivalent channel gain of the qth path of user k’s channel. Assume that N p subcarriers are used for channel estimation, and the corresponding index set is

[0090] remember For user k in N p The frequency domain channel of subcarriers can be expressed as

[0091]

[0092] in H p，k is called the space-frequency domain channel (SFC) of user k. In addition, d p，k It can be further expressed as

[0093]

[0094] in for

[0095]

[0096] Note that the equivalent propagation delay of each user satisfies The interval [0, T g ) is divided into the following intervals

[0097]

[0098] in τ l =lT d / (N d N p Δf) is the lth grid point, where and Note μ d =N d / L d is the delay refinement factor. Using N in (17) d subintervals, d p，k can be re-expressed as

[0099]

[0100] in When N d When large enough, of can be approximated as p(τ l ), then d p，k can be re-expressed as

[0101]

[0102] in Note that N d express Approximately p(τ l ) accuracy. d The larger it is, the more accurate the approximation is. p，k It can be expressed as

[0103]

[0104] Afterwards, t，k It is called the angle-delay domain channel (ADC) of user k.

[0105] Assume α k，e It can be modeled as where β k represents the large-scale fading parameter, η k，l represents the small-scale fading parameter of the l-th path of user k. In addition, assuming that η k，l Obey Gaussian distribution in is called the power delay profile (PDP) of user k. And, α k，lFor different users k and multipath l, they are independent of each other, that is,

[0106]

[0107] Time domain correlation matrix for

[0108]

[0109] in And ω k，l =β k γ k，l Assume that the satellite and the user can accurately obtain the slowly changing channel parameters

[0110] (3) Obtaining statistical channel information

[0111] Statistical channel information such as spatial angle information and average channel energy can be obtained through the uplink detection process or through feedback information from each user terminal. In the uplink detection process, each user periodically sends a detection signal. The satellite estimates the spatial angle information, average channel energy, mean vector and variance matrix of the user-side channel component of each user based on the received detection signal. Specifically, the spatial angle The estimated value of can be obtained through the classic arrival angle estimation algorithm, such as MUSIC algorithm, ESPRIT algorithm, Unitary ESPRIT algorithm, etc.; the average channel energy w k，l The estimated value of can be obtained by the following method

[0112]

[0113] in is the vector α k，l The nth estimated value, N L For estimated times.

[0114] The feedback information of each user terminal is the user's geographical location information, spatial angle information and average channel energy. The feedback information of each user terminal can be obtained by using the uplink synchronization signal or the detection signal through the channel parameter estimation method, wherein the geographical location information can also be obtained with the help of the global positioning system. In the case where the terminal feeds back the geographical location information, the satellite side uses the terminal's geographical location information and the satellite's position information to obtain the spatial angle information of each user.

[0115] (4) Pilot multiplexing and pilot allocation

[0116] The number of available pilots is S, and the pilot set is In satellite communications, the number of available pilots is much smaller than the number of users, that is, S < < K. Therefore, pilots need to be reused between different users. p，s The user set is recorded as in

[0117] remember For satellites in N p The uplink received signal of subcarriers can be expressed as

[0118]

[0119] where p P is the transmission power of user k, s k represents the pilot index used by user k. In addition, is additive complex Gaussian noise, whose elements are independent and identically distributed

[0120] Using the characteristics of the phase-shifted pilot, the sth pilot X p，s It can be expressed as

[0121]

[0122] in satisfy In addition, X c =diag{x c} is a common pilot with unit modulus elements and satisfies For example, X c It can be constructed based on the Zadoff-Chu sequence. In addition, the maximum number of pilots satisfies

[0123] Since the satellite side knows the spatial angle accurately That is, the known array direction vector Estimation of SFC The estimate can be transformed into a vector In addition, since ADC can better describe SFC with fewer parameters, we first consider estimating ADC

[0124] remember and From formula (24), we can get:

[0125]

[0126] The following equation is used

[0127]

[0128] in for

[0129]

[0130] z p Follow the distribution When we get the vector d t After the estimation, the vector The estimate is

[0131]

[0132] d t The minimum mean square error (MMSE) of

[0133]

[0134] in

[0135] The corresponding channel estimation error Follow the distribution in for

[0136]

[0137] The MMSE of channel estimation can be expressed as

[0138]

[0139] When N p As it approaches infinity, the MMSE of channel estimation approaches J→J asy , where J asy for

[0140]

[0141] in When the following conditions are met

[0142]

[0143] J asy The following minimum values ​​can be achieved

[0144]

[0145] It can be found that when N p As it approaches infinity, the interference between users using different pilots will disappear, and only the interference caused by the multiplexed pilot will exist. In addition, if the array direction vectors of users using the same pilot are orthogonal to each other, the interference caused by the multiplexed pilot will also disappear, and the asymptotic MMSE reaches its minimum value.

