CN111478724A - Three-dimensional wave beam searching method for millimeter wave platform of unmanned aerial vehicle - Google Patents

Abstract

The invention discloses a three-dimensional wave beam searching method facing an unmanned aerial vehicle millimeter wave platform, which comprises the steps of establishing a communication channel model when a millimeter wave plane array is adopted at a transmitting end and a receiving end; establishing a three-dimensional layered search model during millimeter wave planar array; calculating an ideal beam forming vector as an initial vector; calculating a hybrid beamforming vector; and searching according to the three-dimensional layered search model. According to the invention, the search domain is expanded into the three-dimensional space, so that the beam search and alignment in the three-dimensional space can be realized, and the method is more suitable for the millimeter wave communication scene of the three-dimensional flying unmanned aerial vehicle. The invention designs an ideal beam forming vector by combining beam design and a Fourier series method, and designs a millimeter wave beam design method for a hybrid beam forming system under the condition of considering system complexity and efficiency.

Description

Three-dimensional wave beam searching method for millimeter wave platform of unmanned aerial vehicle

Technical Field

The invention belongs to the technical field of beam forming of millimeter wave communication of unmanned aerial vehicles, and particularly relates to a three-dimensional beam searching method for a millimeter wave platform of an unmanned aerial vehicle.

Background

Millimeter waves attract people's attention due to the huge license-free continuous bandwidth (30-300 GHz) possessed by the millimeter waves and high-speed data transmission capacity, and the millimeter waves are reliable choices for high-data-rate applications such as wireless local area networks, fifth-generation cellular networks and vehicle-mounted networks. In millimeter wave communication systems, beam forming technology is usually adopted to overcome the transmission path loss caused by high frequency band, and in order to ensure good communication performance, it is necessary to ensure that the beams at the transmitting and receiving ends are aligned with each other.

However, when the millimeter wave signal transmitting/receiving end is located on the platform of the unmanned aerial vehicle, three-dimensional relative motion between the transmitting and receiving ends easily causes beam mismatch, which affects communication quality, and therefore, beam search is required to be adopted to achieve beam alignment before data transmission is performed at the transmitting and receiving ends. The beam searching and aligning method for millimeter wave communication of the unmanned aerial vehicle faces many challenges, firstly the searching method needs to be suitable for a three-dimensional space, secondly the load of the unmanned aerial vehicle is very limited, the hardware weight and the energy consumption of a communication system are reduced as much as possible, and finally, in order to ensure the reliability of the communication system, the searching and aligning method needs to have a low mismatching rate.

Exhaustive search is a reliable beam search method, however, the search complexity of this method is too high, and therefore hierarchical search is usually employed. The hierarchical searching adopts an equal division method to gradually reduce the searching angle range, in the searching process of each layer, a transmitting terminal equally divides a searching area into a plurality of parts, transmitting beams are sequentially generated to different searching areas, and a receiving terminal generates receiving beams by adopting the same method. However, the traditional layered search method is only suitable for a two-dimensional search space and cannot work normally in the three-dimensional space, so that a fast and efficient three-dimensional beam search method for unmanned aerial vehicle millimeter wave communication needs to be researched, a hybrid beam forming system is adopted at both the transmitting end and the receiving end of an unmanned aerial vehicle communication system, the hybrid beam system integrates the architectures of an analog beam forming system and a digital beam forming system, and accurate design of beams can be achieved without consuming a large number of components.

Disclosure of Invention

The technical problem to be solved by the invention is to provide a three-dimensional beam searching method facing to a millimeter wave platform of an unmanned aerial vehicle aiming at the defects of the prior art, and the method aims at realizing real-time searching and aligning of three-dimensional beams by taking low complexity and high searching efficiency as targets in a millimeter wave directional communication scene of the unmanned aerial vehicle.

