CN115208442A - Two-stage beam training method of ultra-large scale array communication system - Google Patents
Two-stage beam training method of ultra-large scale array communication system Download PDFInfo
- Publication number
- CN115208442A CN115208442A CN202210597131.7A CN202210597131A CN115208442A CN 115208442 A CN115208442 A CN 115208442A CN 202210597131 A CN202210597131 A CN 202210597131A CN 115208442 A CN115208442 A CN 115208442A
- Authority
- CN
- China
- Prior art keywords
- beam training
- channel
- stage
- codebook
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/0413—MIMO systems
- H04B7/0426—Power distribution
- H04B7/043—Power distribution using best eigenmode, e.g. beam forming or beam steering
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/391—Modelling the propagation channel
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/0413—MIMO systems
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/0413—MIMO systems
- H04B7/0456—Selection of precoding matrices or codebooks, e.g. using matrices antenna weighting
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Abstract
The invention discloses a two-stage beam training method of a super-large-scale array communication system. In the first stage of beam training, each sub-array of the super-large-scale multiple-input multiple-output system is subjected to analog combination by using a series of far-field steering vectors. In the second stage of beam training, aiming at the line-of-sight path channel estimation, firstly, a dedicated digital combiner is designed for each code word in a preset codebook. Then, the digital processing unit performs digital combining processing on the signals subjected to the first-stage analog combining by using a dedicated digital combiner. And finally, outputting the code words capable of realizing the maximum combining power in the preset codebook as the result of beam training. And aiming at the multipath channel estimation, performing digital processing on the signals after analog combination in the first phase of beam training to complete the reconstruction of a channel transmission matrix. On the premise of greatly reducing the training overhead, the method can approach the performance of the existing mixed field beam scanning scheme and complete high-precision channel estimation.
Description
Technical Field
The invention belongs to the field of millimeter wave wireless communication, and relates to a two-stage beam training method of a super-large-scale array communication system.
Background
Massive mimo is a key technology of conventional fifth generation wireless communication. The base station is provided with a large number of antennas, and the spectrum efficiency can be obviously improved through a beam forming technology. In the sixth generation of wireless communication in the future, a super-large-scale multi-input multi-output system with more antennas is adopted, and the spectrum efficiency is further improved.
Due to the large energy consumption of the rf chain, it is not practical to configure one antenna with one dedicated rf chain for very large-scale mimo systems. Therefore, hybrid precoding/combining results are typically used in very large scale mimo systems, i.e. a small number of rf chains are connected to a large number of antennas. Hybrid precoding can be generally divided into a full-connection structure and a partial-connection structure according to a connection mode of a radio frequency chain and an antenna. In a fully connected configuration, each rf chain is connected to an antenna individually, which is inefficient due to the high insertion loss of the circuit. In contrast, in the partial connection result, each rf chain is connected to only a partial connection structure of one separate sub-array, which is more practical in terms of circuit configuration and system performance.
The super-large-scale multi-input multi-output system is provided with a super-large number of antennas, so that the characteristics of the system are different from those of the traditional large-scale multi-input multi-output system. According to the distance between a radiation source and a super-large-scale multi-input multi-output system, a radiation field can be divided into a near field and a far field by taking Rayleigh distance as a boundary. At a specific carrier frequency, the rayleigh distance increases quadratically with the number of antennas, and thus the far field assumption in the conventional massive mimo system may not be applicable to the near field model of the ultra-massive mimo system. The channel state information acquisition method suitable for the near-field channel of the ultra-large-scale multi-input multi-output system becomes a research hotspot.
A direct method for acquiring channel state information is hybrid field beam scanning, which is to exhaust all codewords in a hybrid codebook to acquire a best matching codeword, but this scheme consumes a large amount of training overhead and reduces communication efficiency. Document [1] proposes a P-SOMP algorithm to estimate the near-field channel, however, directional beamforming with high beam gain is not formed in the training process of this method, which may degrade the channel estimation performance in the case of low signal-to-noise ratio. (document [1]: M.Cui and L.Dai, "Channel estimation for extreme large-scale MIMO. Document [2] proposes a codebook-based far-field exhaustive beam training method that does not consider near-field channels, although high beamforming gain can be obtained in the far field. (document [2]: A. Alkhateeb, G. Leus, and R.W.Heath, "Limited feedback hybrid coding for Multi-user millimeter wave systems," IEEE trans. Wireless Commun., vol.14, no.11, pp.6481-6494, nov.2015.).
