A millimeter wave MIMO system model extraction positioning method

Abstract

The application discloses a millimeter wave MIMO system model extraction positioning method, comprising the following steps: in a millimeter wave MIMO system, based on a mixed Gaussian noise model, using a message passing algorithm, estimating TOA, AOA and AOD of a signal to obtain an estimated value, and determining the positions of user equipment and scatterers in the millimeter wave MIMO system according to the estimated value of the TOA, AOA and AOD, so that the accurate positioning of the user equipment and the base station in the millimeter wave MIMO system is realized.

Description

Technical Field

[0001] This invention relates to a positioning method, specifically a positioning method for extracting models of millimeter-wave MIMO systems. Background Technology

[0002] Millimeter-wave technology is one of the key technologies in wireless communication. Due to the large available bandwidth in the millimeter-wave band, it can support high-speed transmission and also gives millimeter-wave signals high time resolution. However, compared to lower-frequency systems, millimeter-wave systems require larger directional gain to overcome higher path loss. Furthermore, rainwater, oxygen, and other factors can cause additional losses to millimeter-wave systems. Larger directional gain typically requires large physical apertures or large antenna arrays. Because millimeter-wave signals have shorter wavelengths, massive MIMO (Massively Interactive Antennas) can achieve large beamforming gain at a relatively low cost. Large antenna arrays generate narrow beams through beamforming, thus enabling millimeter-wave systems to achieve high resolution in the angular domain. At millimeter-wave frequencies, the received power from scattering and multi-reflector reflections is much lower than that from direct paths and single-reflector reflections. Therefore, it can be assumed that only direct paths and a small number of non-direct paths are available at the receiver, making the millimeter-wave channel sparse in the angular domain.

[0003] Based on the above characteristics, millimeter-wave signals exhibit high resolution in both the angular and time domains within the beam space, overcoming the shortcomings of traditional positioning schemes and thus attracting increasing attention in recent years. Theoretically, positioning can be achieved by obtaining information such as the TOA, AOA, and AOD of the signal, and this information can be acquired through channel state information. Considering the sparse characteristics of millimeter-wave channels, many millimeter-wave positioning algorithms are based on compressed sensing.

[0004] Channel noise is an unavoidable factor affecting the performance of communication systems. To achieve high-precision positioning, it is necessary to minimize its interference with the channel estimation process. The Gaussian model is currently the most widely used noise model in communication systems. However, physical research shows that channel noise often exhibits non-Gaussian properties during actual transmission, thus severely impacting the accuracy of user equipment and base station positioning. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a model extraction and positioning method for millimeter-wave MIMO systems, which enables accurate positioning of user equipment and base stations in millimeter-wave MIMO systems.

[0006] To achieve the above objectives, this invention discloses a method for model extraction and localization of millimeter-wave MIMO systems, comprising:

[0007] In millimeter-wave MIMO systems, based on a Gaussian mixture noise model and using a message passing algorithm, the estimated values ​​of TOA, AOA, and AOD of the signal are obtained. and

[0008] Based on the estimated values ​​of TOA, AOA, and AOD and Determine the location of user equipment and scatterers in a millimeter-wave MIMO system.

[0009] AOD estimate for:

[0010]

[0011] Among them, Ind T,k This represents the index of the AOD in the beam space, where d is the distance between antenna elements, and N is the index of the AOD. T K represents the number of transmitting antennas, and K represents the number of observable paths between the base station (BS) and the user equipment (UE).

[0012] AOD index in beam space Ind T,k for:

[0013]

[0014] Among them, Ind k express The index corresponding to the k-th largest value in N R This represents the number of receiving antennas.

[0015] AOA estimate for:

[0016]

[0017] Among them, Ind R,k This represents the index of the AOA in the beam space, where d is the distance between antenna elements, and N is the index of the AOA. T and N R These represent the number of transmitting antennas and receiving antennas, respectively.

[0018] AOA index in beam space Ind R,k for:

[0019] Ind R,k =mod(Ind) k -1,N R )+1.

[0020] TOA estimate for:

[0021]

[0022] in, This is an estimate of the channel fading coefficient at the k-th subcarrier. This represents the grid candidate value for TOA. B represents bandwidth.

[0023] Location of user equipment in millimeter-wave MIMO system for:

[0024]

[0025] Where c is the speed of light.

[0026] The location of the scatterer in a millimeter-wave MIMO system is:

[0027]

[0028]

[0029] in, This refers to the antenna direction of the user equipment.