In order to achieve the technical purpose, the technical scheme adopted by the invention is as follows:

a three-dimensional beam searching method for an unmanned aerial vehicle millimeter wave platform comprises the following steps:

the method comprises the following steps: establishing a communication channel model when the transmitting end and the receiving end both adopt millimeter wave planar arrays;

step two: establishing a three-dimensional layered search model during millimeter wave planar array, and calculating a search area required by the three-dimensional layered search model;

step three: calculating an ideal beam forming vector as an initial vector;

step four: calculating a hybrid beamforming vector;

step five: and searching according to the three-dimensional layered search model.

In order to optimize the technical scheme, the specific measures adopted further comprise:

the transmitting end and the receiving end both adopt a hybrid beam forming system, and the hybrid beam forming system fuses the architectures of the analog beam forming system and the digital beam forming system.

In the first step, when the two ends of the transmitter both adopt millimeter wave planar arrays, the received signals are represented as:

wherein, y tableDenotes a received signal, P denotes a transmit power, H denotes a channel matrix, r denotes a transmission symbol,

representing beamforming vectors of a transmitting-side hybrid beamforming system, precoding vectors by digital baseband

And a radio frequency precoding matrix

Composition of, wherein N_BSRepresents the number of Radio Frequency (RF) chains of the transmit-side beamforming system;

representing beamforming vectors of a receiving-side hybrid beamforming system, precoding vectors by digital baseband

And a radio frequency precoding matrix

Composition of, wherein N_MSRepresenting the number of RF chains of the receiving end beamforming system; n is the mean 0 and the variance is sigma²Additive white Gaussian noise, M_BSAnd M_MSRespectively representing the number of antennas at the transmitting end and the receiving end.

The communication channel model when the transmitting end and the receiving end which are established in the first step adopt the millimeter wave planar array is as follows:

wherein L denotes the number of channel paths, q_lIs the channel gain of the l-th path, a_BS(ψ_h,ψ_v) And a_MS(ψ'_h,ψ'_v) Array responses of a transmitting end and a receiving end, respectively, wherein (psi)_h,ψ_v) Indicating signal transmissionThe phase of the Angle (AOD) in the horizontal and vertical domains, and (ψ'_h,ψ'_v) Representing the phase of the angle of arrival (AOA) of the signal in the horizontal and vertical domains [. ]]^HA conjugate transpose transform representing a matrix;

for M_BS＝M_h×M_vFor a Uniform Planar Array (UPA) of array elements, the array response at the receiving end is expressed as:

wherein the content of the first and second substances,

denotes the kronecker product, a_BS(ψ_h) And a_BS(ψ_v) Respectively expressed as:

wherein [ ·]^TTranspose transform of representation matrix, for M_MS＝M_h×M_vThe receiving end array response of the uniform planar array of the array elements is expressed as:

wherein, a_MS(ψ′_h) And a_MS(ψ′_v) Respectively expressed as:

in the second step, the established three-dimensional layered search model consists of S layers, and the S-th layer (S is more than or equal to 1 and less than or equal to S) consists of 2^s-1×2^s-1Sub-sets, each sub-set corresponding to a part of the spatial region, let k_hRepresents the k-th in the horizontal domain_hSubset, k_vDenotes the kth in the vertical domain_vA subset of(k_h,k_v) The spatial regions corresponding to the subsets are represented as:

wherein

Wherein, [ psi_hb,ψ_he]Is the coverage of the entire model in the horizontal domain, [ psi_vb,ψ_ve]Is the coverage of the entire model on the vertical domain;

dividing each subset coverage area into 4 parts in average, then in the s-th search, the (k) th_h,k_v) The (b) th of the subset_h,b_v) The coverage of each part is as follows:

in the searching process, the transmitting end adopts a beam forming vector

Generating beams corresponding to the range, the receiving end adopting

A corresponding range of beams is generated.