Disclosure of Invention
The purpose of the invention is as follows: aiming at the problems, the invention provides a two-stage beam training method of a super-large-scale array communication system. The method comprises two stages: in the first stage of beam training, each subarray of the very large scale multiple input multiple output system is combined in an analog mode by using a series of far field steering vectors. In the second stage of beam training, for the line-of-sight path channel estimation, first, a dedicated digital combiner is designed for each codeword in the preset codebook. Then, the digital processing unit performs digital combining processing on the signals subjected to the first-stage analog combining by using a dedicated digital combiner. And finally, outputting the code word capable of realizing the maximum combining power in the preset codebook as a beam training result. And for multipath channel estimation, performing digital processing on the signals after analog combination in the first stage of beam training to complete reconstruction of a channel transmission matrix. On the premise of greatly reducing training overhead, the method can approach the performance of the existing mixed field beam scanning scheme and complete high-precision channel estimation.
The technical scheme is as follows: in order to realize the purpose of the invention, the technical scheme adopted by the invention is as follows: a two-stage beam training method of a super-large scale array communication system comprises the following steps:
(1) Setting basic parameters of a super-large-scale multi-input multi-output system;
(2) Constructing a super-large-scale multi-input multi-output system between a base station and a user and a channel model in the system;
(3) Designing a hybrid codebook for estimating the channel in step (2);
(4) Based on the mixed codebook in the step (3), aiming at the sight distance path channel estimation, and aiming at determining the most suitable guide vector of the channel, two-stage beam training is carried out on the super-large-scale multi-input multi-output system;
(5) And (5) performing two-stage beam training on the ultra-large scale multi-input multi-output system for channel reconstruction aiming at multi-path channel estimation based on the first-stage beam training in the step (4).
Preferably, in the step (1), the method for setting basic parameters of the very large scale multiple input multiple output system is as follows: in the uplink beam training scene arranged between a base station and a user, the base station end antenna arrays are uniform linear arrays with the interval of half-wavelength, the number of the antennas is N, a partial connection hybrid combination structure is adopted, the structure comprises analog combination and digital combination, and the number of radio frequency links is N RF The antenna comprises N RF A plurality of non-overlapping sub-arrays; each subarray has M = N/N RF The root antenna is connected with a radio frequency chain after analog combination; all N RF The radio frequency chains are connected to a digital processing unit for digital combination; the user terminal adopts a single antenna.
Preferably, in the step (2), the method for constructing the super-large-scale multiple-input multiple-output system model comprises the following steps:
(2.1) constructing a super-large-scale multi-input multi-output system model between a base station and a user
In the uplink beam training, the training symbol sent by the user terminal is x k K =1,2, \ 8230;, K, where K is the signal length and the channel between the base station and the user is denoted as h, the received signal obtained after the base station has undergone mixing and combining is denoted as:
y k ＝v k W k hx k +v k W k η
wherein, W k Represents a simulated merge, v k Denotes a combination of numbers, η denotes additive white Gaussian noise, andnoise satisfaction
Mean is μ variance is σ 2 Complex gaussian distribution of (a);
(2.2) constructing a channel model in a very large-scale MIMO system
Setting a multi-path channel composed of a main path and multiple sub-paths between a user and a base station, wherein N antennas of the base station are arranged along the y-axis of a Cartesian coordinate system, and the coordinates of the nth antenna are (0, delta) n λ), where N =1,2, \ 8230;, N,
λ represents the wavelength, and the coordinates of the center of the t-th sub-array are (0, Δ) t λ), wherein t =1,2, \ 8230;, N RF ，Δ t ＝[(2t-1)M-N]4, the coordinates of the user are represented as p 1 ＝(r 1 cosθ 1 ,r 1 sinθ 1 ) Wherein r is 1 Representing the distance between the user and the origin of coordinates, theta 1 ∈[-π/2,π/2]Representing the angle of the user with respect to the positive half-axis of the x-axis, and the coordinates of the scattering points in the ith path are denoted p l ＝(r l cosθ l ,r l sinθ l ) Wherein l>2，r l Representing the distance, theta, between the user and the origin of coordinates l ∈[-π/2,π/2]Representing the angle, p, of the user with respect to the positive half-axis of the x-axis l And the nth antenna is expressed as
Wherein, the first and the second end of the pipe are connected with each other,
is the sine of the angle, and Ω l ∈[-1,1](ii) a The channel between the base station and the user is modeled as follows:
wherein, L and g l Respectively representing the number of paths and the channel gain of the ith path, and α (·) represents a channel steering vector defined as:
the near and far fields are typically distinguished using rayleigh distances, expressed as
Wherein D = N λ/2 denotes an antenna array aperture; when the distance between the radiation source and the base station exceeds Z, the wireless channel is defined as a far-field channel; conversely, the wireless channel is defined as a near-field channel;
when the distance r l >2D 2 At/λ, the following approximation is used,
wherein the steering vector of the far-field channel is defined as beta (N, omega) l )。
Preferably, in step (3), the method for designing the hybrid codebook for estimating the channel in step (2) is as follows:
(3.1) setting C h Representing a hybrid codebook, C f Representing a far-field codebook, C n Representing a near-field codebook;
(3.2) far-field codebook C described in step (3.1) f The nth codeword in (1) is represented as
(3.3) step (3.1)The near-field codebook C described in (1) n The code word in (1) is designed by the following steps:
(1) dividing the near field into N parts which are equal in angle dimension and S parts which are unequal in distance dimension;
(2) the nth quantization angle is theta n ＝(2n-1-N)/N；
(3) At the nth angle, the s quantized distance is
(4) Near field codebook C n Is shown as
Wherein, [ C ] n ] :,s ＝α(N,Θ n ,d n,s )；
(5) The hybrid codebook described in step (3.1) is represented as
Preferably, in step (4), based on the hybrid codebook in step (3), aiming at the line-of-sight path channel estimation, two-stage beam training is performed on the very large scale multiple input multiple output system, and the method is as follows:
(4.1) for each subarray, a common beam training DFT codebook of
Wherein phi is m ＝(2m-1-M)/M，m＝1,2,…,M；
(4.2) in the first phase of beam training, the user end sends training symbols to the base station, the training symbols last for M time slots, the base station receives the training symbols sequentially, and for the kth beam training, the received signals without digital combination are expressed as
Wherein the kth analog combination is represented as
blkdig {. Denotes a block diagonalization operation, so far, the first stage in the two-stage beam training scheme is completed;
(4.3) analog combining is designed in the first stage, and digital combining v is designed in the second stage by testing NS + N code words covered by the mixed codebook p P =1,2, \ 8230n, NS + N, representing the p-th codeword in the mixed codebook as
Wherein, the first and the second end of the pipe are connected with each other,
representing a lower bound on an element value;
(4.4) for the p-th codeword, the sine of the angle of the corresponding quantization position with respect to the center of the t-th sub-array is
Wherein, delta t ＝[(2t-1)M-N]/4；
(4.5) setting
Wherein
(4.6) numerical incorporation of
Wherein the content of the first and second substances,
(4.7) digital combining of the received signals of the first stage
After is shown as
Wherein the content of the first and second substances,
(4.8) comparing all NS + N combined signals, and selecting the quantization position corresponding to the signal with the maximum energy, wherein the quantization position is expressed as
(4.9) the finally selected most suitable channel codeword is
Preferably, in the step (5), based on the first-stage beam training in the step (4), aiming at the multipath channel estimation, two-stage beam training is performed on the very large-scale multiple-input multiple-output system.
(5.1) on the basis of the first stage, representing the analog combined signal as
Definition of
Wherein, the first and the second end of the pipe are connected with each other,
(5.2) initializing residual to R 0 Y, index collection Y 0 = phi, where phi denotes an empty set;
(5.3) the first path, wherein
For a predetermined number of path passes, typically
The method comprises the following specific steps:
(1) computing a correlation matrix Γ l ＝Ψ H R l-1 ；
(2) Obtaining the index p corresponding to the maximum module value in the correlation matrix * ＝argmax p |[Γ l ] p L, p =1,2, \8230;, NS + N, where l · | denotes modulo;
(3) update index set y l ＝γ l-1 ∪p * Wherein, U represents union operation;
(4) updating orthogonal functions
Wherein the content of the first and second substances,
representing pseudo-inverse calculation;
(5) updating residual errors
(5.4) repeating step (5.4) until all iterations are traversed
The strip path is finished to obtain
The channel matrix estimate is
Has the advantages that: compared with the prior art, the technical scheme of the invention has the following beneficial technical effects:
(1) Compared with the existing mixed field beam scanning scheme, the method can greatly reduce the training overhead;
(2) The method can obtain equivalent speed performance, can approach the mixed field beam scanning performance at the full signal-to-noise ratio, is superior to the performance of the method in the literature [1] at the low signal-to-noise ratio, and is superior to the performance of the method in the literature [2] at the near field.
(3) The method can obtain a channel estimation result with considerable precision, and is superior to the estimation precision of the method in the literature [1 ].
(4) The digital combination in the method can be completed by off-line calculation before the beam training, thereby greatly reducing the calculated amount in the beam training process.
(5) The method can realize the combination of a plurality of numbers in parallel, thereby greatly improving the operation efficiency.