[0030] Antenna direction of user equipment for:

[0031]

[0032] The present invention has the following beneficial effects:

[0033] In practical operation, the millimeter-wave MIMO system model extraction and localization method described in this invention, based on a Gaussian mixture noise model, uses a message passing algorithm to estimate the TOA, AOA, and AOD of the signal. and It has greater versatility, and the use of message passing algorithms can improve positioning accuracy, further enhancing the accuracy based on the estimated values ​​of the signal's TOA, AOA, and AOD. and Determining the location of user equipment and scatterers in a millimeter-wave MIMO system is simple, convenient, and easy to implement. Attached Figure Description

[0034] Figure 1 This is a structural diagram of a millimeter-wave positioning system;

[0035] Figure 2 A comparison chart of algorithm performance under different models;

[0036] Figure 3 A comparison chart of channel estimation performance with different numbers of antennas;

[0037] Figure 4 The image shows the results of scatterer position estimation under different SNR values. Detailed Implementation

[0038] To enable those skilled in the art to better understand the present invention, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present invention, not all embodiments, and are not intended to limit the scope of the present invention. Furthermore, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessary confusion regarding the concepts disclosed in the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort should fall within the scope of protection of the present invention.

[0039] The accompanying drawings show structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not drawn to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0040] refer to Figure 1 Consider a millimeter-wave MIMO system with N transmitting antennas and N receiving antennas. T and N R The system includes one base station (BS) and one user equipment (UE), wherein the location of the base station (BS) is p. T Given that the direction of its antenna is known, the location p of the user equipment (UE) R and its antenna direction Unknown. Additionally, the position of the scatterer in the system is denoted by s.

[0041] 1) Send signal

[0042] Considering the transmitted signal is an OFDM symbol, and assuming the base station BS transmits M OFDM symbols on N subcarriers, then the m-th symbol on the n-th subcarrier is represented as:

[0043]

[0044] Where, N F This indicates the number of signal clusters.

[0045] 2) Channel

[0046] Suppose there are K ≥ 1 observable paths between the base station (BS) and the user equipment (UE). A LOS path is denoted by k = 0, and an NLOS path is denoted by k > 0. Assume that an NLOS path originates only from a single scattering or reflection. Each path is associated with three channel parameters. For the k-th path, θ... T,k and θ R,k They represent AOD and AOA respectively, d k The path distance is given, therefore the arrival time (TOA) is... Let c be the speed of light, and let d k1 =||s k -p T ||2,d k2 =||p R -s k ||2, Assuming the channel remains unchanged during the transmission of M signals, the channel matrix of the nth subcarrier is:

[0047]

[0048] in

[0049] A T [n] = [a T,n (θ T,0 ),...,a T,n (θ T,K (3)

[0050] A R [n] = [a R,n (θ R,0 ),...,a T,n (θ R,K (4)

[0051]

[0052] Where, β k and ρ k Let these represent the complex channel gain and path loss of the k-th path, respectively. B represents bandwidth.

[0053] Consider a uniform linear array (ULA), where vector a in equations (3) and (4) T,n (θ T,k ) and a R,n (θ R,k They are respectively:

[0054]

[0055]

[0056] Where, λn Let d be the wavelength of the nth subcarrier, and d be the distance between the antenna elements.

[0057] 3) Receiving signals

[0058] After removing the cyclic prefix and performing a Fast Fourier Transform, the signal on the nth subcarrier of the mth OFDM symbol is represented as:

[0059]

[0060] Among them, F m [n] is N T ×N F The beamforming matrix is ​​known to the user equipment (UE). Additionally, N... m [n] represents the channel noise with zero mean. Considering its non-Gaussian property, a Gaussian mixture noise model is adopted:

[0061]

[0062] To estimate the location of the scatterer, the location of the user equipment (UE), and the antenna orientation, it is first necessary to know the AOD, AOA, and TOA. Since AOD, AOA, and TOA are all channel parameters, it is necessary to estimate the channel matrix.

[0063] The estimation process for AOD, AOA, and TOA is as follows:

[0064] set up Let the angular domain beam space matrix be represented, then we have

[0065]

[0066] Among them, U R and U T For the transformation matrix, in the ULA antenna model, to uniformly sample the angles across the entire beam space, let U be the transformation matrix. T for:

[0067] U T =[u T (v T,0 ),...,u T (v T,0 (11)

[0068]

[0069]

[0070] The angle of the beam space is determined by the grid. Therefore, the AOD angular resolution in the beam space is 1 / N. T Matrix UR Structure and U T Similarly, U R Represented as:

[0071] U R =[u R (v R,0 ),...,u R (v R,0 (14)

[0072]

[0073]

[0074] The angle of the beam space is determined by the grid. Therefore, the AOA angular resolution in the beam space is 1 / N. R .

[0075] Due to U R and U T Since all are unitary matrices, from equation (10), we get:

[0076]

[0077] Substituting (17) into formula (8), we get:

[0078]

[0079] Vectorizing equation (18) yields:

[0080]

[0081] Depend on It can be seen that equation (19) is equivalent to:

[0082]

[0083] in,

[0084] y[n]=vec[(Y1[n]) T ,...,(Y M [n]) T ] T (twenty one)

[0085] T[n]=[T 0,n ,...,T M-1,n ] T (twenty two)

[0086]

[0087]

[0088]

[0089] Where h[n] is an approximate sparse vector.

