A method and system for verifying the location and angle information of 5G cluster scattering channel modeling

Abstract

A method and system for verifying the location and angle information of 5G cluster scattering channel modeling includes: signal generation: a signal generator sends a channel sounding signal; access configuration: the channel sounding signal is connected to the channel simulator or channel modeling software under test, and parameters are configured; channel estimation: the channel is estimated based on the channel sounding signal; hybrid beamforming: a hybrid beamforming matrix is ​​obtained using the estimated channel; antenna radiation display: the radiation pattern of the shaped antenna is recalculated based on the hybrid beamforming matrix; the basic principle of the above method and system based on the hybrid beamforming matrix is ​​that the optimal and best-matched beam matrix is ​​derived from the angle of the channel ray, which can align the energy of the antenna with the receiver; if the energy of the shaped beam is not concentrated, or is only a two-dimensional plane, it indicates that the channel used for beamforming is incorrect.

Description

Technical Field

[0001] This invention relates to wireless channel modeling verification technology, and in particular to a method and system for verifying the location and angle information of 5G cluster scattering channel modeling. Background Technology

[0002] In wireless channel modeling methods, cluster scattering-based channel modeling is the closest to real wireless channels. Both 3GPP 38.901 and the 802.11 TGN series protocols employ cluster scattering-based channel modeling, which primarily simulates multipath delay and fading by simulating the superposition of multiple scattering propagation path vectors. Its main modeling principles are as follows:

[0003]

[0004]

[0005] Where n represents the nth cluster, and m represents the mth ray within cluster n. F rx,u,θ F rx,u,φ Let denot u be the radiation field strength of the receiving antenna array u in spherical coordinates θ and φ. φ is the angle of arrival on the horizontal plane. n,m,AOA and the angle θ of the vertical plane n,m,ZOA Convert to Cartesian coordinates:

[0006]

[0007] F tx,u,θ F tx,u,φ Let denot u be the radiation field intensity of the transmitting antenna array u in spherical coordinates θ and φ. To open the angle φ to the horizontal plane n,m,AOD and the angle θ of the vertical plane n,m,ZOD Convert to Cartesian coordinates:

[0008]

[0009] Let u be the Cartesian coordinates of the receiving antenna array. Let λ be the Cartesian coordinates of the receiving antenna array s. λ0 is the wavelength corresponding to the center frequency. κ represents the UE's movement speed. n,m This represents the cross-polarization ratio. All are random initial phases.

[0010] As can be seen from the modeling formula above, the phase of channel H is mainly affected by the arrival angle and departure angle. If the angle information is calculated incorrectly, then the channel is not the expected channel. While there are many existing measurement and verification methods for channel multipath number, multipath energy, and multipath delay, there is no unified method for verifying position and angle information. During R&D and testing, the alignment of the base station and terminal in channel modeling can be indirectly verified by verifying the RSRP (Received Signal Power Reduction) at different horizontal angles of the base station. However, in massive MIMO systems, beamforming is calculated using channel state information. If the position and angle information used in channel modeling is incorrect, then beamforming will not be aligned with the terminal, and the gain of the array antenna beamforming cannot be obtained. Therefore, the verification and testing of the position and angle information in channel modeling are crucial.

[0011] One existing method for measuring phase information is the PN sequence correlation method. This involves sending a pseudo-random sequence through the channel, correlating it with a local pseudo-random sequence, calculating the phase at each peak point, and then calculating the combined phase generated by different departure and arrival angles in the channel model. This combined phase is then compared with the measured value. The problem with the PN sequence correlation method is that it only calculates the correctness of the arrival and departure angles within the channel template parameters. However, in actual modeling, the location information of the UE and the base station needs to be considered, and the actual arrival and departure angles need to be converted based on this location information. Furthermore, the related calculation methods are not very accurate in setting the multipath energy threshold. In noisy environments, if the threshold is set too high, some multipath information will be lost. Summary of the Invention

[0012] Therefore, it is necessary to have an easily implementable method for verifying the location and angle information of 5G cluster scattering channel modeling.

[0013] Meanwhile, an easy-to-implement verification system for the location and angle information of 5G cluster scattering channel modeling is provided.

[0014] A method for verifying the location and angle information of 5G cluster scattering channel modeling includes:

[0015] Signal generation: The signal generator sends a channel sounding signal;

[0016] Access configuration: Connect the channel sounding signal to the channel simulator or channel modeling software under test and configure the parameters;

[0017] Channel estimation: Estimating the channel based on channel sounding signals;

[0018] Hybrid beamforming: Calculate the hybrid beamforming matrix using the estimated channel;

[0019] Antenna radiation demonstration: The radiation pattern of the shaped antenna is recalculated based on the hybrid beamforming matrix.