Drawings
FIG. 1 is a schematic diagram of a very large scale multiple input multiple output system model used in an embodiment of the invention;
FIG. 2 is a schematic diagram of a channel model in a very large scale multiple input multiple output system used by an embodiment of the present invention;
FIG. 3 is a comparison of the spectral efficiency of the present invention with documents [1], [2] and mixed field beam scanning methods;
FIG. 4 is a comparison of the beamforming gain of the present invention with documents [1], [2] and the mixed field beam scanning method;
figure 5 is a normalized mean square error comparison of the channel estimated by the method of the present invention and the document [1] with the actual channel.
Detailed Description
The technical solution of the present invention is further described in detail with reference to the accompanying drawings and embodiments.
(1) Basic parameters of the super-large-scale multi-input multi-output system used by the invention are as follows:
in the uplink beam training scene arranged between a base station and a user, the base station end antenna arrays are uniform linear arrays with the interval of half-wavelength, the number of the antennas is N, a partial connection hybrid combination structure is adopted, the structure comprises analog combination and digital combination, and the number of radio frequency links is N RF The antenna comprises N RF A plurality of non-overlapping sub-arrays; each sub-array has M = N/N RF The root antenna is connected with a radio frequency chain after analog combination; all N RF The radio frequency chains are connected to a digital processing unit for digital combination; the user terminal adopts a single antenna.
(2) The super-large-scale multi-input multi-output system model between the base station and the user is described as follows:
in the uplink beam training, the training symbol sent by the user terminal is x k K =1,2, \8230, where K is the length of the signal and the channel between the base station and the user is denoted as h, the received signal obtained after the mixing and combining of the signals at the base station is denoted as:
y k ＝v k W k hx k +v k W k η
wherein, W k Represents a simulated merge, v k Represents the digital combination, eta represents the additive white Gaussian noise, and the noise is satisfied
Mean is μ and variance is σ 2 Complex gaussian distribution of (a);
as shown in fig. 2, the channel model in the very large scale mimo system of the present invention is described as follows:
setting a multi-path channel composed of a main path and multiple sub-paths between a user and a base station, wherein N antennas of the base station are arranged along the y-axis of a Cartesian coordinate system, and the coordinates of the nth antenna are (0, delta) n λ), where N =1,2, \8230;, N,
λ represents the wavelength, and the coordinates of the center of the t-th sub-array are (0, Δ) t λ), wherein t =1,2, \ 8230;, N RF ，Δ t ＝[(2t-1)M-N]4, the user's coordinates are denoted p 1 ＝(r 1 cosθ 1 ,r 1 sinθ 1 ) Wherein r is 1 Representing the distance, theta, between the user and the origin of coordinates 1 ∈[-π/2,π/2]Representing the angle of the user with respect to the positive x-axis half-axis, and the coordinates of the scattering points in the ith path are denoted as p l ＝(r l cosθ l ,r l sinθ l ) Wherein l>2，r l Representing the distance, theta, between the user and the origin of coordinates l ∈[-π/2,π/2]Representing the angle, p, of the user with respect to the positive half-axis of the x-axis l And the nth antenna is represented as
Wherein, the first and the second end of the pipe are connected with each other,
is the sine of the angle, and Ω l ∈[-1,1](ii) a The channel between the base station and the user is modeled as follows:
wherein, L and g l Denotes the number of paths and the channel gain of the ith path, respectively, and α (·) denotes a channel steering vector, defined as:
the near and far fields are typically distinguished using rayleigh distances, expressed as
Wherein D = N λ/2 represents an antenna array aperture; when the distance between the radiation source and the base station exceeds Z, the wireless channel is defined as a far-field channel; conversely, the wireless channel is defined as a near-field channel;
when the distance r l >2D 2 At/λ, the following approximation is used,
wherein the steering vector of the far-field channel is defined as beta (N, omega) l )。
(3) The design method of the hybrid codebook for estimating the channel in the step (2) provided by the invention is as follows:
(3.1) setting C h Representing a hybrid codebook, C f Representing a far-field codebook, C n Representing a near-field codebook;
(3.2) far-field codebook C described in step (3.1) f Is represented as
(3.3) near-field codebook C described in step (3.1) n The code word in (1) is designed by the following steps:
(1) dividing the near field into N parts which are equal in angle dimension and S parts which are unequal in distance dimension;
(2) the nth quantization angle is theta n ＝(2n-1-N)/N；
(3) At the nth angle, the s quantization distance is
(4) Near field codebook C n Is shown as
Wherein, [ C ] n ] :,s ＝α(N,Θ n ,d n,s )；