Calculating the ideal beamforming vector as an initial vector in the third step includes:

3.1) calculating an ideal beam forming matrix, wherein the calculation formula is as follows:

wherein, A (psi)_h,ψ_v) Is arrayed in (psi)_h,ψ_v) Array gain of (d), X (m)_h) Is the (m) th_h,m_v) An antenna array element is horizontalRatio of coordinates to antenna spacing, Y (m)_v) Is the (m) th_h,m_v) The ratio of the longitudinal coordinate of each antenna array element to the antenna interval;

3.2) calculating the array gain, and the formula is as follows:

wherein (ω)_h0,ω_v0) Is the midpoint of the beam coverage area, ω_hbAnd ω_vbThe widths of the midpoint from the horizontal domain boundary and the vertical domain boundary are respectively, and the beam coverage area is obtained in the second step;

3.3) calculating an ideal beam forming vector, wherein the formula is as follows:

will be provided with

Conversion to

This vector is the ideal beamforming vector.

The calculating the hybrid beamforming vector in the fourth step includes:

4.1) initializing the RF precoding matrix C_RFAnd residual vector c_res：C_RF＝[]，c_res＝c_opt；

4.2) setting the number N of RF chains_RFWhen I is less than or equal to N_RFRepeating the steps 4.3) -4.7);

4.3) updating the radio frequency precoding matrix: c_RF＝[C_RF,υ(c_res)]Wherein upsilon (c)_res) Is c_resThe quantized vector has a set of quantized values of { e }^-jπ/2,1,e^jπ/2,e^jπ}；

4.4) calculating c_resMaximum element Ma and minimum element Mi in (1):

4.5) calculating the coefficient: ' mean [ c_res(J)/υ(c_res)(J)]Wherein J ═ find [ | c [ ]_res|≥(Ma+Mi)/2]；

4.6) calculating residual coefficients: if (| ' > (Ma + Mi)/2, ('/| ' |) ((Ma + Mi)/2), otherwise ═ f;

4.7) update residual: c. C_res＝c_res-υ(c_res)；

4.8) after finishing the iteration, calculating a baseband precoding vector:

4.9) normalized baseband precoding vector: c. C_BB＝c_BB/||c_BB||₂；

4.10) according to the latest C_RFAnd c_BBObtaining the final mixed beam forming vector C ═ C_RFc_BB。

The searching according to the three-dimensional hierarchical searching model in the fifth step includes:

5.1) initializing parameters, making s equal to 0, (k)_h,k_v,k′_h,k′_v)＝(1,1,1,1)；

5.2) transmitting end adoption

b_h∈{1,2}，b_v∈ {1,2} to generate 4 transmitting beams in sequence, the receiving end adopts

b′_h∈{1,2}，b′_v∈ {1,2} sequentially generate 4 receive beams;

5.3) according to the beam obtained in step 5.2), 4 × 4 received signals y can be generated in total, the transmitting and receiving beam forming vector which can generate the maximum signal power is selected, and the b at the moment is recorded_h，b_v，b′_hAnd b'_vA value of (d);

5.4) update the value, k_h＝2(k_h-1)+b_h，k_v＝2(k_v-1)+b_v，k′_h＝2(k′_h-1)+b′_h，k′_v＝2(k′_v-1)+b′_v，s＝s+1；

5.5) go to the next layer search and repeat steps 5.2) -5.4) until the desired beam resolution is obtained.

The invention has the following beneficial effects:

1) according to the invention, the search domain is expanded into the three-dimensional space, so that the beam search and alignment in the three-dimensional space can be realized, and the method is more suitable for the millimeter wave communication scene of the three-dimensional flying unmanned aerial vehicle.

2) The invention designs an ideal beam forming vector by combining beam design and a Fourier series method, and designs a millimeter wave beam design method for a hybrid beam forming system under the condition of considering system complexity and efficiency.

Drawings

Fig. 1 is a block diagram of a millimeter wave hybrid beam forming system based on the present invention;

FIG. 2 is a flow chart of the method of the present invention;

FIG. 3 is a spatial region partition diagram in the search model of the present invention;

FIG. 4 is an ideal beamforming matrix of the present invention;

fig. 5 is a simulation diagram of beam directions after the beam forming vectors of the present invention are used.