[0146] It can be seen that the MMSE performance of channel estimation depends on the pilot allocation. In order to further improve the performance of channel estimation, it is necessary to allocate pilots reasonably. From the above analysis, it can be seen that users whose array direction vectors are close to orthogonal to each other should use the same pilot, while users whose array direction vectors are highly linearly correlated should use different pilots. In other words, the inner product of the array direction vectors of any two users using the same pilot should be as close to zero as possible. Intuitively, the pilot allocation problem can be formulated as the following optimization problem:

[0147]

[0148] in Note that the problem in (36) can be expressed as a max-S-cut problem on a graph. in represents a set of vertices, is a weight set. Specifically, any edge between a vertex i and a vertex k has a weight W i，k , and satisfy W i，k =W k，i , and record W i，i = 0. It can be seen that problem (36) is actually to divide the vertex set into S disjoint subsets Minimize the sum of the weights of the edges connecting vertices in the same subset.

[0149] Note that the following equation

[0150]

[0151] Problem (36) can be equivalently transformed into the following problem

[0152]

[0153] It is worth noting that question (38) is exactly Specifically, problem (38) attempts to find the set of vertices An S partition is constructed to maximize the sum of weights of endpoints in different subsets. Inspired by this, an efficient graph-based pilot allocation algorithm is designed.

[0154] The specific steps of the graph-based pilot allocation algorithm are as follows:

[0155] Input: Space angle and the average channel energy

[0156] Output: pilot allocation result

[0157] Step 1: Initialization and

[0158] Step 2: For any Calculate s as follows k

[0159]

[0160] Step 3: Update and

[0161] Step 4: If Is an empty set, exit the execution step; if If it is not an empty set, proceed to step 2.

[0162] (5) Space-frequency two-level channel estimation method

[0163] A space-frequency two-stage channel estimation method is proposed below. The channel estimation is obtained by first performing subcarrier-by-subcarrier spatial domain processing on the satellite-side received signal and then performing user-by-user frequency domain processing. The spatial domain processing is the same for all subcarriers. By minimizing the asymptotic MMSE, the optimal linear subcarrier-by-subcarrier spatial domain processing vector is obtained. Using the fast Toeplitz system solver, user-by-user frequency domain processing can be quickly implemented.

[0164] To simplify the analysis, assume that each user’s R t，k Can be expressed as

[0165] R t，k =β k Γ, (39)

[0166] where Γ = diag{γ}, satisfy This means that the PDP is the same for all users. Therefore, the satellite only needs to obtain the large-scale fading parameters and PDP for all users This can significantly reduce the overhead of acquiring statistical channel information.

[0167] In the space-frequency two-level channel estimation method, the channel estimate can be obtained as follows

[0168]

[0169] in and They represent the subcarrier-by-subcarrier spatial processing vector and the user-by-user frequency processing matrix respectively. First, the satellite-side received signal undergoes the following spatial processing:

[0170]

[0171] in Follow the distribution Then, t，k The estimation of can be obtained by frequency domain processing per user as follows

[0172]

[0173] By minimizing The optimal frequency domain processing matrix for user k is:

[0174]

[0175] in for

[0176]

[0177] Diagonal matrix Λ w，k for

[0178]

[0179] The MMSE matrix is

[0180]

[0181] in The corresponding MMSE is

[0182]

[0183] When N p As J approaches infinity, w，k tend to in for

[0184]

[0185] Among them A k，l =G k γ e +σ 2 W k , W k =||w k || 2 and Make The minimized spatial processing vector should satisfy

[0186]

[0187] where u k satisfy

[0188]

[0189] It can be seen that the asymptotically optimal spatial domain processing vector should have the form of regularized zero-breaking processing. k，l With w k It is usually difficult to obtain v through formula (50) k According to formula (50), v k Equal to the equation F(v k )=0, where F(u k )for

[0190]

[0191] Equation F(u k The roots of )=0 can be obtained by Newton's method.