(5) The hybrid codebook described in step (3.1) is represented as
(4) Based on the mixed codebook in the step (3), aiming at the line-of-sight path channel estimation, the two-stage beam training method of the super-large scale multi-input multi-output system provided by the invention comprises the following steps:
(4.1) for each subarray, the common beam training DFT codebook is
Wherein phi m ＝(2m-1-M)/M，m＝1,2,…,M；
(4.2) in the first phase of beam training, the user end sends training symbols to the base station, the training symbols last for M time slots, the base station receives the training symbols sequentially, and for the kth beam training, the received signals without digital combination are expressed as
Wherein the kth analog combination is represented as
blkdig {. Denotes a block diagonalization operation, so far, the first stage in the two-stage beam training scheme is completed;
(4.3) analog combining is designed in the first stage, and digital combining v is designed in the second stage by testing NS + N code words covered by the mixed codebook p P =1,2, \8230ns + N, representing the pth codeword in the mixed codebook as
Wherein, the first and the second end of the pipe are connected with each other,
representing a lower bound on an element value;
(4.4) for the p-th codeword, the sine of the angle of the quantization position corresponding to the p-th codeword with respect to the center of the t-th sub-array is
Wherein, delta t ＝[(2t-1)M-N]/4；
(4.5) setting
Wherein
(4.6) numerical incorporation of
Wherein the content of the first and second substances,
(4.7) digital combining of the received signals of the first stage
Is represented by
Wherein, the first and the second end of the pipe are connected with each other,
(4.8) comparing all NS + N combined signals, and selecting the quantization position corresponding to the signal with the maximum energy, wherein the quantization position is expressed as
(4.9) the finally selected most suitable channel codeword is
(5) And (5) performing two-stage beam training on the ultra-large scale multi-input multi-output system aiming at the multi-path channel estimation based on the first-stage beam training in the step (4).
(5.1) on the basis of the first stage, the analog combined signal is represented as
Definition of
Wherein the content of the first and second substances,
(5.2) initializing the residual to R 0 = Y, index set γ 0 = phi, where phi denotes an empty set;
(5.3) the first path, wherein
For a predetermined number of path passes, typically
The method comprises the following specific steps:
(1) computing a correlation matrix r l ＝Ψ H R l-1 ；
(2) Obtaining the index p corresponding to the maximum module value in the correlation matrix * ＝argmax p |[Γ l ] p L, p =1,2, \8230;, NS + N, where l |, denotes modulo;
(3) updating the index set gamma l ＝γ l-1 ∪p * Wherein, U represents union operation;
(4) updating orthogonal functions
Wherein the content of the first and second substances,
representing pseudo-inverse calculation;
(5) updating residual errors
(5.4) repeating step (5.4) until all steps are traversed
The strip path is finished to obtain
The channel matrix estimate is
The invention is further described below by combining simulation conditions and results:
considering a super-large-scale multi-input multi-output system configuration with N =256 antennas, the whole antenna array is divided into N RF =4 sub-arrays, each having M =64 antennas, the wavelength set λ =0.003M. The channel between the base station and the user is composed of 1 line-of-sight path and 2 non-line-of-sight paths, and the total number of paths L =3. The path gain beam gain conforms to complex Gaussian distribution
Channel angle omega of the l-th path l Conform to
Uniformly distributed therein. Hybrid codebook C h S =6. Document [1]]The method of (3) wherein the pilot length is set to 64.
(1) As shown in FIG. 3, the two-stage beam training method proposed by the present invention is compared with document [1]]Method of (2), mixed field beam scanning scheme and document]The spectral efficiency of the method of (1). The distance from the base station to the user or the scattering point is consistent with 5m,10m]Even distribution within. It can be seen that the mixed field beam scanning scheme can achieve higher performance than the other three approaches because the mixed field beam scanning scheme exhaustively enumerates the mixed codebook C h But the mixed field beam scan will use a much higher number of pilots than the other three schemes. Document [1]]The method of (2) performs the worst under low signal-to-noise ratio conditions, e.g., -10dB, because of the comparison with the other three methods [1]]The smaller the received signal-to-noise ratio of the method of (2), the performance of the scheme is obviously reduced under the condition of low signal-to-noise ratio. Document [2]]The method of (3) performs the worst under high signal-to-noise ratio conditions because the steering vector of the far field is gain attenuated in the near field model. However, the two-stage beam training method proposed by the present invention can approach the performance of mixed-field beam scanning schemes in low and high signal-to-noise environments.