[0020] In a preferred embodiment, the signal generation includes: a signal generator transmitting a channel sounding signal, wherein the number of antennas in the transmitting antenna array is N. t The number of antennas in the receiving antenna array is N. r The number of signal transmission streams is N s The detection signal is an orthogonal frequency division multiplexing symbol, using a comb pilot structure, with N available subcarriers. sc The number of transformation points for the Fast Fourier Transform is N. fft The length of the cyclic prefix is ​​N cp For each antenna, first generate a length of N. sc A pseudo-random sequence pSeq is used to probe the channels of all transmit antennas. pSeq is then multiplied by an orthogonal mask matrix P, where P is N. t ×N t A three-dimensional matrix is ​​used to generate the detection signal on each transmitting antenna. The elements of the mask matrix P are either 1 or -1, and each column is orthogonal. P(i,j) represents the j-th element in the i-th row of matrix variable P, and P(i,j,k) represents the element in the three-dimensional matrix whose first dimension is i, second dimension is j, and third dimension is k. H The expression represents the conjugate transpose operation of matrix P, and svd(P) represents the svd decomposition of matrix P. The first OFDM symbol for each antenna is generated as follows:

[0021] sig 1j =ifft(pSeq*P(j,1)),j=1:N t ,

[0022] y t (1:N fft +N cp ,j)=[sig 1j (N fft -N cp +1:N fft )sig 1j ],j=1:N t ;

[0023] The second OFDM symbol for each antenna is generated as follows:

[0024] sig 2j =ifft(pSeq*P(j,2))j=1:N t ,

[0025] y t (N fft +N cp +1:2(N fft +N cp ),j)=[sig 2j(N fft -N cp +1:N fft )sig 2j ],j=1:N t ;

[0026] And so on, each antenna needs to generate N t One OFDM symbol, the final channel sounding signal y t For N t (N fft +N cp )×N t The matrix.

[0027] In a preferred embodiment, the access configuration includes: connecting the channel sounding signal to the channel simulator or channel modeling software to be tested, configuring the transmit and receive antenna parameters, transmit and receive position parameters, and channel template related parameters, wherein the signal dimension after passing through the channel simulator or channel modeling software is y. r (i,j),i=1:N t (N t +N cp ),j=1:N r .

[0028] In a preferred embodiment, the channel estimation includes: estimating the channel based on the probe signal, performing CP removal and IFT transformation on the signal of each receiving antenna, extracting the signal at the location of the subcarrier, and obtaining the frequency domain resource signal as rxGrid(k,i,j), k=1:N sc i = 1: N t j=1:N r The channel estimation uses the LS channel estimation algorithm, and the estimated channel is denoted as H(k,i,j), k=1:N sc i = 1: N t j=1:N r The LS calculation method is as follows:

[0029] h(k,i,j)=rxGrid(k,1:N t ,j)*P(1:N t ,i) / pSeq,k=1:N sc i = 1: N t j=1:N r .

[0030] In a preferred embodiment, the hybrid beamforming includes: using the estimated channel h(k,j,i) to calculate the hybrid beamforming matrix, wherein the digital beamforming matrix is ​​F. bb F bb For N s ×N sThe matrix; the simulated beamforming matrix is ​​F. rf , for N t ×N s The matrix is ​​initialized with all 1s. For F rf The matrix after deleting column j has dimension N. t ×(N s -1), Hybrid beamforming first calculates the simulated beamforming matrix F. rf Based on this, the digital beamforming matrix F is calculated. bb The calculation method is as follows:

[0031] First, calculate the target matrix M, which is N on each subcarrier. t ×N t matrix:

[0032] M = h(k,:,:) * (h(k,:,:)) H

[0033] For each stream j, calculate:

[0034] D j For (N) s -1)×(N s -1) Matrix

[0035] G j For N t ×N t matrix

[0036] For each receiving antenna i, we have:

[0037] η ij =∑ l≠i G j (i,l)F rf (l,j)

[0038]

[0039] After calculating the simulated beamforming matrix, the equivalent channel h is... eff =h T F rf Perform singular value decomposition:

[0040] [U,T,V]=svd(h eff )

[0041] h eff The unitary matrix obtained from singular value decomposition is used as the digital beamforming matrix F. bb :F bb =V.

[0042] In a preferred embodiment, the antenna radiation demonstration includes: based on the hybrid beamforming matrix F bb and F rf Recalculate the radiation pattern of the shaped antenna, draw the radiation pattern of the vertical plane, and the angle at which the main lobe is aligned is the relative angle between the receiver and the base station.

[0043] A verification system for the location and angle information of 5G cluster scattering channel modeling includes:

[0044] Signal generation module: The signal generator sends channel sounding signals;

[0045] Access configuration module: Connects the channel sounding signal to the channel simulator or channel modeling software under test and configures the parameters;

[0046] Channel estimation module: Estimates the channel based on the channel sounding signal;

[0047] Hybrid beamforming module: Calculates the hybrid beamforming matrix using the estimated channel;

[0048] Antenna radiation display module: Recalculates the radiation pattern of the shaped antenna based on the hybrid beamforming matrix.