[0192] In the user-by-user frequency domain processing in equation (43), there exists Note that T w，k is a Toeplitz matrix, which is equivalent to solving the Toeplitz system T w，k x w，k =y w，k Here, the classic Levinson recursive algorithm is used to solve the Toeplitz system. For an n-dimensional Toeplitz matrix, the number of multiplications required by the Levinson algorithm is 4n 2 . To solve the Toeplitz system T n Take s=b as an example, where is a positive Toeplitz matrix, b=[b1…b n ] T is an arbitrary vector. The steps of Levinson algorithm are as follows.

[0193] First, T n Re-expressed as T n =T0L n , where T0 = [T n ] 0，0 , L n It can be expressed as

[0194]

[0195] where r n-1 =[ρ1...ρ n-1 ] T For convenience, remember r m=[ρ1...ρ m ] T Variable α m , m , The initial values ​​of are α1=-ρ1, ζ1=1, x1=b1, y1=-ρ1. For 1≤m≤n-1, ζ m and x m Update as follows

[0196] ζ m+1 =(1-|α m | 2 ) m , (53)

[0197]

[0198]

[0199] in Represents the vector obtained by rearranging the elements in vector x in reverse order. For 1≤m≤n-2, α m and m Update as follows

[0200]

[0201]

[0202] In getting x n After that, Toeplitz system T n The solution of s=b can be expressed as

[0203] When the solution After that, in (42) It can be expressed as

[0204]

[0205] in IFFT(·) stands for Inverse Fast Fourier Transform.

[0206] The specific steps of the space-frequency two-level channel estimation algorithm are as follows.

[0207] Input: Space angle Average channel energy and common PDP

[0208] Output: Channel estimation result

[0209] for do

[0210] Calculate w according to formula (49): k ;

[0211] calculate

[0212] Calculate according to equations (53)-(57)

[0213] According to formula (58),

[0214] end for

[0215] (6) Dynamic Update

[0216] As the satellite or user moves, the channel information such as the spatial angle and average channel energy of each user is dynamically updated, and the pilot allocation and channel estimation values ​​change adaptively accordingly. Information such as Doppler frequency shift, minimum propagation delay, frequency and time compensation of the user terminal also change adaptively accordingly.

[0217] Figure 3 The channel estimation performance curves of the method proposed in this embodiment in different scenarios are given.

[0218] Based on the same inventive concept, Figure 4 As shown, an embodiment of the present invention discloses a satellite-side device for estimating a large-scale MIMO satellite mobile communication channel, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is loaded into the processor, the method for estimating a large-scale MIMO satellite mobile communication channel applied to a satellite or a gateway is implemented.

[0219] In a specific implementation, the device includes a processor, a communication bus, a memory, and a communication interface. The processor may be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the present invention. The communication bus may include a path to transmit information between the above components. The communication interface uses any transceiver-like device for communicating with other devices or communication networks. The memory may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPRO M), a read-only compact disk (CD-ROM) or other optical disk storage, a disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may be independent and connected to the processor via a bus. The memory may also be integrated with the processor.

[0220] The memory is used to store the application code for executing the scheme of the present invention, and the execution is controlled by the processor. The processor is used to execute the application code stored in the memory, thereby realizing the communication method provided by the above embodiment. The processor may include one or more CPUs, or may include multiple processors, each of which may be a single-core processor or a multi-core processor. The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0221] Based on the same inventive concept, Figure 5 As shown, a large-scale MIMO satellite mobile communication channel estimation user terminal device disclosed in an embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is loaded into the processor, the large-scale MIMO satellite mobile communication channel estimation method applied to the user terminal is implemented. In a specific implementation, the user terminal device includes a processor, a communication bus, a memory, and a communication interface, and its form may include various handheld devices with wireless communication functions, vehicle-mounted devices, wearable devices, computing devices, or other processing devices connected to a wireless modem.