(2) As shown in fig. 4, we compare the beamforming gains of the two-stage beam training method proposed by the present invention, the method of document [1], the mixed-field beam scanning scheme, and the method of document [2 ]. The distance of the base station to the user or scattering point conforms to a uniform distribution within [5m, r ], where r varies from 10m to 120 m. The signal-to-noise ratio is set to-5 dB. As can be seen in fig. 4, the mixed-field beam scanning scheme has the highest beam gain at all distances. As the distance decreases, severe beamforming gain attenuation occurs with the far-field beam scanning scheme, since this scheme only considers far-field channels. The other three schemes consider both near-field and far-field channels, so that they are robust in distance. In addition, the two-stage beam training proposed by the present invention can approach the performance of the mixed-field beam scanning scheme, with only minor beam-forming gain losses at short and long distances.
(3) As shown in FIG. 5, we compare the two-stage beam training method proposed by the present invention with the document [1]]The method of (3) estimates the normalized mean square error of the channel and the actual channel. Defining a mean square error of
Wherein the content of the first and second substances,
is a channel estimate, H is the actual channel,
representing the calculation of the two norms. In the NLOS model, the number of paths L =6. As can be seen from the figure, the following document [1]]Compared with the method, the method can obtain higher-precision channel estimation because the method adopts a partial connection structure, can obtain higher-dimensional signals by utilizing multiple radio frequency chains, and can obtain higher estimation precision under the same training cost by carrying out combined digital processing on the signals on the multiple radio frequency chains.
(4) The training overheads of the different schemes are compared. Hybrid field Beam scanning scheme, document [2]]Method of (1)]The training overhead of the method and the two-stage beam training method provided by the invention are respectively N (S + 1), N P And M. For example, under the simulation parameter settings of fig. 3 and 4, mix the field beam scanning scheme, document [2]]Method of (1)]The method of (3) and the two-stage beam training method proposed by the present invention use 1792, 256, 64 and 64, respectively, training pilots. In general, compared with the mixed field beam scanning scheme, the two-stage beam training method provided by the invention reduces the training overhead by 96.43%, and approaches the mixed field beam scanning scheme in performance.
Claims (6)
1. A two-stage beam training method for a very large scale array communication system is characterized by comprising the following steps:
(1) Setting basic parameters of a super-large-scale multi-input multi-output system;
(2) Constructing a super-large-scale multi-input multi-output system between a base station and a user and a channel model in the system;
(3) Designing a hybrid codebook for estimating the channel in step (2);
(4) Based on the mixed codebook in the step (3), aiming at the sight distance path channel estimation, and determining the most suitable guide vector of the channel, carrying out two-stage beam training on the super-large scale multi-input multi-output system;
(5) And (5) performing two-stage beam training on the ultra-large scale multi-input multi-output system for channel reconstruction aiming at multi-path channel estimation based on the first-stage beam training in the step (4).
2. The two-stage beam training method for very large scale array communication system according to claim 1, wherein in step (1), the method for setting basic parameters of the very large scale mimo system comprises the following steps: in the uplink wave beam training scene arranged between a base station and a user, the antenna arrays at the base station end are all uniform linear arrays with the interval of half wavelength, the number of the antennas is N, a partial connection mixed combination structure is adopted, the structure comprises analog combination and digital combination, the number of radio frequency links is N RF The antenna comprises N RF A plurality of non-overlapping sub-arrays; each sub-array has M = N/N RF The root antenna is connected with a radio frequency chain after analog combination; all N RF The radio frequency chains are connected to a digital processing unit for digital combination; the user terminal adopts a single antenna.
