Joint optimization method for beam spatial MIMO system downlink signal processing based on smart reflecting surface

Abstract

The present application relates to the technical field of wireless communication signal processing, in particular to a kind of beam space Multi-Input Multi-Output (MIMO) system downlink signal processing joint optimization method based on intelligent reflecting surface (Reconfigurable Intelligent Surface, RIS), by the related parameters based on base station (Base Station, BS) end, user and RIS, RIS assisted beam space MIMO system reception signal model is constructed;According to the sparse characteristics of beam space channel and beam space channel power, a kind of beam selection algorithm is designed to generate beam selection matrix B, to select dominant beam;By centralized algorithm, maximum ratio combining optimization phase shift matrix is used;Then the design of precoding matrix is carried out to eliminate interference.The present application can greatly reduce system complexity, reduce the number of radio frequency chain, and can reduce hardware cost and energy loss.

Description

Technical Field

[0001] This invention relates to the field of wireless communication signal processing technology, and in particular to a joint optimization method for downlink signal processing in a beamspace MIMO system based on a smart reflector. Background Technology

[0002] Reconfigurable Intelligent Surfaces (RIS) represent a revolutionary and transformative technology for achieving low-cost, spectrum- and energy-efficient wireless communication. Specifically, an RIS consists of numerous low-cost passive components, each reflecting the incident signal with a specific phase shift, collaboratively achieving beamforming and interference suppression at one or more designated receivers. Existing technologies maximize the total received signal power at the user end by jointly optimizing active transmit beamforming at the base station (BS) and passive reflective beamforming at the RIS. This is achieved using a centralized algorithm based on semi-definite relaxation techniques, assuming known global channel state information at the RIS. However, in millimeter-wave massive MIMO systems, the sheer number of antennas at the base station, coupled with the equal number of radio frequency links, leads to extremely high hardware complexity, cost, and energy consumption, thus impacting the signal processing performance of the wireless communication system. Summary of the Invention

[0003] To address this, the present invention provides a joint optimization method for downlink signal processing in beamspace MIMO systems based on intelligent reflectors, which solves the problems of high complexity, high hardware cost, and high energy consumption in beamforming of existing millimeter-wave massive MIMO systems.

[0004] According to the design scheme provided by this invention, on the one hand, a joint optimization method for downlink signal processing in a beamspace MIMO system based on a smart reflector is provided, comprising:

[0005] Based on the relevant parameters of the BS, users and RIS, a beam spatial multiple input multiple output MIMO system received signal model is constructed. The relevant parameters include the channel matrix G between the RIS and the BS, the channel matrix H between the users and the RIS, the lens antenna array matrix U, and the phase shift matrix Θ and beam selection matrix B of the RIS.

[0006] Based on the sparsity characteristics and power of the beam space channel, a beam selection algorithm is designed to generate a beam selection matrix B to select the dominant beam; a phase shift matrix Θ is designed based on maximum ratio combining and using semidefinite relaxation techniques.

[0007] The BS precoding matrix is ​​designed based on the channel matrix G between the RIS and BS, the channel matrix H between the user and the RIS, the beam selection matrix B, the phase shift matrix Θ, and the array steering vector U. Interference between users is eliminated through the precoding matrix.

[0008] As a joint optimization method for downlink signal processing in beamspace MIMO systems based on intelligent reflectors according to the present invention, the received signal model of the beamspace multiple-input multiple-output MIMO system is further represented as follows: Where x represents the transmitted signal vector from the base station, and n represents the additional Gaussian noise vector. For beam space channels, and

[0009] As part of the joint optimization method for downlink signal processing in the beamspace MIMO system based on the intelligent reflector of this invention, a beam selection algorithm is further designed based on the sparsity characteristics and power of the beamspace channel, comprising:

[0010] First, the beam space channel power is calculated column by column using a preset sum of squares function;

[0011] Next, all the calculated powers in the beam space channel are sorted, and the top K powers are selected. RF (Number of RF connectors) sequence;

[0012] The dominant beam of the RIS is obtained based on the column index corresponding to the highest power sequence, and the beam selection matrix B is designed based on the dominant beam.