[0222] like Figure 6 As shown, the massive MIMO OFDM satellite mobile communication channel estimation system disclosed in the example of the present invention includes a satellite and a user terminal, wherein the satellite is equipped with an antenna array to communicate with a user terminal equipped with multiple antennas or a single antenna in its coverage area; the satellite or a gateway station associated with the satellite is used to:

[0223] The sequence of basic pilots modulated by different phase modulation factors is used as a pilot set; the satellite or the gateway station implements pilot allocation using statistical channel information including spatial angle information and average channel energy, determines the pilot sequence of each user, and sends its sequence number to each user; the user sends a pilot signal, and the satellite or the gateway station performs channel estimation based on the received pilot signal, and obtains the estimated value of the channel parameter using a space-frequency two-level channel estimation method;

[0224] The satellite or gateway uses a graph-based pilot allocation algorithm to allocate pilots. The graph-based pilot allocation algorithm divides the user set into S groups, where S represents the number of available pilots, so that users in the same group use the same pilot and users in different groups use different pilots. In the graph-based pilot allocation algorithm, the vertex set of the graph represents the user set, and the weights between the vertices of the graph represent the interference between users. The weights between the vertices are calculated by spatial angle information and average channel energy;

[0225] The satellite or the gateway implements channel estimation using a space-frequency two-stage channel estimation method based on the received pilot signal. In the space-frequency two-stage channel estimation method, the channel estimation value is obtained by first performing subcarrier-by-subcarrier spatial domain processing on the satellite-side received signal and then performing user-by-user frequency domain processing. The spatial domain processing is the same for all subcarriers. The user-by-user frequency domain processing can be quickly implemented using a Toeplitz system solver;

[0226] During the movement of the satellite or each user terminal, the channel estimation result is dynamically updated as the statistical channel information changes;

[0227] The user terminal is used to: the user terminal periodically sends a detection signal to the satellite, or feeds back the user's geographical location information, spatial angle information, and average channel energy to the satellite, so that the satellite or the gateway station can implement pilot allocation and channel estimation;

[0228] The user terminal uses the Doppler frequency shift caused by satellite movement and the minimum propagation delay of long-distance propagation to perform frequency and time compensation on the uplink transmission signal;

[0229] Each user terminal sends a pilot signal using a pilot sequence assigned by a satellite or a gateway.

[0230] The above-mentioned massive MIMO satellite mobile communication channel estimation system embodiment and the massive MIMO satellite mobile communication channel estimation method embodiment belong to the same inventive concept, and the specific technical means implementation details can refer to the method embodiment, which will not be repeated here. The contents not involved in the present invention are all prior art.

[0231] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for estimating a massive MIMO OFDM satellite mobile communication channel, the method being applied to a satellite mobile communication system, characterized in that: The satellite is equipped with an antenna array to communicate with user terminals equipped with multiple antennas or a single antenna within its coverage area; The sequence of basic pilots modulated by different phase modulation factors is used as a pilot set; the satellite or the gateway station implements pilot allocation using statistical channel information including spatial angle information and average channel energy, determines the pilot sequence of each user, and sends its sequence number to each user; The user sends a pilot signal, and the satellite or gateway performs channel estimation based on the received pilot signal, using a space-frequency two-level channel estimation method to obtain the estimated value of the channel parameters; The satellite or the gateway uses a graph-based pilot allocation algorithm to allocate pilots; the graph-based pilot allocation algorithm divides the user set into S groups, of which S represents the number of available pilots, so that users in the same group use the same pilot and users in different groups use different pilots; in the graph-based pilot allocation algorithm, the vertex set of the graph represents the user set, and the weights between the vertices of the graph represent the interference between the users; the weights between the vertices are calculated by spatial angle information and average channel energy; the graph-based pilot allocation algorithm comprises the following steps: Initialize a user set to which pilots have been allocated and a user set to which pilots have not been allocated; Allocate a pilot that minimizes the sum of weights between any two vertices in the group to each user to which a pilot is not allocated, and update the user set to which the pilot is allocated and the user set to which the pilot is not allocated; In the space-frequency two-level channel estimation method, the channel estimation value is obtained by first performing subcarrier-by-subcarrier space-domain processing on the satellite side received signal and then performing user-by-user frequency-domain processing; the space-domain processing is the same for all subcarriers; and the user-by-user frequency-domain processing can be quickly implemented using a Toeplitz system solver.

2. The method for estimating a massive MIMO OFDM satellite mobile communication channel according to claim 1, characterized in that: The statistical channel information is obtained by the uplink detection process or through the feedback information of each user terminal; during the uplink detection process, each user periodically sends a detection signal, and the satellite estimates the spatial angle information and average channel energy of each user based on the received detection signal; the feedback information of each user terminal is the user's geographical location information, spatial angle information, and average channel energy.