[0049] In a preferred embodiment, the signal generation module includes: a signal generator that transmits channel sounding signals, wherein the number of antennas in the transmitting antenna array is N. t The number of antennas in the receiving antenna array is N. r The number of signal transmission streams is N s The detection signal is an orthogonal frequency division multiplexing symbol, using a comb pilot structure, with N available subcarriers. sc The number of transformation points for the Fast Fourier Transform is N. fft The length of the cyclic prefix is ​​N cp For each antenna, first generate a length of N. sc A pseudo-random sequence pSeq is used to probe the channels of all transmit antennas. pSeq is then multiplied by an orthogonal mask matrix P, where P is N. t ×N t A three-dimensional matrix is ​​used to generate the detection signal on each transmitting antenna. The elements of the mask matrix P are either 1 or -1, and each column is orthogonal. P(i,j) represents the j-th element in the i-th row of matrix variable P, and P(i,j,k) represents the element in the three-dimensional matrix whose first dimension is i, second dimension is j, and third dimension is k. H The expression represents the conjugate transpose operation of matrix P, and svd(P) represents the svd decomposition of matrix P. The first OFDM symbol for each antenna is generated as follows:

[0050] sig 1j =ifft(pSeq*P(j,1)),j=1:Nt ,

[0051] y t (1:N fft +N cp ,j)=[sig 1j (N fft -N cp +1:N fft )sig 1j ],j=1:N t ;

[0052] The second OFDM symbol for each antenna is generated as follows:

[0053] sig 2j =ifft(pSeq*P(j,2))j=1:N t ,

[0054] y t (N fft +N cp +1:2(N fft +N cp ),j)=[sig 2j (N fft -N cp +1:N fft )sig 2j ],j=1:N t ;

[0055] And so on, each antenna needs to generate N t One OFDM symbol, the final channel sounding signal y t For N t (N fft +N cp )×N t The matrix.

[0056] In a preferred embodiment, the access configuration module includes: connecting the channel sounding signal to the channel simulator or channel modeling software to be tested, configuring the transceiver antenna parameters, transceiver position parameters, and channel template related parameters, wherein the signal dimension after passing through the channel simulator or channel modeling software is y. r (i,j),i=1:N t (N t +N cp ),j=1:N r ;

[0057] The channel estimation module includes: estimating the channel based on the probe signal, performing CP removal and IFT transformation on the signal of each receiving antenna, extracting the signal at the location of the subcarrier, and obtaining the frequency domain resource signal as rxGrid(k,i,j), k=1:Nsc i = 1: N t j=1:N r The channel estimation uses the LS channel estimation algorithm, and the estimated channel is denoted as H(k,i,j), k=1:N sc i = 1: N t j=1:N r The LS calculation method is as follows: h(k,i,j)=rxGrid(k,1:N) t ,j)*P(1:N t ,i) / pSeq,k=1:N sc i = 1: N t j=1:N r .

[0058] In a preferred embodiment, the hybrid beamforming module includes: calculating the hybrid beamforming matrix using the estimated channel h(k,j,i), wherein the digital beamforming matrix is ​​F. bb F bb For N s ×N s The matrix; the simulated beamforming matrix is ​​F. rf , for N t ×N s The matrix is ​​initialized with all 1s. For F rf The matrix after deleting column j has dimension N. t ×(N s -1), Hybrid beamforming first calculates the simulated beamforming matrix F. rf Based on this, the digital beamforming matrix F is calculated. bb The calculation method is as follows:

[0059] First, calculate the target matrix M, which is N on each subcarrier. t ×N t matrix:

[0060] M = h(k,:,:) * (h(k,:,:)) H

[0061] For each stream j, calculate:

[0062] D j For (N) s -1)×(N s -1) Matrix

[0063] G j For N t ×N t matrix

[0064] For each receiving antenna i, we have:

[0065] η ij =∑ l≠i G j (i,l)F rf (l,j)

[0066]

[0067] After calculating the simulated beamforming matrix, the equivalent channel h is... eff =h T F rf Perform singular value decomposition:

[0068] [U,T,V]=svd(h eff )

[0069] h eff The unitary matrix obtained from singular value decomposition is used as the digital beamforming matrix F. bb :

[0070] F bb =V;

[0071] The antenna radiation demonstration includes: based on the hybrid beamforming matrix F bb and F rf Recalculate the radiation pattern of the shaped antenna, draw the radiation pattern of the vertical plane, and the angle at which the main lobe is aligned is the relative angle between the receiver and the base station.

[0072] The aforementioned method and system for verifying the location and angle information of 5G cluster scattering channel modeling first generates a channel sounding signal with a special structure. After the signal passes through a channel simulator / channel modeling software, the received signal is used to estimate the wireless channel and calculate the hybrid beamforming matrix. The radiation pattern after beamforming of the transmitting antenna array is then calculated. If the radiation pattern of the transmitting antenna array is not aligned with the expected location and angle, it indicates an error in the calculation of the ray angle information related to channel modeling. This invention provides a relatively accurate and easily implemented verification scheme for 5G cluster scattering channel modeling location and angle information. The basic principle of the hybrid beamforming matrix is ​​the optimal and best-matched beamforming matrix derived from the angle of the channel rays, which can align the antenna energy with the receiver. If the beam energy after beamforming is not concentrated, or is only a two-dimensional plane, it indicates an error in the channel used for beamforming. This invention's method for verifying the location and angle information of 5G cluster scattering channel modeling can be implemented in both hardware and software. Attached Figure Description

[0073] Figure 1 A flowchart illustrating a method for verifying the location and angle information of 5G cluster scattering channel modeling according to an embodiment of the present invention;

[0074] Figure 2 This is a schematic diagram of the channel detection signal structure of the present invention;

[0075] Figure 3 This is the vertical radiation section of the hybrid beamforming method at the horizontal angle of the terminal in a specific embodiment of the present invention;

[0076] Figure 4 This is a shaping result with incorrect position and angle information in a specific embodiment of the present invention. Detailed Implementation

[0077] The following examples are provided to help better understand the present invention, but are not intended to limit the invention.