[0013] As a joint optimization method for downlink signal processing in a beamspace MIMO system based on a smart reflector according to the present invention, further comprising, based on maximum ratio combining and simultaneously utilizing semi-definite relaxation to design the phase shift matrix Θ, the method also includes:

[0014] First, global channel state information (CSI) is set to be valid at RIS. Using maximum ratio combining, the channel vector is projected onto the direction-matched phase matrix. With the goal of maximizing the total downlink sum rate for all users, an optimization problem model of the phase matrix is ​​constructed.

[0015] Then, the semidefinite relaxation technique is used to solve the optimization problem model of the phase matrix.

[0016] As a joint optimization method for downlink signal processing in beamspace MIMO systems based on intelligent reflectors according to the present invention, the optimization problem model of the phase matrix is ​​further expressed as follows: Where M is the number of users per antenna, h m Let θ be the channel between the m-th user and the RIS, N be the number of reflection units in the RIS, and θ be the channel between the m-th user and the RIS. nLet n be the phase shift of the nth reflecting unit in the RIS. Where θ n ∈[0,2π), β∈[0,1], Indicates the N×K corresponding to the selected beam RF Beam space channel matrix.

[0017] As a joint optimization method for downlink signal processing in a beamspace MIMO system based on a smart reflector, the BS-end precoding matrix designed based on the channel matrix G between the RIS and BS ends, the channel matrix H between the user and the RIS ends, the beam selection matrix B, the phase shift matrix Θ, and the lens antenna array matrix U is expressed as: V = H H Θ(UG) H B.

[0018] As a joint optimization method for downlink signal processing in a beamspace MIMO system based on a smart reflector according to the present invention, the received signal of the user obtained by optimizing the received signal model of the beamspace MIMO system through a precoding matrix is ​​further represented as follows: Where M is the number of users per antenna, and n m For each element in the additional Gaussian noise vector n in the received signal model of the beamspace MIMO system corresponding to the m-th user, Let ρ be the transmit BS precoding vector obtained from the estimated channel for the m-th user, and let ρ be the transmit power. Let ξ be the precoding matrix obtained from the estimated channel. j Let s be the channel estimation error matrix for user j. j This is for transmitting signals from the base station to user j.

[0019] The beneficial effects of this invention are:

[0020] This invention replaces the BS antenna in a millimeter-wave massive MIMO system with a lens antenna array. It selects the dominant beam in the millimeter-wave channel through a beam selection algorithm and maximizes the total received signal power at the user by using active and passive beamforming. Based on the joint optimization of the beam selection matrix, phase shift matrix and precoding matrix, it achieves good downlink and rate performance. With almost no performance loss, it can greatly reduce system complexity and the number of RF chains, thereby reducing hardware costs and energy consumption. Attached image description:

[0021] Figure 1 This is a schematic diagram of the joint optimization framework for downlink (DL) signals in a beamspace MIMO system based on a smart reflector, as illustrated in the embodiment.

[0022] Figure 2This is a schematic diagram of a beam space MIMO system assisted by a smart reflector in the embodiment;

[0023] Figure 3 This example illustrates the performance comparison of DL and rate versus DL signal-to-noise ratio (SNR) under different K values.

[0024] Figure 4 This example illustrates the performance comparison of DL and rate with DL SNR under different N values.

[0025] Figure 5 This example illustrates the performance comparison of DL and rate versus DL SNR under different phase shift matrices. Detailed implementation method:

[0026] To make the objectives, technical solutions, and advantages of this invention clearer and more understandable, the invention will be further described in detail below with reference to the accompanying drawings and technical solutions.