Detailed Description

Embodiments of the present invention are described in further detail below with reference to the accompanying drawings.

Referring to fig. 2, the three-dimensional beam search method for the millimeter wave platform of the unmanned aerial vehicle of the invention comprises:

the method comprises the following steps: establishing a communication channel model when the transmitting end and the receiving end both adopt millimeter wave planar arrays; and the two transmitting and receiving ends adopt a hybrid beam forming system, and the hybrid beam forming system fuses the architectures of the analog beam forming system and the digital beam forming system.

Step two: establishing a three-dimensional layered search model during millimeter wave planar array, and calculating a search area required by the three-dimensional layered search model;

step three: calculating an ideal beam forming vector as an initial vector;

step four: calculating a hybrid beamforming vector;

step five: and searching according to the three-dimensional layered search model.

In an embodiment, when both ends of the transmitter in step one adopt the millimeter wave planar array, the received signal is represented as:

where y denotes a received signal, P denotes a transmit power, H denotes a channel matrix, r denotes a transmission symbol,

representing beamforming vectors of a transmitting-side hybrid beamforming system, precoding vectors by digital baseband

And a radio frequency precoding matrix

Composition of, wherein N_BSRepresenting the number of radio frequency chains of the wave beam shaping system at the transmitting end;

representing beamforming vectors of a receiving-side hybrid beamforming system, precoding vectors by digital baseband

And a radio frequency precoding matrix

Composition of, wherein N_MSRepresenting the number of RF chains of the receiving end beamforming system; n is the mean value of 0, squareThe difference is sigma²Additive white Gaussian noise, M_BSAnd M_MSRespectively representing the number of antennas at the transmitting end and the receiving end.

In an embodiment, the communication channel model when the transmitting and receiving ends established in the first step both use the millimeter wave planar array is as follows:

wherein L denotes the number of channel paths, q_lIs the channel gain of the l-th path, a_BS(ψ_h,ψ_v) And a_MS(ψ'_h,ψ'_v) Array responses of a transmitting end and a receiving end, respectively, wherein (psi)_h,ψ_v) Phase representing emission angle of signal in horizontal and vertical domains, and (ψ'_h,ψ'_v) Phase representing angle of arrival of signal in horizontal and vertical domains [ ·]^HA conjugate transpose transform representing a matrix;

for M_BS＝M_h×M_vThe array response of the receiving end of the uniform planar array of the array elements is expressed as follows:

wherein the content of the first and second substances,

denotes the kronecker product, a_BS(ψ_h) And a_BS(ψ_v) Respectively expressed as:

wherein [ ·]^TTranspose transform of representation matrix, for M_MS＝M_h×M_vThe receiving end array response of the uniform planar array of the array elements is expressed as:

wherein, a_MS(ψ′_h) And a_MS(ψ′_v) Respectively expressed as:

in the embodiment, in the second step, the established three-dimensional layered search model consists of S layers, and the S-th layer (S is more than or equal to 1 and less than or equal to S) consists of 2^s-1×2^s-1Sub-sets, each sub-set corresponding to a part of the spatial region, let k_hRepresents the k-th in the horizontal domain_hSubset, k_vDenotes the kth in the vertical domain_vA subset, then (k) th_h,k_v) The spatial regions corresponding to the subsets are represented as:

wherein

Wherein, [ psi_hb,ψ_he]Is the coverage of the entire model in the horizontal domain, [ psi_vb,ψ_ve]Is the coverage of the entire model on the vertical domain;

dividing each subset coverage area into 4 parts in average, then in the s-th search, the (k) th_h,k_v) The (b) th of the subset_h,b_v) The coverage of each part is as follows:

in the searching process, the transmitting end adopts a beam forming vector

Generating beams corresponding to the range, the receiving end adopting

A corresponding range of beams is generated.