3. A method for estimating a massive MIMO OFDM satellite mobile communication channel, the method being applied to a user terminal, characterized in that: The user terminal periodically sends a detection signal to the satellite, or feeds back the user's geographical location information, spatial angle information, and average channel energy to the satellite for the satellite or the gateway to implement pilot allocation and channel estimation; The user terminal uses the Doppler frequency shift caused by satellite movement and the minimum propagation delay of long-distance propagation to perform frequency and time compensation on the uplink transmission signal; Each user terminal sends a pilot signal using a pilot sequence assigned by a satellite or a gateway; The Doppler frequency shift caused by the satellite movement and the minimum propagation delay of the long-distance propagation are estimated by the user terminal based on the received synchronization signal, or calculated based on the position information of the user terminal and the satellite; As the satellite or user terminal moves, the Doppler frequency shift and minimum propagation delay information are dynamically updated, and the frequency and time compensation amounts change adaptively accordingly; The satellite or the gateway uses a graph-based pilot allocation algorithm to allocate pilots; the graph-based pilot allocation algorithm divides the user set into S groups, of which S represents the number of available pilots, so that users in the same group use the same pilot and users in different groups use different pilots; in the graph-based pilot allocation algorithm, the vertex set of the graph represents the user set, and the weights between the vertices of the graph represent the interference between the users; the weights between the vertices are calculated by spatial angle information and average channel energy; the graph-based pilot allocation algorithm comprises the following steps: Initialize a user set to which pilots have been allocated and a user set to which pilots have not been allocated; Allocate a pilot that minimizes the sum of weights between any two vertices in the group to each user to which a pilot is not allocated, and update the user set to which the pilot is allocated and the user set to which the pilot is not allocated; In the space-frequency two-level channel estimation method, the channel estimation value is obtained by first performing subcarrier-by-subcarrier space-domain processing on the satellite side received signal and then performing user-by-user frequency-domain processing; the space-domain processing is the same for all subcarriers; the user-by-user frequency-domain processing can be quickly implemented using a Toeplitz system solver.

4. A satellite-side device for channel estimation of massive MIMO OFDM satellite mobile communication, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is loaded into a processor, the method for estimating a massive MIMO OFDM satellite mobile communication channel is implemented according to any one of claims 1 to 2.

5. A massive MIMO OFDM satellite mobile communication channel estimation user terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is loaded into a processor, the method for estimating a massive MIMO OFDM satellite mobile communication channel according to claim 3 is implemented.

6. A massive MIMO OFDM satellite mobile communication channel estimation system, comprising a satellite and a user terminal, characterized in that: The satellite is equipped with an antenna array to communicate with user terminals equipped with multiple antennas or a single antenna within its coverage area; the satellite or a gateway station associated with it is used to: The sequence of basic pilots modulated by different phase modulation factors is used as a pilot set; the satellite or the gateway station implements pilot allocation using statistical channel information including spatial angle information and average channel energy, and feeds back the pilot allocation result to each user; The user sends a pilot signal, and the satellite or gateway performs channel estimation based on the received pilot signal; The satellite or the gateway uses a graph-based pilot allocation algorithm to allocate pilots; the graph-based pilot allocation algorithm divides the user set into S groups, of which S represents the number of available pilots, so that users in the same group use the same pilot and users in different groups use different pilots; in the graph-based pilot allocation algorithm, the vertex set of the graph represents the user set, and the weights between the vertices of the graph represent the interference between the users; the weights between the vertices are calculated by the spatial angle information and the average channel energy; The satellite or the gateway station implements channel estimation by using a space-frequency two-stage channel estimation method according to the received pilot signal; in the space-frequency two-stage channel estimation method, the channel estimation value is obtained by first performing subcarrier-by-subcarrier spatial domain processing on the satellite side received signal and then performing user-by-user frequency domain processing; the spatial domain processing is the same for all subcarriers; the user-by-user frequency domain processing is quickly implemented using a Toeplitz system solver; During the movement of the satellite or each user terminal, the channel estimation result is dynamically updated as the statistical channel information changes; The user terminal is used to: periodically send a detection signal to the satellite, or feed back the user's geographical location information, spatial angle information, and average channel energy to the satellite, so that the satellite or the gateway station can implement pilot allocation and channel estimation; use the Doppler frequency shift caused by satellite movement and the minimum propagation delay of long-distance propagation to perform frequency and time compensation on the uplink transmission signal; and use the pilot sequence allocated by the satellite or the gateway station to send the pilot signal.

Images