[0027] Addressing the issues of high complexity, high hardware cost, and high energy consumption in beamforming of existing millimeter-wave massive MIMO systems described in the background section, this invention provides a joint optimization method for downlink signal processing in beamspace MIMO systems based on intelligent reflectors. (See also...) Figure 1 As shown, it contains the following content:

[0028] S101. Construct a beam spatial multiple input multiple output MIMO system receiving signal model based on the relevant parameters of the BS, user and RIS. The relevant parameters include the channel matrix G between RIS and BS, the channel matrix H between user and RIS, the lens antenna array matrix U, and the phase shift matrix Θ and beam selection matrix B of RIS.

[0029] In this embodiment of the RIS-assisted beamspace MIMO system, the BS end can employ an antenna with K lens antennas and K... RF A uniform planar array of radio frequency chains simultaneously serves M single-antenna users. The RIS can be equipped with N reflector elements, each with its reflection coefficient individually adjustable. Since millimeter-wave links are highly susceptible to environmental congestion, it is assumed that the direct link between the BS and the users is blocked.

[0030] make This represents the channel between RIS and BS. h represents the channels between all M users and the RIS. m The size is N×1, representing the channel from the m-th user to the RIS, where m = 1, 2, ..., M. The channel G is represented using the widely used Saleh Valenzuela channel model, i.e., denoted as...

[0031]

[0032] Where L G This represents the number of paths between RIS and BS. and This represents the complex gain of path l1, including path loss, and the azimuth (elevation) angles at BS and RIS. and These represent the normalized array steering vectors associated with BS and RIS, respectively.

[0033] For a typical K1×K2 (K=K1×K2)UPA, It can be represented as

[0034]

[0035] in, It is the azimuth steering vector. It is the elevation steering vector, and then we define... Where λ is the carrier wavelength and d is the antenna spacing, typically satisfying d = λ / 2. The array steering vector of the RIS can also be represented in the same way, using N instead of K. Similarly, the channel h... m It can be represented in the following way

[0036]

[0037] In the formula L m This represents the number of paths between the m-th user and the RIS. This represents the complex gain (including path loss) and azimuth (elevation) of the l2 path, along with the RIS.

[0038] To reduce the number of RF links, i.e. to achieve K RF <<K, thereby reducing hardware costs and energy consumption, lens antenna arrays can be used at the BS end. Traditional wireless channels can be converted into beam space channels through lens antenna arrays, and such lens antenna arrays act as spatial DFT matrices. The function of this is as follows. The array steering vector, which contains K orthogonal directions (beams) covering the entire angular space, can be expressed as:

[0039]

[0040] Assume all operations are completed within the channel coherence time, during which the CSI remains constant. Define the RIS phase shift matrix as follows: The beam selection matrix is ​​B. The size of B is K×K. RF Its elements are either 0 or 1. Let θ = [θ1, ..., θ2]N ], Where θ n ∈[0,2π), β∈[0,1]. Received signals of M mobile users Represented as

[0041]

[0042] in Let be a vector of additive Gaussian noise with zero mean, and its covariance matrix be... Right now This represents a BS transmission signal with power ρ. This signal is composed of unit power complex-valued information symbols s selected from the discrete constellation set of the m-th user. m Multiply by the precoding vector p m Composition. Furthermore, A DL beam space channel matrix with K columns corresponding to K orthogonal beams is defined.

[0043] It is worth noting that, due to the limited number of dominant scatterers in the millimeter-wave propagation environment, the beam space channel... It possesses a sparse structure. Utilizing this sparsity characteristic of the beamspace channel, and through beam selection, a low-complexity beamspace precoder with near-optimal performance can be designed. The beam selection matrix B should satisfy... and Where k = 1, 2, ..., K, k RF =1,2,…,K RF To ensure RIS and K RF Different RF chain matches. [A] i,: Let [A] represent the i-th row of matrix A. :,j Let [A] represent the j-th column of matrix A. i,j This represents the element in the i-th row and j-th column of matrix A.

[0044] S102. Based on the sparsity characteristics and power of the beam space channel, a beam selection algorithm is designed to generate a beam selection matrix B to select the dominant beam; the phase shift matrix Θ is designed based on maximum ratio combining and using semi-definite relaxation.