In an embodiment, the calculating an ideal beamforming vector as the initial vector in step three includes:

3.1) calculating an ideal beam forming matrix, wherein the calculation formula is as follows:

wherein, A (psi)_h,ψ_v) Is arrayed in (psi)_h,ψ_v) Array gain of (d), X (m)_h) Is the (m) th_h,m_v) The ratio of the abscissa of the individual antenna elements to the antenna spacing, Y (m)_v) Is the (m) th_h,m_v) The ratio of the longitudinal coordinate of each antenna array element to the antenna interval;

3.2) calculating the array gain, and the formula is as follows:

wherein (ω)_h0,ω_v0) Is the midpoint of the beam coverage area, ω_hbAnd ω_vbThe widths of the midpoint from the horizontal domain boundary and the vertical domain boundary are respectively, and the beam coverage area is obtained in the second step;

3.3) calculating an ideal beam forming vector, wherein the formula is as follows:

will be provided with

Conversion to

This vector is the ideal beamforming vector.

In the embodiment, the calculating of the hybrid beamforming vector in the step four, the frame implemented by the invention in combination with the hybrid beamforming system specifically includes:

4.1) initializing the RF precoding matrix C_RFAnd residual vector c_res：C_RF＝[]，c_res＝c_opt；

4.2) setting the number N of RF chains_RFWhen I is less than or equal to N_RFRepeating the steps 4.3) -4.7);

4.3) updating the radio frequency precoding matrix: c_RF＝[C_RF,υ(c_res)]Wherein upsilon (c)_res) Is c_resThe quantized vector has a set of quantized values of { e }^-jπ/2,1,e^jπ/2,e^jπ}；

4.4) calculating c_resMaximum element Ma and minimum element Mi in (1):

4.5) calculating the coefficient: ' mean [ c_res(J)/υ(c_res)(J)]Wherein J ═ find [ | c [ ]_res|≥(Ma+Mi)/2]；

4.6) calculating residual coefficients: if (| ' > (Ma + Mi)/2, ('/| ' |) ((Ma + Mi)/2), otherwise ═ f;

4.7) update residual: c. C_res＝c_res-υ(c_res)；

4.8) after finishing the iteration, calculating a baseband precoding vector:

4.9) normalized baseband precoding vector: c. C_BB＝c_BB/||c_BB||₂；

4.10) according to the latest C_RFAnd c_BBObtaining the final mixed beam forming vector C ═ C_RFc_BB。

In an embodiment, the searching according to the three-dimensional hierarchical search model in step five includes:

5.1) initializing parameters, making s equal to 0, (k)_h,k_v,k′_h,k′_v)＝(1,1,1,1)；

5.2) transmitting end adoption

b_h∈{1,2}，b_v∈ {1,2} to generate 4 transmitting beams in sequence, the receiving end adopts

b′_h∈{1,2}，b′_v∈ {1,2} sequentially generate 4 receive beams;

5.3) according to the beam obtained in step 5.2), 4 × 4 received signals y can be generated in total, the transmitting and receiving beam forming vector which can generate the maximum signal power is selected, and the b at the moment is recorded_h，b_v，b′_hAnd b'_vA value of (d);

5.4) update the value, k_h＝2(k_h-1)+b_h，k_v＝2(k_v-1)+b_v，k′_h＝2(k′_h-1)+b′_h，k′_v＝2(k′_v-1)+b′_v，s＝s+1；

5.5) go to the next layer search and repeat steps 5.2) -5.4) until the desired beam resolution is obtained.

The three-dimensional beam searching method for the millimeter wave unmanned aerial vehicle communication platform of the invention is specifically described below by the drawings and the embodiments.