3. The two-stage beam training method for very large scale array communication system of claim 2, wherein in step (2), the method for constructing the very large scale mimo system model comprises the following steps:
(2.1) constructing a super-large-scale multi-input multi-output system model between a base station and a user
In the uplink beam training, the training symbol sent by the user terminal is x k K =1,2, \8230, where K is the length of the signal and the channel between the base station and the user is denoted as h, the received signal obtained after the mixing and combining of the signals at the base station is denoted as:
y k ＝v k W k hx k +v k W k η
wherein, W k Represents a simulated merge, v k Represents digital combination, eta represents additive white Gaussian noise, and noise satisfies
Mean is μ variance is σ 2 Complex gaussian distribution of (a);
(2.2) constructing a channel model in a very large-scale multiple-input multiple-output system
Setting a multi-path channel composed of a main path and multiple sub-paths between a user and a base station, wherein N antennas of the base station are arranged along the y-axis of a Cartesian coordinate system, and the coordinates of the nth antenna are (0, delta) n λ), where N =1,2, \ 8230;, N,
n-1)/4, wherein lambda represents the wavelength, and the coordinate of the center of the t-th sub-array is (0, delta) t λ), wherein t =1,2, \ 8230;, N RF ，Δ t ＝[(2t-1)M-N]4, the user's coordinates are denoted p 1 ＝(r 1 cosθ 1 ,r 1 sinθ 1 ) Wherein r is 1 Representing the distance between the user and the origin of coordinates, theta 1 ∈[-π/2,π/2]Representing the angle of the user with respect to the positive x-axis half-axis, and the coordinates of the scattering points in the ith path are denoted as p l ＝(r l cosθ l ,r l sinθ l ) Wherein l>2，r l Representing the distance between the user and the origin of coordinates, theta l ∈[-π/2,π/2]Representing the angle of the user, p, with respect to the positive x-axis half l And the nth antenna is expressed as
Wherein the content of the first and second substances,
is the sine of the angle, and Ω l ∈[-1,1](ii) a The channel between the base station and the user is modeled as follows:
wherein, L and g l Denotes the number of paths and the channel gain of the ith path, respectively, and α (·) denotes a channel steering vector, defined as:
the near and far fields are typically distinguished using rayleigh distances, expressed as
Wherein D = N λ/2 denotes an antenna array aperture; when the distance between the radiation source and the base station exceeds Z, the wireless channel is defined as a far-field channel; conversely, the wireless channel is defined as a near-field channel;
when the distance r l >2D 2 At/λ, the following approximation is used,
wherein the content of the first and second substances,the steering vector of the far-field channel is defined as beta (N, omega) l )。
4. The two-stage beam training method of very large scale array communication system of claim 3, wherein in step (3), the method for designing the hybrid codebook for estimating the channel in step (2) is as follows:
(3.1) setting C h Representing a hybrid codebook, C f Representing a far-field codebook, C n Representing a near-field codebook;
(3.2) far-field codebook C described in step (3.1) f The nth codeword in (1) is represented as
(3.3) near-field codebook C described in step (3.1) n The code word in (1) is designed by the following steps:
(1) dividing the near field into N parts which are equal in angle dimension and S parts which are unequal in distance dimension;
(2) the nth quantization angle is theta n ＝(2n-1-N)/N；
(3) At the nth angle, the s quantization distance is
(4) Near field codebook C n Is shown as
Wherein, [ C ] n ] :,s ＝α(N,Θ n ,d n,s )；
(5) The hybrid codebook described in step (3.1) is represented as
5. The two-stage beam training method of the very large scale array communication system of claim 4, wherein in step (4), based on the hybrid codebook of step (3), aiming at the line-of-sight path channel estimation, the two-stage beam training is performed on the very large scale multiple input multiple output system, and the method comprises the following steps:
(4.1) for each subarray, the common beam training DFT codebook is
Wherein phi m ＝(2m-1-M)/M，m＝1,2,…,M；
(4.2) in the first phase of beam training, the user end sends training symbols to the base station, the training symbols last for M time slots, the base station receives the training symbols sequentially, and for the kth beam training, the received signals without digital combination are expressed as
Wherein the kth analog combination is represented as
blkdig {. Denotes a block diagonalization operation, so far, the first stage in the two-stage beam training scheme is completed;
(4.3) the first stage designs analog combining, and the second stage designs digital combining v by testing NS + N code words covered by the mixed codebook p P =1,2, \ 8230n, NS + N, representing the p-th codeword in the mixed codebook as
Wherein the content of the first and second substances,
representing a lower bound on an element value;
(4.4) for the p-th codeword, the sine of the angle of the corresponding quantization position with respect to the center of the t-th sub-array is
Wherein, delta t ＝[(2t-1)M-N]/4；
(4.5) setting
Wherein
(4.6) numerical merger represented as
Wherein the content of the first and second substances,
(4.7) digital combining of the received signals of the first stage
After is shown as
Wherein, the first and the second end of the pipe are connected with each other,
(4.8) comparing all NS + N combined signals, and selecting the quantization position corresponding to the signal with the maximum energy, wherein the quantization position is expressed as
(4.9) the codeword of the most suitable channel finally selected is