[0045] Among them, based on the sparsity characteristics and power of the beam space channel, a beam selection algorithm is designed to generate the beam selection matrix B, which can be designed to include the following:

[0046] First, the beam space channel power is calculated column by column using a preset sum of squares function;

[0047] Next, all the calculated powers in the beam space channel are sorted, and the top K powers are selected. RF Sequence;

[0048] The dominant beam of the RIS is obtained based on the column index corresponding to the highest power sequence, and the beam selection matrix B is updated based on the dominant beam.

[0049] like Figure 2 The illustrated lens antenna array can convert a spatial channel into a beamspace channel. A maximum power (MP) beam selection scheme can be used to select K. RF Beams, specifically, utilize the characteristics of beam space channels. Specifically, the beam space channel matrix... It exhibits sparsity, with a few elements of the matrix having dominant values ​​near the line-of-sight propagation direction from BS to RIS.

[0050] Although RIS has K selectable beams, the number of dominant beams is only K. RF To reduce the effective channel dimension without significant performance loss, a beam selection scheme called the MP algorithm is designed. The algorithm process can be described as follows: First, through... Beam spatial channel calculation column by column The power, of which express The sum of squares of the elements in the k-th column vector. Then, we calculate the sum of squares of all... Sort in descending order and select the top K with the highest power. RF indivual Finally, based on the obtained top K RF indivual The corresponding column index design selects matrix B. After obtaining B, the received signal can be re-represented as...

[0051]

[0052] in Indicates the N×K corresponding to the selected beam RF Beam space channel matrix.

[0053] Among them, the phase shift matrix Θ, designed based on maximum ratio merging and using semi-definite relaxation, also includes:

[0054] First, global channel state information (CSI) is set to be valid at RIS. Using maximum ratio combining, the channel vector is projected onto the direction-matched phase matrix. With the goal of maximizing the total downlink sum rate for all users, an optimization problem model of the phase matrix is ​​constructed.

[0055] Then, the semidefinite relaxation technique is used to solve the optimization problem model of the phase matrix.

[0056] The goal of phase matrix estimation is to maximize the total depth and rate of all users by optimizing the reflected beamforming of the phase shifter at the RIS. We employ a centralized algorithm based on a positive semi-definite relaxation technique, assuming global CSI is available at the RIS. To simplify the design, specific phase matrices Θ are customized for different precoding matrices. Maximum ratio combining is used to project the channel vector onto the direction-matched phase matrix. The phase matrix optimization problem can be expressed as:

[0057]

[0058] By assuming that global CSI is available on RIS, a semidefinite relaxation technique can be applied to solve the problem (7). Then, a centralized algorithm is used to implement the problem. Let t = [t1,…,t N ] H ,in The constraint in equation (7) is equivalent to By applying variables in Therefore, problem (7) is equivalent to

[0059]

[0060] in, because in Therefore, problem (8) can be further expressed as

[0061]

[0062] in Define T = tt H T must satisfy the condition that it is positive semidefinite and rank(T) = 1. Since the rank-one constraint is non-convex, semidefinite relaxation can be further applied to relax this constraint. Therefore, problem (9) is simplified to

[0063]

[0064] It can be observed that problem (10) is a standard convex semidefinite programming problem. Therefore, it can be optimized and solved using convex optimization solvers such as CVX. In general, relaxing problem (10) may not lead to a rank-one solution, i.e., rank(T) ≠ 1, which means that the optimal objective value of (10) is only an upper bound of the objective value of (9). Therefore, an additional step is needed to construct a rank-one problem from the optimal solution of problem (10). Specifically, the eigenvalues ​​of T need to be decomposed into T = ΩΣΩ H , where Ω=[ω1,…,ω N ],Σ=diag([λ1,…,λ N]) are an identity matrix and a diagonal matrix, both of size N×N. Then, the suboptimal solution of (9) can be obtained as t′=ΩΣ 1 / 2 r, where It is based on The generated random vector. The objective value of (9) can be approximated by taking the maximum value of the best t obtained from a set of independently generated Gaussian random vectors r. Finally, the solution t of problem (8) can be obtained by t = e jarg(t′) recover.

[0065] S103. Based on the channel matrix G between the RIS and BS ends, the channel matrix H between the user and the RIS, the beam selection matrix B, the phase shift matrix Θ, and the array steering vector U, design the BS end precoding matrix to eliminate interference between users.

[0066] Using symbols and v m =[V H ] :,m The received signal of the m-th user, i.e., the m-th element of y, can be represented as:

[0067]

[0068] Where: n m Let n be the m-th element. The last two terms on the right-hand side of equation (11) represent interference plus noise, which is a variable with a mean of zero and a variance of . A random variable. It can be approximated as a variable with respect to s. m Irrelevant additive Gaussian noise.

[0069] Any channel estimator suffers from channel estimation error, which affects achievable deep learning and rate performance. The estimation error matrices of G and H are defined as follows: and The actual matrix V is

[0070]

[0071] in Using these formulas, the model for the received signal of the estimated channel is derived as follows:

[0072]

[0073] in The BS precoding vector is obtained based on the estimated channel. in, ξ j =[Ξ] j,: The first term in formula (13) is the desired signal, and the remaining terms are considered as interference plus noise. The power of the third term in (13) can be defined as...

[0074] The beam selection matrix B depends on the estimated channel; that is, B is actually obtained by substituting the actual channel with the estimated channel. The downlink DL achievable rate for the m-th user is calculated as follows:

[0075]

[0076] For perfect channel estimation, the downlink DL achievable rate for the m-th user is:

[0077]

[0078] To eliminate interference between different users, the ZF precoding scheme sets the precoding matrix as follows, given a channel estimate: Under ZF precoding, assuming the system has a known estimated channel, the downlink DL achievable rate for the m-th user can be expressed as:

[0079]

[0080] Under perfect CSI conditions, the achievable downlink DL rate for the m-th user is given by the following formula.

[0081]

[0082] To verify the effectiveness of this solution, the following explanation is based on experimental data:

[0083] Given the estimated CSI, simulation results verify the effectiveness of the proposed joint optimization method, demonstrating its ability to achieve good downlink DL and rate performance. When evaluating the implemented downlink DL and rate, unless otherwise specified, the system parameters are set as follows: K = 16, M = N = 8, L... G =8,K RF =L G And set the uplink SNR to 20dB.

[0084] like Figure 3 As shown, the relationship between achievable DL sum rate and DL SNR performance is presented for three different K values ​​(K∈{16,32,64}) and a known estimated CSI. It can be seen that the achievable DL sum rate increases with the number of BS antennas K. This is because the training overhead increases with increasing K, resulting in better channel estimation performance at larger K values, thus achieving a higher sum rate.

[0085] like Figure 4 As shown, when K=32, M=T=P=10, L GThe DL and rate performance are obtained under system settings of =10, N∈{10,12,14} and known CSI. It can be seen that at low DL SNR, given different N values, DL and rate are almost the same. However, when DL SNR is greater than 0dB, DL and rate performance deteriorate as N increases. This is because the number of channel coefficients to be estimated in G and H increases with N, leading to poorer channel estimation performance, which in turn deteriorates DL and rate performance.

[0086] like Figure 5 As shown, this study verifies the impact of phase shift matrix optimization on achievable deep learning (DL) and rate performance, and compares the differences in DL and rate performance between random and optimized (estimated) phase shift matrices. When the DL SNR is less than 0 dB, the two phase shift matrices exhibit almost identical sum and rate performance. However, when the DL SNR is greater than 0 dB, the DL sum and rate obtained by the optimized phase shift matrix are superior to those obtained by the random DFT phase shift matrix, further validating the necessity of phase shift matrix optimization.

[0087] The above data verifies that the joint optimization of beam selection, phase shift matrix design, and precoding design in the intelligent reflector-assisted beam space MIMO system can achieve good downlink and rate performance, reduce system complexity, hardware cost, and energy consumption, and facilitate the practical application of signal processing in wireless communication systems.