This embodiment takes the first layer search as an example, where the transmitting end and the receiving end both use the array element number M_BS＝M _MS21 × 21, the number of RF chains is N_BS＝N_MSThe UPA array (as shown in fig. 1) with carrier frequency of 60GHz millimeter wave, array element spacing of half wavelength, transmission channel of block fading channel, L being 1, q being 1, and transmission power being P_BS30dBm, the beam forming vector of the transmitting and receiving end meets the requirement | | | c | | non-woven phosphor screen₂＝||w||₂1, phase of the transmitting end (ψ)_h,ψ_v) (pi/18 ), phase (ψ 'at the receiving end'_h,ψ′_v) The search ranges of the transmitting and receiving ends are [ psi [ (. pi./18, Pi./18) ]_hb,ψ_he]＝[-π/2,π/2]，[ψ_vb,ψ_ve]＝[-π/2,π/2]。

The embodiment comprises the following steps:

the method comprises the following steps: establishing a channel model of millimeter wave beam communication, which comprises the following steps:

1.1) calculate the array response of the transmitting end 21 × 21 array elements UPA:

likewise, receive end array response is calculated:

1.2) calculating a communication channel model of which the transmitting end and the receiving end adopt UPA (unified power automation) respectively:

step two: establishing a three-dimensional layered search model during millimeter wave planar array, and calculating a search area required by the three-dimensional layered search model, wherein the method specifically comprises the following steps:

2.1) first-tier search where s is 1, we can get 2^s-11, so the numbering of the subsets is initialized to k_h＝k_v＝1。

2.2) calculating the coverage area of the subset:

wherein

2.3) dividing the subset into 4 parts, calculating the coverage area of each part:

step three: calculating an ideal beamforming vector as follows:

3.1) calculating the ideal beam forming matrix:

wherein m is_h＝1:21，m_vWith 1:21, beam forming matrices corresponding to 4 coverage areas are calculated according to step 3.1) and step 3.2), and the obtained matrices with the size of 21 × 21 are modulo by each point, so as to obtain the scatter diagram as shown in fig. 4.

3.2) calculating the array gain:

3.3) ideally beamforming matrix

Conversion to

The desired beamforming vector is the beamforming vector.

Step four: calculating a hybrid beamforming vector of the hybrid beamforming system, specifically as follows:

4.1) initializing the RF precoding matrix C_RFAnd residual vector c_res：C_RF＝[]，c_res＝c_opt。

4.2) setting the number N of RF chains_RFWhen I is less than or equal to N_RFThe following steps 4.3) -4.7) are repeated.

4.3) updating the radio frequency precoding matrix: c_RF＝[C_RF,υ(c_res)]。

4.4) calculating c_resMaximum and minimum elements of (1):

4.5) calculating the coefficient: ' mean [ c_res(J)/υ(c_res)(J)]。

4.6) calculating residual coefficients.

4.7) update residual: c. C_res＝c_res-υ(c_res)。

4.8) after finishing the iteration, calculating a baseband precoding vector:

4.9) normalized baseband precoding vector: c. C_BB＝c_BB/||c_BB||₂。

4.10) according to the latest C_RFAnd c_BBObtaining the final mixed beam forming vector C ═ C_RFc_BB。

Step five: and carrying in a beam forming vector, and operating a search process, wherein the search process specifically comprises the following steps:

5.1) if s is 0, (k)_h,k_v,k′_h,k′_v)＝(1,1,1,1)。

5.2) transmitting end adoption

b_h∈{1,2}，b_v∈ {1,2}, the receiving end adopts

b′_h∈{1,2}，b′_v∈{1,2}。

5.3) from the beams obtained in step 5.2), 4 × 4 received signals y can be generated, wherein

y＝w^HHc+w^Hn

The selection may result in the bestTransmit and receive beamforming vectors for large signal power, recording b at that time_h，b_v，b′_hAnd b'_vA value of (a) to (b)_h＝2，b_v＝2，b′_h＝2，b′_v＝2。

5.4) update the value, k_h＝2(k_h-1)+b_h＝2，k_v＝2(k_v-1)+b_v＝2，k′_h＝2(k′_h-1)+b′_h＝2，k′_v＝2(k′_v-1)+b′_v＝2，s＝s+1＝2。

5.5) enter next layer search, recalculate beamforming vector according to step two to step four, and repeat steps 5.2-5.4) until the desired beam resolution is obtained.

The effect obtained by the present embodiment can be further illustrated by the beam shape obtained in the simulation experiment of fig. 5. Fig. 5 corresponds to beamforming vector c_(3,3,3,1,1)The designed beam can be seen that the beam covering shape meets the requirement of a search model, the beam shape gradually approaches to an ideal situation when the number of array elements is increased, higher average beam gain is possessed, the beam gain outside the covering area is almost zero, the interference to other three beams can be reduced, and the search effect is effectively improved.

The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above-mentioned embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may be made by those skilled in the art without departing from the principle of the invention.

Claims (8)

1. A three-dimensional beam searching method for a millimeter wave platform of an unmanned aerial vehicle is characterized by comprising the following steps:

the method comprises the following steps: establishing a communication channel model when the transmitting end and the receiving end both adopt millimeter wave planar arrays;

step two: establishing a three-dimensional layered search model during millimeter wave planar array, and calculating a search area required by the three-dimensional layered search model;

step three: calculating an ideal beam forming vector as an initial vector;

step four: calculating a hybrid beamforming vector;

step five: and searching according to the three-dimensional layered search model.

2. The method of claim 1, wherein a hybrid beam forming system is used at both the transmitting end and the receiving end, and the hybrid beam system combines architectures of an analog beam forming system and a digital beam forming system.

3. The three-dimensional beam searching method for the millimeter wave platform of the unmanned aerial vehicle as claimed in claim 2, wherein when the millimeter wave planar array is used at both ends of the transmitter in the step one, the received signal is represented as:

where y denotes a received signal, P denotes a transmit power, H denotes a channel matrix, r denotes a transmission symbol,

representing beamforming vectors of a transmitting-side hybrid beamforming system, precoding vectors by digital baseband

And a radio frequency precoding matrix

Composition of, wherein N_BSRepresenting the number of radio frequency chains of the wave beam shaping system at the transmitting end;

indicating receive side hybrid beamformingBeamforming vectors for systems precoded by digital baseband

And a radio frequency precoding matrix

Composition of, wherein N_MSRepresenting the number of RF chains of the receiving end beamforming system; n is the mean 0 and the variance is sigma²Additive white Gaussian noise, M_BSAnd M_MSRespectively representing the number of antennas at the transmitting end and the receiving end.

4. The three-dimensional beam searching method for the millimeter wave platform of the unmanned aerial vehicle as claimed in claim 3, wherein the communication channel model when the millimeter wave planar array is adopted at both the transmitting end and the receiving end established in the first step is:

wherein L denotes the number of channel paths, q_lIs the channel gain of the l-th path, a_BS(ψ_h,ψ_v) And a_MS(ψ'_h,ψ'_v) Array responses of a transmitting end and a receiving end, respectively, wherein (psi)_h,ψ_v) Phase representing emission angle of signal in horizontal and vertical domains, and (ψ'_h,ψ'_v) Phase representing angle of arrival of signal in horizontal and vertical domains [ ·]^HA conjugate transpose transform representing a matrix;

for M_BS＝M_h×M_vThe array response of the receiving end of the uniform planar array of the array elements is expressed as follows:

wherein the content of the first and second substances,

denotes the kronecker product, a_BS(ψ_h) And a_BS(ψ_v) Respectively expressed as:

wherein [ ·]^TTranspose transform of representation matrix, for M_MS＝M_h×M_vThe receiving end array response of the uniform planar array of the array elements is expressed as:

wherein, a_MS(ψ′_h) And a_MS(ψ′_v) Respectively expressed as:

。

5. the method as claimed in claim 4, wherein in the second step, the three-dimensional layered search model is composed of S layers, and the S-th layer (S is greater than or equal to 1 and less than or equal to S) is composed of 2^s-1×2^s-1Sub-sets, each sub-set corresponding to a part of the spatial region, let k_hRepresents the k-th in the horizontal domain_hSubset, k_vDenotes the kth in the vertical domain_vA subset, then (k) th_h,k_v) The spatial regions corresponding to the subsets are represented as:

wherein

Wherein, [ psi_hb,ψ_he]Is the coverage of the entire model in the horizontal domain, [ psi_vb,ψ_ve]Is the coverage of the entire model on the vertical domain;

dividing each subset coverage area into 4 parts in average, then in the s-th search, the (k) th_h,k_v) The (b) th of the subset_h,b_v) The coverage of each part is as follows:

in the searching process, the transmitting end adopts a beam forming vector

Generating beams corresponding to the range, the receiving end adopting

A corresponding range of beams is generated.

6. The method of claim 5, wherein the step three of calculating an ideal beamforming vector as an initial vector comprises:

3.1) calculating an ideal beam forming matrix, wherein the calculation formula is as follows:

wherein, A (psi)_h,ψ_v) Is arrayed in (psi)_h,ψ_v) Array gain of (d), X (m)_h) Is the (m) th_h,m_v) The ratio of the abscissa of the individual antenna elements to the antenna spacing, Y (m)_v) Is the (m) th_h,m_v) The ratio of the longitudinal coordinate of each antenna array element to the antenna interval;

3.2) calculating the array gain, and the formula is as follows:

wherein (ω)_h0,ω_v0) Is the midpoint of the beam coverage area, ω_hbAnd ω_vbThe widths of the midpoint from the horizontal domain boundary and the vertical domain boundary are respectively, and the beam coverage area is obtained in the second step;

3.3) calculating an ideal beam forming vector, wherein the formula is as follows:

will be provided with

Conversion to

This vector is the ideal beamforming vector.

7. The method of claim 6, wherein the step four of calculating the hybrid beamforming vector comprises:

4.1) initializing the RF precoding matrix C_RFAnd residual vector c_res：C_RF＝[]，c_res＝c_opt；

4.2) setting the number N of RF chains_RFWhen I is less than or equal to N_RFRepeating the steps 4.3) -4.7);

4.3) updating the radio frequency precoding matrix: c_RF＝[C_RF,υ(c_res)]Wherein upsilon (c)_res) Is c_resThe quantized vector has a set of quantized values of { e }^-jπ/2,1,e^jπ/2,e^jπ}；

4.4) calculating c_resMaximum element Ma and minimum element Mi in (1):

4.5) calculating the coefficient: ' mean [ c_res(J)/υ(c_res)(J)]Wherein J ═ find [ | c [ ]_res|≥(Ma+Mi)/2]；

4.6) calculating residual coefficients: if (| ' > (Ma + Mi)/2, ('/| ' |) ((Ma + Mi)/2), otherwise ═ f;

4.7) update residual: c. C_res＝c_res-υ(c_res)；

4.8) after finishing the iteration, calculating a baseband precoding vector:

4.9) normalized baseband precoding vector: c. C_BB＝c_BB/||c_BB||₂；

4.10) according to the latest C_RFAnd c_BBObtaining the final mixed beam forming vector C ═ C_RFc_BB。

8. The method of claim 7, wherein the step five of searching according to the three-dimensional hierarchical search model comprises:

5.1) initializing parameters, making s equal to 0, (k)_h,k_v,k′_h,k′_v)＝(1,1,1,1)；

5.2) transmitting end adoption

b_h∈{1,2}，b_v∈ {1,2} to generate 4 transmitting beams in sequence, the receiving end adopts

b′_h∈{1,2}，b′_v∈ {1,2} sequentially generate 4 receive beams;

5.3) according to the beam obtained in step 5.2), 4 × 4 received signals y can be generated in total, the transmitting and receiving beam forming vector which can generate the maximum signal power is selected, and the b at the moment is recorded_h，b_v，b′_hAnd b'_vA value of (d);

5.4) update the value, k_h＝2(k_h-1)+b_h，k_v＝2(k_v-1)+b_v，k′_h＝2(k′_h-1)+b′_h，k′_v＝2(k′_v-1)+b′_v，s＝s+1；

5.5) go to the next layer search and repeat steps 5.2) -5.4) until the desired beam resolution is obtained.

Images