[0078] To better simulate real-world wireless scattering environments, protocols such as 3GPP 38.901 and 802.11 TGN employ cluster scattering models. In this model, the channel is composed of the superposition of propagation path vectors from multiple scatterers. Different scatterer positions result in different departure and arrival angles for the rays, leading to multipath propagation with varying delays and energies. Channel modeling is primarily used for performance testing during the communication system development phase, and the accuracy of the modeling data significantly impacts system performance. However, there is no unified standard or method for verifying the modeling parameters. This invention presents a verification method for the position and angle information of 5G cluster scattering channel modeling, used to verify the correctness of the model's position and angle information.

[0079] like Figure 1 As shown, a method for verifying the location and angle information of 5G cluster scattering channel modeling according to an embodiment of the present invention includes:

[0080] Step S110, Signal generation: The signal generator sends a channel probe signal;

[0081] Step S120, Access Configuration: Connect the channel sounding signal to the channel simulator or channel modeling software under test, and configure the parameters;

[0082] Step S130, Channel estimation: Estimate the channel based on the channel sounding signal;

[0083] Step S140, Hybrid Beamforming: Calculate the hybrid beamforming matrix using the estimated channel;

[0084] Step S150, Antenna radiation display: Recalculate the radiation pattern of the shaped antenna based on the hybrid beamforming matrix.

[0085] Furthermore, in this embodiment, the signal generation involves a signal generator transmitting a channel sounding signal with a special structure. Let the number of antennas in the transmitting antenna array be N. t The number of antennas in the receiving antenna array is N. r The number of signal transmission streams is N sThe detection signal is an orthogonal frequency-division multiplexing (OFDM) symbol, employing a comb pilot structure, with N available subcarriers. sc The Fast Fourier Transform (fft) has N transformation points. fft The length of the cyclic prefix (CP) is N. cp Each antenna.

[0086] First, generate a length of N. sc The pseudo-random sequence pSeq is used. To probe the channels of all transmit antennas, pSeq needs to be multiplied by an orthogonal mask matrix P, where P is N. t ×N t A three-dimensional matrix is ​​used to generate the detection signal on each transmitting antenna. Elements in the mask matrix P can only be 1 or -1, and each column is orthogonal. In this invention, P(i,j) represents the j-th element in the i-th row of the matrix variable P. P(i,j,k) represents the element in the three-dimensional matrix whose first dimension is i, second dimension is j, and third dimension is k. H This represents the conjugate transpose operation of matrix P. `svd(P)` represents the SVD decomposition of matrix P. The channel sounding signal structure for each antenna is as follows: Figure 2 As shown.

[0087] The first OFDM symbol for each antenna is generated as follows:

[0088] sig 1j =ifft(pSeq*P(j,1)),j=1:N t ,

[0089] y t (1:N fft +N cp ,j)=[sig 1j (N fft -N cp +1:N fft )sig 1j ],j=1:N t ;

[0090] The second OFDM symbol for each antenna is generated as follows:

[0091] sig 2j =ifft(pSeq*P(j,2))j=1:N t ,

[0092] y t (N fft +N cp+1:2(N fft +N cp ),j)=[sig 2j (N fft -N cp +1:N fft )sig 2j ],j=1:N t ;

[0093] And so on, each antenna needs to generate N t One OFDM symbol. Final channel sounding signal y t For N t (N fft +N cp )×N t The matrix. ifft() is the inverse fast Fourier transform.

[0094] Furthermore, the access configuration steps in this embodiment are as follows: The probe signal is connected to the channel simulator or channel modeling software under test, and the transmit / receive antenna parameters, transmit / receive position parameters, and channel template-related parameters are configured. The signal dimension after passing through the channel simulator or channel modeling software is y. r (i,j),i=1:N t (N t +N cp ),j=1:N r .

[0095] Furthermore, the channel estimation in this embodiment is based on the probe signal. First, the signal from each receiving antenna is subjected to CP (Cyclic Prefix) removal and IFT (Inverse Fast Fourier Transform) to extract the signal at the subcarrier location. The resulting frequency domain resource signal is rxGrid(k,i,j), k=1:N sc i = 1: N t j=1:N r Channel estimation can be performed using the LS channel estimation algorithm, denoted as H(k,i,j), k=1:N. sc i = 1: N t j=1:N r The LS calculation method is as follows:

[0096] h(k,i,j)=rxGrid(k,1:N t ,j)*P(1:N t ,i) / pSeq,k=1:N sc i = 1: N t j=1:N r .