6. The two-stage beam training method of very large scale array communication system according to claim 4, wherein in step (5), based on the first stage beam training of step (4), aiming at the multipath channel estimation, the two-stage beam training is performed on the very large scale mimo system, and the method is as follows:
(5.1) on the basis of the first stage, the analog combined signal is represented as
Definition of
Wherein the content of the first and second substances,
(5.2) initializing the residual to R 0 = Y, index set γ 0 = phi, where phi denotes an empty set;
(5.3) the first path, wherein
For a predetermined number of path passes, typically
The method comprises the following specific steps:
(1) computing a correlation matrix r l ＝Ψ H R l-1 ；
(2) Obtaining the index p corresponding to the maximum module value in the correlation matrix * ＝argmax p |[Γ l ] p L, p =1,2, \8230;, NS + N, where l |, denotes modulo;
(3) updating the index set gamma l ＝γ l-1 ∪p * Wherein, U represents union operation;
(4) updating orthogonal functions
Wherein, the first and the second end of the pipe are connected with each other,
representing pseudo-inverse calculation;
(5) updating residual errors
(5.4) repeating step (5.4) until all steps are traversed
The strip path is finished to obtain
The channel matrix estimate is
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210597131.7A CN115208442A (en) | 2022-05-30 | 2022-05-30 | Two-stage beam training method of ultra-large scale array communication system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210597131.7A CN115208442A (en) | 2022-05-30 | 2022-05-30 | Two-stage beam training method of ultra-large scale array communication system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115208442A true CN115208442A (en) | 2022-10-18 |
Family
ID=83575498
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210597131.7A Pending CN115208442A (en) | 2022-05-30 | 2022-05-30 | Two-stage beam training method of ultra-large scale array communication system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115208442A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115996392A (en) * | 2023-03-22 | 2023-04-21 | 南京邮电大学 | Orthogonal position design method in near field ultra-large scale planar array communication system |
| CN117335848A (en) * | 2023-11-06 | 2024-01-02 | 国家工业信息安全发展研究中心 | Beam training method for ultra-large-scale MIMO space non-stationary channel |
-
2022
- 2022-05-30 CN CN202210597131.7A patent/CN115208442A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115996392A (en) * | 2023-03-22 | 2023-04-21 | 南京邮电大学 | Orthogonal position design method in near field ultra-large scale planar array communication system |
| CN115996392B (en) * | 2023-03-22 | 2023-08-25 | 南京邮电大学 | Orthogonal position design method in near field ultra-large scale planar array communication system |
| CN117335848A (en) * | 2023-11-06 | 2024-01-02 | 国家工业信息安全发展研究中心 | Beam training method for ultra-large-scale MIMO space non-stationary channel |
| CN117335848B (en) * | 2023-11-06 | 2024-04-16 | 国家工业信息安全发展研究中心 | Beam training method for ultra-large-scale MIMO space non-stationary channel |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3577776B1 (en) | Transmitter and method for transmitting symbols over wireless communication channel | |
| CN107453795B (en) | Beam allocation method of multi-user millimeter wave communication system, device and system thereof | |
| CN115208442A (en) | Two-stage beam training method of ultra-large scale array communication system | |
| CN107294590B (en) | Digital-analog hybrid beam forming method based on uplink training | |
| CN111049557B (en) | Millimeter wave MIMO system hybrid precoding method based on statistical channel information | |
| CN109327918B (en) | Low-overhead FDD large-scale MIMO downlink channel reconstruction method | |
| CN110138425B (en) | Low-complexity array antenna multi-input multi-output system hybrid precoding algorithm | |
| CN108418617B (en) | Large-scale MIMO system verification method based on multiple sub-antenna arrays | |
| CN105790811A (en) | High-resolution analog beam rapid training method and device | |
| CN105891771A (en) | Continuous distribution-based angle estimation method and device for improving estimation precision | |
| CN104698430A (en) | Virtual array antenna based angle estimation method for improving accuracy | |
| CN107171985B (en) | Channel estimation method and system | |
| CN114285702A (en) | Sparse cascade channel estimation method for millimeter wave IRS (inter-Range instrumentation System) cooperation system | |
| CN105743559A (en) | Hybrid beam-forming and space-time coding multi-user downlink transmission method in Massive MIMO (Multiple Input Multiple Output) system | |
| CN114157331A (en) | Large-scale MIMO channel state information feedback method based on pseudo-complex value input | |
| CN112769462B (en) | Millimeter wave MIMO broadband channel estimation method based on joint parameter learning | |
| CN107171705B (en) | Joint analog beam and user scheduling method in digital-analog hybrid communication | |
| Zou et al. | Analog beamforming for millimeter-wave mimo systems via stochastic optimization | |
| CN109951219B (en) | Low-cost large-scale non-orthogonal multi-access method | |
| Wang et al. | Antenna Selection Strategies for Massive MIMO Systems with Limited-Resolution ADCs/DACs | |
| Lee et al. | Efficient channel AoD/AoA estimation using widebeams for millimeter wave MIMO systems | |
| You et al. | Bayesian matching pursuit based estimation of off-grid channel for millimeter wave massive MIMO system | |
| Tan et al. | Hybrid precoding codebook design in millimetre‐wave massive MIMO systems with low‐resolution phase shifters | |
| Chiang et al. | Hybrid beamforming strategy for wideband millimeter wave channel models | |
| CN112235022B (en) | Low-complexity large-scale MIMO low-orbit satellite precoding method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |