CN115801071A - Pilot pollution elimination method assisted by structural tensor decomposition for unmanned aerial vehicle communication - Google Patents
Pilot pollution elimination method assisted by structural tensor decomposition for unmanned aerial vehicle communication Download PDFInfo
- Publication number
- CN115801071A CN115801071A CN202211475966.1A CN202211475966A CN115801071A CN 115801071 A CN115801071 A CN 115801071A CN 202211475966 A CN202211475966 A CN 202211475966A CN 115801071 A CN115801071 A CN 115801071A
- Authority
- CN
- China
- Prior art keywords
- pilot
- channel
- matrix
- uavs
- pollution
- 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
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 33
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000004891 communication Methods 0.000 title claims abstract description 18
- 230000008030 elimination Effects 0.000 title claims abstract description 14
- 238000003379 elimination reaction Methods 0.000 title claims abstract description 14
- 239000011159 matrix material Substances 0.000 claims abstract description 43
- 230000005540 biological transmission Effects 0.000 claims abstract description 16
- 230000001413 cellular effect Effects 0.000 claims description 10
- 238000001228 spectrum Methods 0.000 claims description 6
- 239000000126 substance Substances 0.000 claims description 6
- 230000002452 interceptive effect Effects 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 239000000654 additive Substances 0.000 claims description 3
- 230000000996 additive effect Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 239000007788 liquid Substances 0.000 claims description 3
- 239000000969 carrier Substances 0.000 claims description 2
- 230000021615 conjugation Effects 0.000 claims description 2
- 239000013256 coordination polymer Substances 0.000 claims 4
- 238000004364 calculation method Methods 0.000 abstract description 6
- 238000012545 processing Methods 0.000 abstract description 2
- 108091006146 Channels Proteins 0.000 description 72
- 238000007430 reference method Methods 0.000 description 6
- 230000006978 adaptation Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Images
Classifications
-
- 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
Landscapes
- Radio Transmission System (AREA)
Abstract
The invention discloses a pilot pollution elimination method assisted by structural tensor decomposition for unmanned aerial vehicle communication. The present invention designs a two-phase pilot transmission scheme and intentionally introduces pilot pollution for pilot pollution cancellation. The method comprises the following steps that a base station expresses received pilot signals as second-order tensors, so that the problem of pilot pollution elimination is converted into a problem of joint channel parameter estimation, and a structural CANDECOMP/PARAFAC decomposition auxiliary (SCPD) method is used for estimating channel parameters by utilizing a Vandermonde structure of a factor matrix, so that a channel is reconstructed and pilot pollution is eliminated. The method only utilizes standard linear algebra, avoids a large amount of iteration, improves the channel estimation precision, reduces the calculation complexity, and has very important significance for processing the pilot frequency pollution problem in a mobile scene, so the method has certain practical value.
Description
Technical Field
The invention relates to the technical field of unmanned aerial vehicle communication in a non-cellular large-scale MIMO system, in particular to a structured CANDICOMP/PARAFAC decomposition-assisted (SCPD) pilot pollution elimination (PDC) scheme for unmanned aerial vehicle communication in the non-cellular large-scale MIMO system.
Background
Unmanned Aerial Vehicles (UAVs) have attracted considerable attention in professional applications, and are considered to be a promising wireless technology after the 5G era. Cellless massive Multiple Input Multiple Output (MIMO) combines the concepts of distributed MIMO and massive MIMO, without sharing instantaneous channel state information among numerous Access Points (APs), with all channel estimation performed in local APs, reducing transmission delay. Because direct path (LoS) propagation is adopted between the unmanned aerial vehicle and the AP, the unmanned aerial vehicle communication has unique sparse channel characteristics. Furthermore, for safety, the drone communication should have a low latency. Therefore, the cellular-free massive MIMO system is considered as a perfect adaptation scheme for drone communication. However, due to the strong LoS channel of UAV-AP (so-called air-to-ground), the remote drones still suffer from pilot Pollution (PC), and therefore the pilot pollution caused by pilot reuse will inevitably become more severe. Therefore, an effective scheme needs to be designed to eliminate the influence caused by pilot pollution, so that the utilization rate of pilot resources for unmanned aerial vehicle communication is improved.
Disclosure of Invention
The technical problem is as follows: in view of the above, an object of the present invention is to provide a pilot pollution cancellation (PDC) scheme for structural CANDICOMP/parafacc (CP) decomposition assistance (SCPD) for drone communication in a cellular-free massive MIMO system, so as to solve the problem of pilot pollution caused by a strong LoS channel in the drone communication. The invention designs a two-stage pilot frequency transmission scheme, deliberately introduces pilot frequency pollution, expresses signals received by a base station as second-order tensor, and provides an SCPD channel estimation scheme for eliminating the pilot frequency pollution. In addition, the superiority of the scheme in terms of accuracy and complexity is demonstrated by calculating the normalized mean square error of the estimated channel parameters.
The technical scheme is as follows: in order to achieve the above object, the pilot pollution elimination method assisted by structured tensor decomposition for unmanned aerial vehicle communication of the present invention is as follows:
in a large-scale MIMO system without honeycomb, U single-antenna Unmanned Aerial Vehicles (UAVs) distributed randomly and R geographically dispersed Access Points (APs) are provided, and each access point adopts Uniform Circular Array (UCA) and is provided with N R The distance between the antennas is half wavelength; is provided with K 0 The OFDM system comprises a plurality of orthogonal frequency division multiple access (OFDM) subcarriers, wherein the first K subcarriers are used for channel parameter estimation; because the unmanned aerial vehicle flies higher than the access point, and in practice, the channel between the unmanned aerial vehicle and the access point is mainly a direct path LoS link, and the channel between the unmanned aerial vehicle and the access point is LoS; the scheme specifically comprises the following steps:
step 1, establishing a channel model of a non-cellular large-scale MIMO system to obtain an LS channel estimation of minimum mean square and an expression of pilot pollution;
step 2, firstly, a two-stage pilot frequency transmission scheme is adopted for channel estimation, then, an AP converts the problem of pilot frequency pollution elimination into the problem of channel parameter estimation according to received pilot frequency signals, a second-order tensor kappa can be obtained according to the pilot frequencies received by all subcarriers, time delay, azimuth angle AoA, elevation angle AoE and complex gain parameters of all L UAV-AP paths are obtained through tensor CANDECOMP/PARAFAC decomposition, and finally, an expression of the signals is reconstructed based on the estimated channel parameters to eliminate the influence of the pilot frequency pollution.
Wherein the content of the first and second substances,
the step S1 specifically includes:
step 101: when each access point serves L drones simultaneously, the channel between drone u and access point r on subcarrier k is represented as:
in the formula (1), K 0 Is the total number of OFDM sub-carriers, f s Representing the sampling rate, theta u,r 、φ u,r 、α u,r And τ u,r Respectively representing AoE, aoA, channel complex gain and time delay between the unmanned aerial vehicle u and the access point r, wherein different channel paths have different channel parameters; a (θ, φ) represents a steering vector, whose nth element is represented as:
step 102: the set of interfering UAVs if sharing the same pilot sequence as the UAVu is denoted as
With mth pilot phi m For example, the pilot signal Φ transmitted by the UAV received by the AP m Associated with the conjugate of the pilot sequence it transmits, the AP can derive a least mean square based channel estimate
In the formula (3), τ p Indicating the length of the pilot signal, p p Representing the power of the pilot signal, N r,k Representing additive white gaussian noise, (.) T Represents a transpose operation, (.) * Denotes a conjugation operation, h u',r.k Representing the channel between the drone u' on subcarrier k and the access point r, the pilot signal transmitted by the UAV is satisfied
Then equation (3) can be rewritten as:
in the formula (4), the first and second groups,
and in part, pilot pollution.
The step 2 specifically comprises:
step 201: regarding the UAVs connected to the same AP as a group, in the first phase of pilot transmission, UAVs in the same group share the same pilot, and UAVs in different groups share different pilot sequences; if the L-shelf UAVs are served by the same AP, the modified equation (4) is:
the L-shelf UAVs in equation (5) are users served by the APr, and assuming (5) is treated as a single-user multipath channel, the pilot pollution cancellation problem can be converted into a joint channel estimation problem for the purpose of combining with that in equation (4)
Distinguishing, for estimating the channel in equation (5)
Represents; obtaining L UAV-AP paths through tensor CP decomposition, considering all subcarriers, obtaining a second order tensor k, wherein the two modes are the antenna number and the subcarrier number of the AP respectively, and obtaining through CP decomposition:
in the formula (6), the first and second groups,
four channel parameters can be obtained by performing CP decomposition on the tensor k;
step 202: the CP decomposition of the tensor k estimates four channel parameters,
(1) And (3) time delay estimation: estimating time delay in the first pilot frequency transmission phase, and defining factor matrix A = [ a (theta) = 1,r ,φ 1,r ),...,a(θ L,r ,φ L,r )]，B＝[α 1,r q(τ 1,r ),...,α L,r q(τ L,r )]Where B is a column weighted Vandermonde matrix, which is generated as
First, a 1-mode expansion is performed on the tensor κ to obtain:
then, singular value expansion (SVD) is carried out on the SVD to obtain:
wherein
(·) H Represents a conjugate transpose operation; since a, B columns are full-rank, there is a non-singular matrix M such that UM = B, the vandermonde structure of B is applied,
wherein Z = diag ([ Z) 1,r ,...,z L,r ])，
And B represents the B matrix with the top row and the bottom row removed respectively, and the Van der Waals characteristic of B is introduced to obtain: u shape 1 M = B and
wherein, U 1 ＝U，
Further, it can be obtained:
wherein, the first and the second end of the pipe are connected with each other,
a pseudo-inverse operation is shown as being performed,
due to B, U 1 And U 2 Rank full, generator
The time delay can be obtained by decomposing the characteristic value, and then the time delay can be extracted as follows:
wherein the content of the first and second substances,
represents the phase angle extraction operation and, therefore,
makes a permutation ambiguity, known as
Can be explicitly expressed as
Thus, the relationship between the true factor matrix and the estimated factor matrix is:
wherein pi is a permutation matrix, { Λ 1 ,Λ 2 Is a scaling matrix, satisfies Λ 1 Λ 2 ＝I，{E 1 ,E 2 Is the estimation error matrix;
(2) AoE, aoA estimation
In the second phase of pilot transmission, it is associated with the same APAnd sending different pilot signals by the connected UAVs, and estimating AoEs and AoAs of the L-frame UAVs in the target set by adopting a MUSIC algorithm. Rewriting AP in equation (3) r Received pilot signal on the k sub-carrier
Comprises the following steps:
wherein X represents the pilot matrix of the L-frame UAVs, N r,k Representing a noise matrix, the correlation matrix of the received pilot signal is represented as:
in the formula (12), the first and second groups of the chemical reaction are shown in the specification,
indicating to take expectation;
to R is S Performing eigenvalue decomposition to obtain a matrix Q = [ Q = [) x |Q n ]，Q x Is listed as the corresponding characteristic value
Characteristic vector of (2), Q n For the noise subspace, a spatial spectrum is defined, since the steering vector a (θ, φ) in the signal subspace is orthogonal to the noise subspace
AoEs and AoAs of UAVs
Will pass through the spatial spectrum P music The L pole of (theta, phi) is estimated;
(3) Permutation ambiguity and complex gain estimation
In practice, the amount of the liquid to be used,
and
the correspondence between them is still unknown, i.e., the arrangement is ambiguous, note that in equations (9), (10),
and
there is another permutation ambiguity between them, i.e.
Having the same permutation matrix Π, using a correlation-based strategy:
wherein, the first and the second end of the pipe are connected with each other,
the AoEs/AoAs will be rearranged according to the time delay so that the permutation ambiguity can be resolved and then the estimated complex gain can be obtained
Step 203: based on estimated channel parameters
And reconstructing the needed L UAV-AP channels.
Has the advantages that: the invention researches the problem of pilot frequency pollution elimination in unmanned aerial vehicle communication of the non-cellular large-scale MIMO system, and converts the problem of pilot frequency pollution elimination into the problem of channel parameter estimation by utilizing the characteristic that a large number of direct paths exist in a channel between an unmanned aerial vehicle and an access point. The channel is expressed as a second-order tensor, and 4 characteristic parameters of the channel are estimated by adopting a tensor CANDECOMP/PARAFAC decomposition algorithm: the time delay, the azimuth angle, the elevation angle and the channel complex gain are further reduced, the channel between the unmanned aerial vehicle and the access point is further restored, the traditional pilot pollution elimination scheme needs to be iterated continuously to eliminate the pilot pollution, the calculation complexity is high, and the cost is large. The method provided by the invention only needs to carry out CANDECOMP/PARAFAC decomposition on the tensor, thereby avoiding multiple iterative operations when the channel between the unmanned aerial vehicle and the AP is recovered and pilot pollution is eliminated, and having low calculation complexity and high calculation speed. Meanwhile, the scheme is not limited to the non-cellular large-scale MIMO architecture adopted by the invention, can be used in any communication systems such as a cellular system and a large-scale MIMO system, and has universality and wide application range.
Drawings
Fig. 1, 2, 3 and 4 are graphs showing the relationship between the normalized mean square error NMSE and SNR for 4 channel parameter delays, aoE, aoA and channel complex gains, respectively. Wherein the normalized mean square error represents an error between the true channel parameter and the estimated channel parameter;
figure 5 is a graph comparing the normalized mean square error, NMSE, versus SNR for a reconstructed channel using different pilot pollution cancellation schemes. The method for removing pilot frequency pollution through continuous iteration is regarded as a reference method, LS represents a minimum mean square channel estimation method, ALS represents a higher-level channel estimation method based on tensor CP decomposition, and 'deployed' represents an SCPD channel estimation method;
fig. 6 shows histograms of the CPU time comparisons for channel estimation by LS, ALS, baseline method and SCPD method.
Detailed Description
The present invention is described in detail below with reference to examples:
assume a cellular-free massive MIMO scenario in a suburban scenario with R =20 APs and U =80 UAVs. Each AP is equipped with N R =64 antennas and is associated with L =4 shelf single-antenna UAVs. The UAVs and APs are randomly distributed within a circle with a radius of 0.500 (km). AoE theta, aoA phi and time delay tau are respectively and uniformly distributed in [ -pi, pi [ -pi [ ]]、
And [0,100]ns. Sampling frequency f s Set to 0.32GHz. The pilot length is 10 in the two-phase pilot transmission strategy. For ease of comparison, the pilot length in the proposed SCPD algorithm is the same as for the other strategies.
Based on the above-mentioned large-scale MIMO scenario without cell, the method for eliminating pilot pollution provided in this embodiment specifically includes the following steps:
step 1: establishing a channel model of a non-cellular massive MIMO system, and obtaining an expression of least mean square (LS) channel estimation and pilot pollution:
in this embodiment, step 1 specifically includes:
step 101: assuming each AP simultaneously serves L-racks UAVu and AP on sub-carrier k r The channels in between are represented as:
in the formula (1), f s Representing the sampling rate, theta u,r 、φ u,r 、α u,r And τ u,r Representing AoE, aoA, complex gain and delay between UAVu and APr, respectively. a (θ, φ) represents a steering vector whose nth element is represented as:
[a(θ,φ)] n ＝exp{-jπsin(θ)cos(φ-γ n )}； (2)
in the formula (2), the first and second groups of the compound,
in practical systems, it is reasonable to have different paths for different UAV-AP channels, and therefore, assuming that there are different paths between all UAVs and all APs, the elevation angle (AoEs), azimuth angle (AoAs), and delay for each path are also different.
Step 102: for no loss of generality, the mth pilot phi of the kth subcarrier m For example, pilot signals transmitted by UAV satisfy
The set of interfering UAVs that are assumed to share the same pilot sequence as the UAVu is denoted as
After associating the pilot signal transmitted by UAV received by AP with the conjugate of pilot sequence transmitted by AP, AP transmits the pilot sequence r LS-based channel estimates can be obtained:
in the formula (3), ρ p Representing the power of the pilot signal, N r,k Representing additive white gaussian noise. Then equation (3) can be rewritten as:
in equation (4), the second part represents pilot pollution, and the accuracy of channel estimation is significantly reduced because the strong LoS interference generated by other UAVs cannot be ignored.
Step 2: first, a two-stage pilot transmission scheme is employed for channel estimation. And then, the AP converts the PDC problem into a channel parameter estimation problem according to the received pilot signals, obtains a second-order tensor kappa according to the pilot signals received by all subcarriers, and estimates and obtains the time delay, the AOA, the AOE and the complex gain parameters of all L UAV-AP paths by carrying out tensor CP decomposition on the second-order tensor kappa. And finally, reconstructing an expression of the signal based on the estimated channel parameters, and eliminating the influence of pilot frequency pollution.
In this embodiment, the step 2 specifically includes:
step 201: firstly, in the first stage of pilot frequency transmission, UAVs connected with the same AP are regarded as one group, UAVs in the same group share the same pilot frequency, and UAVs in different groups share different pilot frequency sequences; assuming that L UAVs are served by the same AP, the modified equation (4) is:
UAVs in equation (5) are APs r Regarding the target user (5) as a single-user multipath channel. Therefore, considering the PDC problem to be converted into the joint channel estimation problem, L UAV-AP paths are obtained through tensor CP decomposition.
Considering all subcarriers, a second order tensor can be obtained
The two modes are the number of antennas and the number of subcarriers of the AP. The CP decomposition is carried out on the obtained product to obtain:
in the formula (6), the first and second groups,
due to the characteristic of sparse scattering, the four-channel-parameter estimation method has an inherent low-rank structure, and four channel parameters can be estimated by performing CP decomposition on tensor kappa
Step 202: the tensor κ is CP decomposed to estimate four channel parameters.
(1) And (3) time delay estimation: estimating time delay in the first pilot frequency transmission phase, and defining factor matrix A = [ a (theta) = 1,r ,φ 1,r ),...,a(θ L,r ,φ L,r )]，B＝[α 1,r q(τ 1,r ),...,α L,r q(τ L,r )]Where B is a column weighted Vandermonde matrix, which is generated as
First, a 1-mode expansion is performed on the tensor κ to obtain:
then, singular value expansion (SVD) is performed on it, resulting in:
since the columns a and B are full-rank, the presence of the non-singular matrix M makes UM = B, and with the vandermonde structure of B, there is a
Wherein Z = diag ([ Z) 1,r ,...,z L,r ])，
And B denotes a B matrix with the top and bottom rows removed, respectively. Using the vandermonde property in UM = B, we obtained: u shape 1 M = B and
wherein, U 1 ＝U，
Further, it is possible to obtain:
due to B, U 1 And U 2 The column is full of rank and is,generator
Can be obtained by characteristic value decomposition. Further, the time delay can be extracted as follows:
wherein the content of the first and second substances,
a phase angle extraction operation is shown. Therefore, the temperature of the molten metal is controlled,
a permutation ambiguity is generated, known as
Can be displayed as
Thus, the relationship between the true factor matrix and the estimated factor matrix is:
wherein pi is a permutation matrix, { Λ 1 ,Λ 2 Is a scaling matrix, satisfies Λ 1 Λ 2 ＝I，{E 1 ,E 2 Is the estimation error matrix.
(2) (AoE, aoA) estimation
In the second phase of pilot transmission, UAVs connected to the same AP transmit different pilot signals. The MUSIC algorithm is adopted to estimate AoEs and AoAs of L-frame UAVs in the target set. Rewriting the pilot signal received by the APr is:
wherein, X represents the pilot matrix of L-frame UAVs, and the correlation matrix of the received pilot signals is represented as:
in the formula (12), the first and second groups,
indicating that it is desired.
To R S Performing eigenvalue decomposition to obtain a matrix Q = [ Q = [) x |Q n ]，Q x Is listed as the corresponding characteristic value
Characteristic vector of (2), Q n For the noise subspace, a spatial spectrum is defined, since the steering vector a (θ, φ) in the signal subspace is orthogonal to the noise subspace
AoEs and AoAs of UAVs
Can pass through the spatial spectrum P music The L pole of (theta, phi) is estimated.
(3) Permutation ambiguity and complex gain estimation
In practice, the amount of the liquid to be used,
and
the correspondence between them is still unknown, i.e. the arrangement is ambiguous. In the equations (9), (10),
and
there is another permutation ambiguity between them, i.e.
There is the same array matrix Π. With a correlation-based strategy:
wherein the content of the first and second substances,
the AoEs/AoAs will be rearranged according to the time delay so that the permutation ambiguity can be resolved. The estimated complex gain can then be obtained
Step 203: and reconstructing the needed L UAV-AP channels according to the estimated channel parameter set.
The whole process of conducting pilot pollution elimination based on the structured CP decomposition assistance in the cellular-free massive MIMO system unmanned aerial vehicle communication by using the method provided by the embodiment is demonstrated above.
Fig. 1-4 show the normalized mean square error NMSE versus SNR for 4 channel parameters delay, aoE, aoA, and channel complex gain, respectively. Where the normalized mean square error represents the error between the true channel parameter and the estimated channel parameter. It can be seen that the SCPD scheme can accurately estimate the channel parameters.
Fig. 5 is a graph showing a comparison of normalized mean square error NMSE versus SNR for a reconstructed channel using different pilot pollution cancellation schemes. It should be noted that the accuracy of the reconstructed channel still needs to be checked due to the problems of CP decomposition and permutation ambiguity introduced by the MISIC algorithm. The method for removing pilot frequency pollution by continuous iteration is taken as a reference method. Comparing LS, a more advanced channel estimation method ALS based on tensor CP decomposition, and pilot pollution elimination performance of the reference method with the SCPD method provided by this embodiment, it can be found that the SCPD scheme has better channel estimation accuracy. This result is generated because LS does not fully utilize angle information, and the reference method, although removing pilot pollution to some extent, still has estimation errors of path loss and phase rotation, and thus the performance of the reference method is inferior to SCPD. The ALS method requires random initialization and thus does not perform as well as the SCPD, but is higher than the reference method, which indicates that tensor processing has an advantage in channel estimation.
Fig. 6 shows the CPU time for different strategies. Since the computational complexity of the LS is lowest, the time it takes is shortest. Since the baseline scheme requires successive iterations to remove the interfering channel, while the ALS requires a factor matrix of the tensor to be restored by iteration, the SCPD is shorter than the CPU time of the baseline and ALS schemes.
In summary, the present invention provides a structured pilot pollution cancellation scheme assisted by CANDICOMP/parafacc decomposition for solving the pilot pollution problem caused by a strong LoS channel in the unmanned aerial vehicle communication, aiming at the pilot pollution cancellation problem of unmanned aerial vehicle communication in the cellular massive MIMO system. In addition, iterative calculation is avoided by decomposing and estimating the parameters of the channel through tensor CANDICOMP/PARAFAC, the calculation complexity is reduced, meanwhile, the precision of parameter estimation is improved, and the method has practical significance.
The invention is not described in detail, but is well known to those skilled in the art.
The foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations could be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions available to those skilled in the art through logic analysis, reasoning and limited experiments based on the prior art according to the concept of the present invention should be within the scope of protection defined by the claims.
Claims (3)
1. A pilot pollution elimination method assisted by structural tensor decomposition of unmanned aerial vehicle communication is characterized by comprising the following steps: in a large-scale multi-input multi-output MIMO system without cellular, a total of U randomly distributed single-antenna Unmanned Aerial Vehicles (UAVs) and R geographically dispersed Access Points (APs) are provided, and each access point adopts Uniform Circular Array (UCA) and is provided with N R The distance between the antennas is half wavelength; is provided with K 0 The OFDM system comprises a plurality of orthogonal frequency division multiple access (OFDM) subcarriers, wherein the first K subcarriers are used for channel parameter estimation; because the unmanned aerial vehicle flies higher than the access point, and in practice, the channel between the unmanned aerial vehicle and the access point is mainly a direct path LoS link, and the channel between the unmanned aerial vehicle and the access point is LoS; the scheme specifically comprises the following steps:
step 1, establishing a channel model of a non-cellular large-scale MIMO system to obtain LS channel estimation of the least mean square and an expression of pilot pollution;
step 2, firstly, a two-stage pilot frequency transmission scheme is adopted for channel estimation, then, an AP converts the problem of pilot frequency pollution elimination into the problem of channel parameter estimation according to received pilot frequency signals, a second-order tensor kappa can be obtained according to the pilot frequencies received by all subcarriers, time delay, azimuth angle AoA, elevation angle AoE and complex gain parameters of all L UAV-AP paths are obtained through tensor CANDECOMP/PARAFAC decomposition, and finally, an expression of the signals is reconstructed based on the estimated channel parameters to eliminate the influence of the pilot frequency pollution.
2. The method according to claim 1, wherein the step S1 specifically includes:
step 101: when each access point serves L drones simultaneously, the channel between drone u and access point r on subcarrier k is represented as:
in the formula (1), K 0 Is the total number of OFDM sub-carriers, f s Representing the sampling rate, theta u,r 、φ u,r 、α u,r And τ u,r Respectively representing AoE, aoA, channel complex gain and time delay between the unmanned aerial vehicle u and the access point r, wherein different channel paths have different channel parameters; a (θ, φ) represents a steering vector, whose nth element is represented as:
step 102: the set of interfering UAVs that share the same pilot sequence as the UAVu is denoted as I u With the mth pilot phi m For example, the pilot signal Φ transmitted by the UAV received by the AP m Associated with the conjugate of its transmitted pilot sequence, the AP can derive a least mean square based channel estimate
In the formula (3), τ p Indicating the length of the pilot signal, p p Representing the power of the pilot signal, N r,k Representing additive white gaussian noise, (.) T Represents a transpose operation, (. Cndot.) * Denotes the conjugation operation, h u',r.k Representing the channel between the drone u' on the subcarrier k and the access point r, the pilot signal transmitted by the UAV satisfying
Then equation (3) can be rewritten as:
in the formula (4), the first and second groups,
and in part, pilot pollution.
3. The method according to claim 1, wherein the step 2 specifically includes:
step 201: regarding the UAVs connected to the same AP as a group, in the first phase of pilot transmission, UAVs in the same group share the same pilot, and UAVs in different groups share different pilot sequences; if the L-shelf UAVs are served by the same AP, the modified equation (4) is:
the L-shelf UAVs in equation (5) are users served by the APr, and assuming (5) is treated as a single-user multipath channel, the pilot pollution cancellation problem can be converted into a joint channel estimation problem for the purpose of combining with that in equation (4)
Distinguishing, for estimating the channel in equation (5)
Represents; obtaining L UAV-AP paths through tensor CP decomposition, considering all subcarriers, obtaining a second order tensor kappa, wherein the two modes are the antenna number and the subcarrier number of the AP respectively, and obtaining the following through CP decomposition:
in the formula (6), the first and second groups,
four channel parameters can be obtained by performing CP decomposition on the tensor k;
step 202: the CP decomposition of the tensor k estimates four channel parameters,
(1) And (3) time delay estimation: estimating time delay in the first pilot frequency transmission phase, and defining factor matrix A = [ a (theta) = 1,r ,φ 1,r ),...,a(θ L,r ,φ L,r )]，B＝[α 1,r q(τ 1,r ),...,α L,r q(τ L,r )]Wherein B is a column-weighted Van der Monte matrix, which is generated as
First, a 1-mode expansion is performed on the tensor κ to obtain:
then, singular value expansion (SVD) is carried out on the SVD to obtain:
wherein
(·) H Represents a conjugate transpose operation; since a, B columns are full-rank, there is a non-singular matrix M such that UM = B, the vandermonde structure of B is applied,
wherein Z = diag ([ Z) 1,r ,...,z L,r ])，
And B represents the B matrix with the top and bottom rows removed, respectively, introducing the vandermonde property of B, resulting in: u shape 1 M = B and
wherein, U 1 ＝U，
Further, it is possible to obtain:
wherein the content of the first and second substances,
a pseudo-inverse operation is shown as being performed,
due to B, U 1 And U 2 Rank full, generator
The time delay can be extracted by decomposing the characteristic value:
wherein the content of the first and second substances,
a phase angle extraction operation is shown, and therefore,
produce rowColumn ambiguity, known as
Can be explicitly expressed as
Thus, the relationship between the true factor matrix and the estimated factor matrix is:
wherein pi is a matrix of arrangement, { Λ 1 ,Λ 2 Is a scaling matrix, satisfies Λ 1 Λ 2 ＝I，{E 1 ,E 2 Is the estimated error matrix;
(2) AoE, aoA estimation
In the second stage of pilot frequency transmission, UAVs connected with the same AP send different pilot frequency signals, and the AoEs and AoAs of the L-frame UAVs in the target set are estimated by adopting an MUSIC algorithm; rewriting the pilot signal on the k-th sub-carrier received by the APr in equation (3)
Comprises the following steps:
wherein X represents the pilot matrix of the L-frame UAVs, N r,k Representing a noise matrix, the correlation matrix of the received pilot signal is represented as:
in the formula (12), the first and second groups,
indicating to take expectation;
to R S Performing eigenvalue decomposition to obtain a matrix Q = [ Q = [) x |Q n ]，Q x Is listed as the corresponding characteristic value
Characteristic vector of (2), Q n For the noise subspace, a spatial spectrum is defined, since the steering vector a (θ, φ) in the signal subspace is orthogonal to the noise subspace
AoEs and AoAs of UAVs
Will pass through the spatial spectrum P music The L pole of (theta, phi) is estimated;
(3) Permutation ambiguity and complex gain estimation
In practice, the amount of the liquid to be used,
and
the correspondence between them is still unknown, i.e., the arrangement is ambiguous, note that in equations (9), (10),
and
there is another permutation ambiguity between them, i.e.
Having the same permutation matrix Π, using a correlation-based strategy:
wherein, the first and the second end of the pipe are connected with each other,
the AoEs/AoAs will be rearranged according to the time delay so that the permutation ambiguity can be resolved and then the estimated complex gain can be obtained
Step 203: based on estimated channel parameters
And reconstructing the needed L UAV-AP channels.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211475966.1A CN115801071A (en) | 2022-11-23 | 2022-11-23 | Pilot pollution elimination method assisted by structural tensor decomposition for unmanned aerial vehicle communication |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211475966.1A CN115801071A (en) | 2022-11-23 | 2022-11-23 | Pilot pollution elimination method assisted by structural tensor decomposition for unmanned aerial vehicle communication |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115801071A true CN115801071A (en) | 2023-03-14 |
Family
ID=85440646
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211475966.1A Pending CN115801071A (en) | 2022-11-23 | 2022-11-23 | Pilot pollution elimination method assisted by structural tensor decomposition for unmanned aerial vehicle communication |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115801071A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080080630A1 (en) * | 2006-09-29 | 2008-04-03 | Samsung Electronics Co., Ltd. | Channel estimation method and apparatus in an ofdm wireless communication system |
| CN108881077A (en) * | 2018-07-03 | 2018-11-23 | 重庆大学 | A kind of Wiener filtering channel estimation methods based on superimposed pilot |
| CN114189408A (en) * | 2022-01-17 | 2022-03-15 | 电子科技大学 | Channel estimation method for intelligent reflector auxiliary millimeter wave system |
-
2022
- 2022-11-23 CN CN202211475966.1A patent/CN115801071A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080080630A1 (en) * | 2006-09-29 | 2008-04-03 | Samsung Electronics Co., Ltd. | Channel estimation method and apparatus in an ofdm wireless communication system |
| CN108881077A (en) * | 2018-07-03 | 2018-11-23 | 重庆大学 | A kind of Wiener filtering channel estimation methods based on superimposed pilot |
| CN114189408A (en) * | 2022-01-17 | 2022-03-15 | 电子科技大学 | Channel estimation method for intelligent reflector auxiliary millimeter wave system |
Non-Patent Citations (1)
| Title |
|---|
| JIAMIN LI; ZHENGGANG WU; ZIQIAN WAN; PENGCHENG ZHU; DONGMING WANG; XIAOHU YOU: "Structured Tensor CP Decomposition-Aided Pilot Decontamination for UAV Communication in Cell-Free Massive MIMO Systems", 《IEEE COMMUNICATIONS LETTER》, 30 September 2022 (2022-09-30) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liao et al. | Terahertz ultra-massive MIMO-based aeronautical communications in space-air-ground integrated networks | |
| CN110099016B (en) | Millimeter wave sparse array surface channel estimation method based on deep learning network | |
| Abdallah et al. | Deep learning-based frequency-selective channel estimation for hybrid mmWave MIMO systems | |
| WO2017219389A1 (en) | Methods for sending and receiving synchronization signals and signals subjected to perfect omnidirectional pre-coding in large-scale mimo system | |
| CN108881074B (en) | Broadband millimeter wave channel estimation method under low-precision hybrid architecture | |
| US11012131B2 (en) | Method for hybrid precoding and communication device | |
| CN112202479A (en) | Low-complexity signal detection method for MIMO-orthogonal time-frequency space system | |
| Zhang et al. | Channel estimation and training design for hybrid multi-carrier mmwave massive MIMO systems: The beamspace ESPRIT approach | |
| CN107465636B (en) | Channel estimation method of millimeter wave large-scale array space-frequency double-broadband system | |
| CN115021843B (en) | Cooperative sensing method for millimeter wave communication multi-user system | |
| CN109861933B (en) | Millimeter wave MIMO channel estimation method based on MUSIC algorithm and precoding | |
| CN110212951B (en) | Large-scale MIMO channel estimation method based on Butler matrix | |
| CN101232316B (en) | Method for estimating wave arriving based on mobile WIMAX | |
| CN106130938B (en) | Multi-user joint channel estimation method for TDD large-scale MIMO system | |
| Cheng et al. | Wideband channel estimation for millimeter wave beamspace MIMO | |
| CN113644941B (en) | Fuzzy phase rapid elimination method based on large-scale MIMO receiving array structure | |
| CN112769462B (en) | Millimeter wave MIMO broadband channel estimation method based on joint parameter learning | |
| CN110086734B (en) | Channel tracking and recovery method in time-varying large-scale MIMO system | |
| CN115801071A (en) | Pilot pollution elimination method assisted by structural tensor decomposition for unmanned aerial vehicle communication | |
| CN112202481A (en) | Compressed sensing channel estimation algorithm based on adaptive sensing matrix and implementation device | |
| Gupta et al. | An Affine Precoded Superimposed Pilot-Based mmWave MIMO-OFDM ISAC System | |
| Zhang et al. | Adaptive codebook-based channel estimation in OFDM-aided hybrid beamforming mmWave systems | |
| Sheng et al. | Blind AoA estimation with hybrid antenna arrays | |
| Li | Parameter estimation of maneuvering target using maximum likelihood estimation for MIMO radar with colocated antennas | |
| Wang et al. | Oversampling based analog beamforming for initial access with a large number of receive antennas |
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 |