[0097] Furthermore, in this embodiment, hybrid beamforming is performed by estimating the channel h(k,j,i) to obtain the hybrid beamforming matrix, where the digital beamforming matrix is ​​F. bb F bb For N s ×N s The matrix; the simulated beamforming matrix is ​​F. rf , for N t ×N s The matrix is ​​initialized with all 1s. For F rf The matrix after deleting column j has dimension N. t ×(N s -1). Hybrid beamforming consists of two steps: first, the simulated beamforming matrix F is calculated. rf Based on this, the digital beamforming matrix F is calculated. bb The calculation method is as follows:

[0098] First, calculate the target matrix M, which is N on each subcarrier. t ×N t matrix:

[0099] M = h(k,:,:) * (h(k,:,:)) H

[0100] For each stream j, calculate:

[0101] D j For (N) s -1)×(N s -1) Matrix

[0102] G j For N t ×N t matrix

[0103] For each receiving antenna i, we have:

[0104] η ij =∑ l≠i G j (i,l)F rf (l,j)

[0105]

[0106] After calculating the simulated beamforming matrix, the equivalent channel h is... eff =h T F rf Perform singular value decomposition:

[0107] [U,T,V]=svd(h eff )

[0108] h eff The unitary matrix obtained from singular value decomposition is used as the digital beamforming matrix F. bb :F bb =V.

[0109] Furthermore, the antenna radiation in this embodiment is demonstrated as follows: based on the hybrid beamforming matrix F bb and F rf Recalculate the radiation pattern of the shaped antenna. Draw the radiation pattern in the vertical plane; the angle at which the main lobe is aligned is the relative angle between the receiver and the base station. The radiation pattern in the vertical plane includes the vertical cross-sectional view at the corresponding horizontal angle.

[0110] This invention can generate the vertical radiation pattern of an antenna by using the pattern in maltab, inputting the calculated precoding matrix and antenna parameters.

[0111] A verification system for the modeling location and angle information of a 5G cluster scattering channel according to an embodiment of the present invention includes:

[0112] Signal generation module: The signal generator sends channel sounding signals;

[0113] Access configuration module: Connects the channel sounding signal to the channel simulator or channel modeling software under test and configures the parameters;

[0114] Channel estimation module: Estimates the channel based on the channel sounding signal;

[0115] Hybrid beamforming module: Calculates the hybrid beamforming matrix using the estimated channel;

[0116] Antenna radiation display module: Recalculates the radiation pattern of the shaped antenna based on the hybrid beamforming matrix.

[0117] Furthermore, in this embodiment, the signal generation module transmits a channel sounding signal with a special structure. Let the number of antennas in the transmitting antenna array be N. t The number of antennas in the receiving antenna array is N. r The number of signal transmission streams is N s The detection signal is an orthogonal frequency-division multiplexing (OFDM) symbol, employing a comb pilot structure, with N available subcarriers. sc The Fast Fourier Transform (fft) has N transformation points. fft The length of the cyclic prefix (CP) is N. cp Each antenna.

[0118] First, generate a length of N. sc The pseudo-random sequence pSeq is used. To probe the channels of all transmit antennas, pSeq needs to be multiplied by an orthogonal mask matrix P, where P is N. t ×N t A three-dimensional matrix is ​​used to generate the detection signal on each transmitting antenna. Elements in the mask matrix P can only be 1 or -1, and each column is orthogonal. In this invention, P(i,j) represents the j-th element in the i-th row of the matrix variable P. P(i,j,k) represents the element in the three-dimensional matrix whose first dimension is i, second dimension is j, and third dimension is k. H This represents the conjugate transpose operation of matrix P. `svd(P)` represents the SVD decomposition of matrix P. The channel sounding signal structure for each antenna is as follows: Figure 2 As shown.

[0119] The first OFDM symbol for each antenna is generated as follows:

[0120] sig 1j =ifft(pSeq*P(j,1)),j=1:N t ,

[0121] y t (1:N fft +N cp ,j)=[sig 1j (N fft -N cp +1:N fft )sig 1j ],j=1:N t ;

[0122] The second OFDM symbol for each antenna is generated as follows:

[0123] sig 2j =ifft(pSeq*P(j,2))j=1:N t ,

[0124] y t (N fft +N cp +1:2(N fft +N cp ),j)=[sig 2j (N fft -N cp +1:N fft )sig 2j ],j=1:N t ;

[0125] And so on, each antenna needs to generate N tOne OFDM symbol. Final channel sounding signal y t For N t (N fft +N cp )×N t The matrix.

[0126] Furthermore, the access configuration module in this embodiment: connects the probe signal to the channel simulator or channel modeling software under test, and configures the transmit and receive antenna parameters, transmit and receive position parameters, and channel template related parameters. The signal dimension after passing through the channel simulator or channel modeling software is y. r (i,j),i=1:N t (N t +N cp ),j=1:N r .

[0127] Furthermore, the channel estimation module in this embodiment estimates the channel based on the probe signal. First, it performs CP (Cyclic Prefix) removal and IFT (Inverse Fast Fourier Transform) on the signal from each receiving antenna to extract the signal at the subcarrier location, resulting in the frequency domain resource signal rxGrid(k,i,j), k=1:N sc i = 1: N t j=1:N r Channel estimation can be performed using the LS channel estimation algorithm, denoted as H(k,i,j), k=1:N. sc i = 1: N t j=1:N r The LS calculation method is as follows:

[0128] h(k,i,j)=rxGrid(k,1:N t ,j)*P(1:N t ,i) / pSeq,k=1:N sc i = 1: N t j=1:N r .