[0088] Unless otherwise specifically stated, the relative steps, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0089] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0090] The units and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations are not considered to be beyond the scope of this invention.

[0091] Those skilled in the art will understand that all or part of the steps in the above methods can be implemented by a program instructing related hardware, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk. Optionally, all or part of the steps in the above embodiments can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiments can be implemented in hardware or as a software functional module. This invention is not limited to any particular combination of hardware and software.

[0092] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A joint optimization method for downlink signal processing in a beamspace MIMO system based on a smart reflector, characterized in that, Include: A received signal model for a beamspace multiple-input multiple-output (MIMO) system assisted by a RIS is constructed based on relevant parameters of the base station (BS), users, and the intelligent reflector (RIS). The relevant parameters include the channel matrix between the RIS and the BS. Channel matrix between user and RIS Lens antenna array matrix , and the phase shift matrix of RIS and beam selection matrix ; Based on the sparsity characteristics and power of the beam space channel, a beam selection algorithm is designed to generate the beam selection matrix. To select the dominant beam, global channel state information (CSI) is set to be valid at the RIS. Maximum ratio combining is used to project the channel vector onto a direction-matched phase matrix. To maximize the total downlink sum rate for all users, an optimization problem model of the phase matrix is ​​constructed, and semi-definite relaxation techniques are used to optimize the phase matrix. The optimization problem model is used to solve the problem. Based on the sparsity characteristics and power of the beam space channel, a beam selection algorithm is designed, including: first, calculating the beam space channel power column by column using a preset sum of squares function; then, sorting all the calculated powers in the beam space channel and selecting the number of RFs in the front RF chain with the highest power. The RIS dominant beam is obtained based on the column index corresponding to the highest power sequence, and the beam selection matrix is ​​updated based on the dominant beam. ; Channel matrix between RIS and BS Channel matrix between user and RIS and beam selection matrix Phase shift matrix and array guide vector Design a precoding matrix for the BS side to eliminate interference between users.

2. The method for joint optimization of downlink signal processing in a beamspace MIMO system based on a smart reflector according to claim 1, characterized in that, The received signal model of a beamspace multiple-input multiple-output (MIMO) system is represented as follows: ,in, This represents the transmitted signal vector from the base station, with a power of... The signal transmitted by the base station is represented as The transmitted signal was sent from the first Select the complex value information symbol of unity power from the discrete constellation set of each user. Multiply by the precoded vector get, Number of users per antenna The number of radio frequency chains, This represents an additive Gaussian noise vector. For beam space channels, and Column correspondence The DL beam space channel matrix of orthogonal beams is represented as follows: , The number of reflective units provided for the RIS.

3. The method for joint optimization of downlink signal processing in a beamspace MIMO system based on a smart reflector according to claim 1, characterized in that, The optimization problem of the phase matrix is ​​represented by the following model: ,in, Number of users per antenna For the first Channel between users and RIS This represents the number of reflection units in the RIS. For RIS Phase shift of each reflecting unit, ,in , , Indicates the selected beam corresponding to Beam space channel matrix.

4. The method for joint optimization of downlink signal processing in a beamspace MIMO system based on a smart reflector according to claim 1, characterized in that, Channel matrix between RIS and BS Channel matrix between user and RIS and beam selection matrix Phase shift matrix and lens antenna array matrix The designed precoding matrix for the transmitting BS end is represented as follows: ,and , Indicates the first The received signal of each user Number of users per antenna This represents the number of radio frequency chains.

5. The method for joint optimization of downlink signal processing in a beamspace MIMO system based on a smart reflector according to claim 1, characterized in that, Interference between users is eliminated using a precoding matrix, and the user's received signal is represented as: ,in, The number of users per antenna. For the corresponding number Additional Gaussian noise vector in the received signal model of a beamspace MIMO system for a single user The elements in For the first The precoding vector of the transmitting base station corresponding to each user, obtained from the estimated channel. To transmit base station power, The precoding matrix is ​​obtained from the estimated channel. For users The channel estimation error matrix, To transmit from the base station to the user The transmitted signal.

Images