[0090] The sparse estimate of h is calculated based on the sparse Bayesian learning algorithm. From equation (20), we get:

[0091]

[0092] Let N TR =N T ×N R .

[0093]

[0094] in,

[0095]

[0096]

[0097]

[0098] Where Ga() represents the gamma distribution, and for Gaussian mixed noise, the following prior model is established:

[0099] p(c|κ)=Mult(c|κ) (31)

[0100]

[0101]

[0102] Where Mult() and Dir() represent the multivariate Gaussian distribution and the Dirichlet distribution, respectively, and combined with the prior model, the joint distribution of all parameters is expressed as:

[0103]

[0104] For p(y|h,c,δ), then:

[0105]

[0106] Based on the aforementioned prior model and factor graph, a message-passing algorithm is used to solve for each unknown parameter. When the iteration converges, the estimated value is obtained. The specific iterative update process is as follows:

[0107] 1)

[0108]

[0109] 2)

[0110] 3)

[0111]

[0112] For q(κ), then we have

[0113]

[0114] in, For E[κ] i ] and E[lnκ i ], then there is

[0115]

[0116]

[0117] Where ψ(·) is the digamma function,

[0118] 6) c: For q(c), then we have

[0119]

[0120] because Therefore, q(c) needs to be standardized, then we have

[0121]

[0122] in,

[0123]

[0124] E[c i ] = o i (41)

[0125] The above is the iterative update formula for all unknown variables. Each unknown parameter is iteratively updated in turn. When the iteration process converges, the expected value of each unknown parameter is the corresponding estimated value.

[0126] Let Ind k express The index corresponding to the k-th largest value in the middle, i.e.

[0127]

[0128] Based on this index, the beam space information for AOD and AOA is obtained as follows:

[0129]

[0130] Ind R,k =mod(Ind) k -1,N R )+1 (44)

[0131] Among them, Ind T,k and Ind R,k Let AOD and AOA represent the indices in the beam space, respectively. Based on these indices, the estimated values ​​for AOD and AOA are as follows:

[0132]

[0133]

[0134] Based on the estimates of AOD and AOA, the estimate of TOA is obtained, where the estimated channel expression for the k-th path is:

[0135]

[0136] in, Corresponding to n subcarriers, Ξ k express The estimation error is then:

[0137]

[0138]

[0139]

[0140] If we disregard the influence of n, then we have

[0141]

[0142] in, This represents the grid candidate value for TOA.

[0143] Therefore, τ is obtained k The estimated value

[0144] Based on the estimated values ​​of AOA, AOD, and TOA, the location of the user equipment (UE) and its antenna orientation, as well as the location of the scattering object, are obtained. The specific process is as follows:

[0145] 1) UE location estimation

[0146] In actual calculations, the position of BS is usually placed at the origin of the coordinate axis, i.e., p. T =[0,0] T The UE location is estimated as follows:

[0147]

[0148] 2) The antenna orientation of the user equipment is estimated as follows:

[0149] 3) The location of the scatterer is estimated as follows:

[0150]

[0151]

[0152] Figure 2 , Figure 3 and Figure 4 The technology of this invention is abbreviated as SMLVMP, see reference. Figure 2 , Figure 3 and Figure 4 Compared with the prior art, the positioning performance of the present invention is superior.

[0153] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for model extraction and localization of a millimeter-wave MIMO system, characterized in that, include: In millimeter-wave MIMO systems, based on a Gaussian mixture noise model, the estimated values ​​of TOA, AOA, and AOD of the signal are obtained using a message passing algorithm. The location of user equipment and scatterers in a millimeter-wave MIMO system is determined based on the estimated values ​​of TOA, AOA, and AOD. AOD estimate for: in, This indicates the index of AOD in the beam space. d The distance between antenna elements. For carrier wavelength, K represents the number of transmit antennas, and K represents the number of observable paths between the base station (BS) and the user equipment (UE). AOA estimate for: in, This represents the index of AOA in the beam space. d The distance between antenna elements. and These refer to the number of transmitting antennas and receiving antennas, respectively. TOA estimate for: in, For the first Estimates of the channel fading coefficient at each subcarrier. This represents the grid candidate value for TOA. , For bandwidth, The number of subcarriers used for positioning. The value of TOA; Location of user equipment in millimeter-wave MIMO system for: in, At the speed of light, ; The location of the scatterer in a millimeter-wave MIMO system is: in, This refers to the antenna direction of the user equipment.

2. The millimeter-wave MIMO system model extraction and localization method according to claim 1, characterized in that, AOD index in beam space for: in, express The Middle The index corresponding to the larger value This represents the number of receiving antennas.

3. The millimeter-wave MIMO system model extraction and localization method according to claim 1, characterized in that, AOA index in beam space for: 。 4. The millimeter-wave MIMO system model extraction and localization method according to claim 1, characterized in that, Antenna direction of user equipment for: 。

Images