[0129] Furthermore, the hybrid beamforming module in this embodiment: uses the estimated channel h(k,j,i) to calculate the hybrid beamforming matrix, where the digital beamforming matrix is ​​F. bb F bb For N s ×N s The matrix; the simulated beamforming matrix is ​​F. rf , for N t ×N s The matrix is ​​initialized with all 1s. For F rf The matrix after deleting column j has dimension N.t ×(N s -1). Hybrid beamforming consists of two steps: first, the simulated beamforming matrix F is calculated. rf Based on this, the digital beamforming matrix F is calculated. bb The calculation method is as follows:

[0130] First, calculate the target matrix M, which is N on each subcarrier. t ×N t matrix:

[0131] M = h(k,:,:) * (h(k,:,:)) H

[0132] For each stream j, calculate:

[0133] D j For (N) s -1)×(N s -1) Matrix

[0134] G j For N t ×N t matrix

[0135] For each receiving antenna i, we have:

[0136] η ij =∑ l≠i G j (i,l)F rf (l,j)

[0137]

[0138] After calculating the simulated beamforming matrix, the equivalent channel h is... eff =h T F rf Perform singular value decomposition:

[0139] [U,T,V]=svd(h eff )

[0140] h eff The unitary matrix obtained from singular value decomposition is used as the digital beamforming matrix F. bb :F bb =V.

[0141] Furthermore, the antenna radiation demonstration module in this embodiment: based on the hybrid beamforming matrix F bb and F rfRecalculate the radiation pattern of the shaped antenna. Draw the radiation pattern in the vertical plane; the angle at which the main lobe is aligned is the relative angle between the receiver and the base station. The radiation pattern in the vertical plane includes the vertical cross-sectional view at the corresponding horizontal angle.

[0142] This invention relates to a method and system for verifying the location and angle information of 5G cluster scattering channel modeling. Based on the fundamental principle of hybrid beamforming matrices, the optimal and best-matched beam matrix is ​​derived from the angle of the channel rays, allowing the antenna energy to be aligned with the receiver. If the beam energy after beamforming is not concentrated, or is merely a two-dimensional plane, it indicates an error in the channel used for beamforming. This method for verifying the location and angle information of 5G cluster scattering channel modeling can be implemented in both hardware and software.

[0143] In a preferred embodiment of the verification method and system for the modeling position and angle information of the 5G cluster scattering channel of the present invention, in a Massive MIMO system with a center frequency of 28 GHz, the transmit antenna array is a two-dimensional uniform array with a total of 64 transmit antennas, 2 receive antennas, and a total of 2 signal streams.

[0144] Signal Generation: First, a channel sounding signal is generated. The sounding signal uses 1024 subcarriers with a CP length of 72, employing a comb pilot. The pilot is a pseudo-random sequence. For different antennas, a time-domain mask is used to ensure that the sounding signals are orthogonal in the time domain. The time-domain mask matrix for 64 antennas is P(i,j), i = 1:64, j = 1:64, where the value of P(i,j) is 1 or -1, its function being to ensure the orthogonality of the time-domain signals of the 64 antennas.

[0145] First, a pseudo-random sequence of length 1024 is generated, where pSeq(i), i = 1:1024, is a complex pseudo-random sequence. For each transmit antenna that needs to probe the channel, a probe signal is generated:

[0146] sig 11 =ifft(pSeq*P(1,1))

[0147] y t (1:1096,1)=[sig 11 (953:1024)sig 11 ]

[0148] sig 12 =ifft(pSeq*P(2,1))

[0149] y t (1:1096,2)=[sig 12 (953:1024)sig 12 ]

[0150] And so on:

[0151] sig 21 =ifft(pSeq*P(1,2))

[0152] y t (1097:2192,1)=[sig 21 (953:1024)sig 21 ]

[0153] Following this logic, the transmitted signal can finally be calculated as follows:

[0154] y t (i,j), i=1:70144,j=1:64.

[0155] Access Configuration: The transmitted signal is sent to the channel modeling software. The channel configuration used in this invention adopts the 3GPP 38.901 CDL-D series. The base station coordinates are [0 0 0], and the UE coordinates have three possibilities:

[0156] [8.4 18.2 27.4], with spherical coordinates relative to the base station of [65.1° 53.8°].

[0157] [71.9 -191.5 -163.7], with spherical coordinates relative to the base station of [-69.4°-38.7°];

[0158] [170.2 51.6 181.3], with spherical coordinates relative to the base station of [16.7°45.6°];

[0159] Let the signal after channel modeling be y. r (i,j), i=1:70144,j=1:2.

[0160] Channel estimation: The received signal is demodulated using OFDM. Let the demodulated signal be:

[0161] sig 11 =y r (73:1096, 1)

[0162] rxGrid(1:1024,1,1)=ifft(sig 11 )

[0163] sig 12 =y r (1169:2192,1)

[0164] rxGrid(1:1024,2,1)=ifft(sig 12 )

[0165] And so on:

[0166] sig 164 =y r (69121:70144,1)

[0167] rxGrid(1:1024,64,1)=ifft(sig 164 )

[0168] Similarly, OFDM demodulation can be performed on the second receiving antenna to obtain its frequency domain data rxGrid(1:1024,1:64,2). From this, the frequency domain data of the receiving antenna can be obtained as rxGrid(k,i,j), k=1:1024,j=1:64,j=1:2. Channel estimation is then performed on the frequency domain data.

[0169] h(1:1024,i,j)=rxGrid(k,1:64,j)*P(1:64,i) / pSeq

[0170] At this point, h on each subcarrier is a 64×2 matrix.

[0171] Hybrid beamforming: Calculate the hybrid beamforming matrix based on the estimated channel.

[0172] For each subcarrier, calculate the target matrix:

[0173] M = h(k,:,:) * h(k,:,:) H M is a 64×64 matrix. Let the noise variance be δ. 2 Let F be the analog beamforming matrix. rf The initial value is a 64×2 matrix of all ones. For F rf The matrix after deleting the j-th column vector is a 64×1 matrix.

[0174] Calculate for each stream:

[0175]

[0176]

[0177] For each receiving antenna i, we have:

[0178]

[0179]

[0180] Calculate the simulated beamforming matrix F rf Then, let the equivalent channel be h. eff =h T Frf By performing SVD decomposition on the equivalent channel, the digital beamforming matrix can be obtained, as follows:

[0181] [U,T,V]=svd(h eff )

[0182] F bb =V

[0183] Antenna radiation demonstration: Substitute the beamforming matrix above into the antenna radiation calculation and draw the vertical cross-section at the corresponding horizontal angle, as shown below. Figure 3 As shown above, Figure 3 In the three figures, the main lobe of the antenna pattern on the vertical section corresponding to the UE's horizontal angle is aligned with the vertical angle of the UE (User Equipment) relative to the base station. However, if the angle of the channel modeling ray is not based on the relative position information of the UE and the base station, the resulting antenna pattern would be as follows: Figure 4 As shown, the vertical beam is not formed at this time; only a planar beam exists. Therefore, the verification method and system for the position and angle information of the 5G cluster scattering channel modeling of this invention can be used to verify the correctness of the position and angle information of the channel modeling.

[0184] For verifying the current channel modeling position and angle information, the verification method for 5G cluster scattering channel modeling position and angle information of this invention is a relatively accurate and easy-to-implement verification method. First, a channel sounding signal with a special structure is generated. After the signal passes through a channel simulator / channel modeling software, the received signal is used to estimate the wireless channel and calculate the hybrid beamforming matrix. Then, the radiation pattern after beamforming of the transmitting antenna array is calculated. If the radiation pattern of the transmitting antenna array is not aligned with the expected position and angle, it indicates that the calculation of the ray angle related information in the channel modeling is incorrect.

[0185] Based on the above-described preferred embodiments according to this application, and through the foregoing description, those skilled in the art can make various changes and modifications without departing from the technical concept of this application. The technical scope of this application is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A method for verifying the location and angle information of 5G cluster scattering channel modeling, characterized in that, include: Signal generation: The signal generator sends a channel sounding signal; Access configuration: Connect the channel sounding signal to the channel simulator or channel modeling software under test and configure the parameters; Channel estimation: Estimating the channel based on channel sounding signals; Hybrid beamforming: Calculate the hybrid beamforming matrix using the estimated channel; Antenna radiation demonstration: The radiation pattern of the shaped antenna is recalculated based on the hybrid beamforming matrix; The hybrid beamforming includes: utilizing an estimated channel Find the hybrid beamforming matrix, where the digital beamforming matrix is: , for The matrix; the simulated beamforming matrix is ,for The matrix is ​​initialized with all 1s. for The first one was deleted. The matrix after the column has dimensions of Hybrid beamforming first calculates the simulated beamforming matrix. Based on this, the digital beamforming matrix is ​​calculated. The calculation method is as follows: First, calculate the target matrix. On each subcarrier matrix: ， For each stream calculate: , for matrix , for matrix For each receiving antenna have: ， ； After calculating the simulated beamforming matrix, the equivalent channel is... Perform singular value decomposition: ， The unitary matrix obtained from singular value decomposition is used as the digital beamforming matrix. : .

2. The method for verifying the location and angle information of 5G cluster scattering channel modeling according to claim 1, characterized in that, The signal generation includes: a signal generator transmitting a channel sounding signal, wherein the number of antennas in the transmitting antenna array is... The number of antennas in the receiving antenna array is The number of signal transmission streams is The detection signal is an orthogonal frequency division multiplexing symbol, employing a comb pilot structure, with the number of available subcarriers being... The number of transform points for the Fast Fourier Transform is The length of the cyclic prefix is Each antenna first generates a length of... pseudo-random sequence Detect the channels of all transmitting antennas, With orthogonal mask matrix Multiply, for A 3D matrix is ​​used to generate the detection signal on each transmit antenna, and a mask matrix is ​​used for this purpose. The elements in the array are either 1 or -1, and the columns are orthogonal. All represent matrix variables The element in the i-th row and j-th position, This represents an element in a three-dimensional matrix whose first dimension is i, second dimension is j, and third dimension is k. Indicates a matrix The conjugate transpose operation. Indicates a matrix The SVD decomposition generates the first OFDM symbol for each antenna as follows: ， ； The second OFDM symbol for each antenna is generated as follows: ， ； And so on, each antenna needs to generate One OFDM symbol, final channel sounding signal for The matrix.

3. The method for verifying the location and angle information of 5G cluster scattering channel modeling according to claim 2, characterized in that, The access configuration includes: connecting the channel sounding signal to the channel simulator or channel modeling software to be tested, configuring the transmit and receive antenna parameters, transmit and receive location parameters, and channel template related parameters, and the signal dimension after passing through the channel simulator or channel modeling software is... .

4. The method for verifying the location and angle information of 5G cluster scattering channel modeling according to claim 3, characterized in that, The channel estimation includes: estimating the channel based on the probe signal, performing CP removal and IFT transformation on the signal from each receiving antenna, extracting the signal at the subcarrier location, and obtaining the frequency domain resource signal. The channel estimation uses the LS channel estimation algorithm, and the estimated channel is denoted as . The LS calculation method is as follows: 。 5. The method for verifying the location and angle information of 5G cluster scattering channel modeling according to claim 1, characterized in that, The antenna radiation demonstration includes: based on the hybrid beamforming matrix and Recalculate the radiation pattern of the shaped antenna, draw the radiation pattern of the vertical plane, and the angle at which the main lobe is aligned is the relative angle between the receiver and the base station.

6. A verification system for the location and angle information of 5G cluster scattering channel modeling, characterized in that, include: Signal generation module: The signal generator sends channel sounding signals; Access configuration module: Connects the channel sounding signal to the channel simulator or channel modeling software under test and configures the parameters; Channel estimation module: Estimates the channel based on the channel sounding signal; Hybrid beamforming module: Calculates the hybrid beamforming matrix using the estimated channel; Antenna radiation display module: Recalculates the radiation pattern of the shaped antenna based on the hybrid beamforming matrix; The hybrid beamforming module includes: utilizing channel estimation Find the hybrid beamforming matrix, where the digital beamforming matrix is: , for The matrix; the simulated beamforming matrix is ,for The matrix is ​​initialized with all 1s. for The first one was deleted. The matrix after the column has dimensions of Hybrid beamforming first calculates the simulated beamforming matrix. Based on this, the digital beamforming matrix is ​​calculated. The calculation method is as follows: First, calculate the target matrix. On each subcarrier matrix: ， For each stream calculate: , for matrix , for matrix For each receiving antenna have: ， ； After calculating the simulated beamforming matrix, the equivalent channel is... Perform singular value decomposition: ， The unitary matrix obtained from singular value decomposition is used as the digital beamforming matrix. : ; The antenna radiation demonstration includes: based on the hybrid beamforming matrix and Recalculate the radiation pattern of the shaped antenna, draw the radiation pattern of the vertical plane, and the angle at which the main lobe is aligned is the relative angle between the receiver and the base station.

7. The verification system for the location and angle information of 5G cluster scattering channel modeling according to claim 6, characterized in that, The signal generation module includes: a signal generator that transmits channel sounding signals, wherein the number of antennas in the transmitting antenna array is . The number of antennas in the receiving antenna array is The number of signal transmission streams is The detection signal is an orthogonal frequency division multiplexing symbol, employing a comb pilot structure, with the number of available subcarriers being... The number of transform points for the Fast Fourier Transform is The length of the cyclic prefix is Each antenna first generates a length of... pseudo-random sequence Detect the channels of all transmitting antennas, With orthogonal mask matrix Multiply, for A 3D matrix is ​​used to generate the detection signal on each transmit antenna, and a mask matrix is ​​used for this purpose. The elements in the array are either 1 or -1, and the columns are orthogonal. All represent matrix variables The element in the i-th row and j-th position, This represents an element in a three-dimensional matrix whose first dimension is i, second dimension is j, and third dimension is k. Indicates a matrix The conjugate transpose operation. Indicates a matrix The SVD decomposition generates the first OFDM symbol for each antenna as follows: ， ； The second OFDM symbol for each antenna is generated as follows: ， ； And so on, each antenna needs to generate One OFDM symbol, final channel sounding signal for The matrix.

8. The verification system for the location and angle information of 5G cluster scattering channel modeling according to claim 7, characterized in that, The access configuration module includes: connecting the channel sounding signal to the channel simulator or channel modeling software to be tested, configuring the transmit and receive antenna parameters, transmit and receive location parameters, and channel template related parameters, and the signal dimension after passing through the channel simulator or channel modeling software is... ; The channel estimation module includes: estimating the channel based on the probe signal, performing CP removal and IFT transformation on the signal from each receiving antenna, extracting the signal at the subcarrier location, and obtaining the frequency domain resource signal. The channel estimation uses the LS channel estimation algorithm, and the estimated channel is denoted as . The LS calculation method is as follows: 。

Images