Hybrid beamforming method and system based on reconfigurable holographic surfaces and symbiotic radio networks

Abstract

This invention discloses a hybrid beamforming method and system for coexisting radio networks based on reconfigurable holographic surfaces (RHS). The method includes: constructing a communication architecture with the RHS as the transmit antenna to eliminate dual path loss; designing a dual-mode signal demodulation strategy. Strategy I employs MMSE-SIC combined with MRC technology to first demodulate the ambient signal, then removes the received signal from the SN via SIC, and demodulates the backscattered signal one by one. Strategy II designs the ambient signal and its transmission frame structure, combined with zero-forcing beamforming, to demodulate the backscattered signal, avoiding the propagation of ambient signal demodulation errors. The method optimizes transmit power using gradient descent, handles the non-convex constraints of digital beamforming using fractional programming, and optimizes the RHS amplitude control matrix using a sequential parameter convex approximation algorithm, achieving efficient alternating optimization of these three aspects. The advantages of this invention are: improved energy efficiency in multi-user scenarios, reduced system power consumption, support for dynamic network topologies while maintaining high reliability, and breakthroughs in the limitations of existing single-user solutions.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a hybrid beamforming method and system for a symbiotic radio network (SRN) based on a reconfigurable holographic surface (RHS). Background Technology

[0002] The surge in IoT connections has posed unprecedented challenges to the spectrum efficiency, access capabilities, and energy consumption of wireless communication systems. Against this backdrop, Symbiotic Radio Networks (SRNs), as a key spectrum-sharing paradigm, are widely considered a crucial technology driving 6G development because they allow backscatter devices (BDs) to multiplex ambient radio frequency signals for data transmission. However, existing SRN technologies suffer from a series of fundamental bottlenecks, severely limiting their practical deployment and application value.

[0003] The core limitation lies in multi-device access and interference management. Current mainstream solutions primarily serve single-active-user scenarios, neglecting the prevalence of concurrent access by multiple users and multiple BD devices in real-world networks. Under this complexity, environmental signals intertwine and superimpose with numerous BD reflected signals, forming an extremely complex multi-source interference environment, leading to signal overwhelming and a sharp deterioration in demodulation reliability. How to effectively support reliable multi-user, multi-BD communication in large-scale, densely deployed scenarios is a core challenge that existing technologies have yet to solve.

[0004] The existing application models of the key enabling technology "Reconfigurable Intelligent Surface" (RIS) have significant shortcomings:

[0005] 1. Signal path loss problem: When RIS is used as a passive relay, the signal needs to undergo two spatial propagation processes, incident and reflection. This inevitably introduces significant double path loss, which greatly restricts the effective range of signal coverage and the efficient use of energy.

[0006] 2. Hardware implementation bottleneck: Traditional RIS uses active phase shifters to achieve beam control, and its high static power consumption and dynamic switching power consumption seriously hinder the feasibility of RIS in battery-powered nodes or large-scale dense deployment scenarios.

[0007] 3. Insufficient adaptability of channel models: Existing RIS beamforming optimization algorithms are mainly designed based on simplified far-field channel models, which cannot accurately characterize the near-field propagation effects (such as non-planar wavefronts and spherical wave effects) in the environments where IoT devices are often located, resulting in significant deviations between actual performance and theoretical expectations.

[0008] The challenges at the algorithmic level are equally severe:

[0009] 1. Optimization Algorithm Scalability and Complexity: Traditional mathematical programming methods (such as those based on continuous convex approximation and Riemann optimization) used for joint optimization of transmitter beamforming and RIS phase suffer from a surge in optimization variables when dealing with dynamic and ever-changing multi-user scenarios. This leads to slow convergence speed, huge computational overhead, and difficulty in guaranteeing real-time performance. More importantly, these algorithms are difficult to effectively scale to cope with changes in topology and the addition of new devices.

[0010] 2. Vulnerability of Demodulation Methods: Mainstream BD signal demodulation mechanisms (such as SiC-based receivers) heavily rely on the accurate recovery and elimination of strong environmental signals. However, in real-world dynamic environments, even minute estimation errors in environmental signals can be amplified and propagated downwards, triggering cascading effects that ultimately lead to the failure of the entire demodulation process and system performance collapse.

[0011] The Challenges of Deep Learning Solutions: While deep learning (DL) technology has shown potential in dimensionality reduction and computational acceleration in recent years, its reliance on large amounts of fixed-scene data for offline training makes it difficult to flexibly adapt to the non-stationarity of the environment caused by dynamic changes in network topology (such as node movement, addition / exit). Insufficient generalization ability limits the robust application of DL in practical dynamic SRNs.

[0012] In summary, current SRN technology faces interconnected and mutually restrictive core contradictions: the contradiction between "the need for dense multi-device access" and "the inherent limitations of single-user support capabilities"; the contradiction between "the potential of RIS implementation and the demand for large-scale low-power deployment" and "its dual path loss, high hardware power consumption, and model mismatch"; the contradiction between "the need for real-time processing in complex environments" and "the computational complexity and low scalability of traditional optimization algorithms"; and the contradiction between "potential methods for improving computational efficiency" and "the insufficient generalization ability of DL in dynamic environments." These deep-seated contradictions collectively constitute a thorny technical dilemma of "multi-device access - insufficient single-user support capabilities - high power consumption - low real-time performance - poor dynamic adaptability." Overcoming this dilemma and achieving low-cost, high-reliability, low-power, scalable, and dynamically adaptable large-scale multi-user SRN is key to propelling this technology towards practical deployment. Summary of the Invention

[0013] To address the shortcomings of existing technologies, this invention provides a hybrid beamforming method and system for symbiotic radio networks based on reconfigurable holographic surfaces.

[0014] To achieve the above-mentioned objectives, the technical solution adopted by the present invention is as follows:

[0015] A hybrid beamforming method for symbiotic radio networks based on reconfigurable holographic surfaces includes:

[0016] Construct a coexisting radio network system comprising a transmitter Tx, a settling node SN, J single-antenna mobile users MU, and L single-antenna backscattering devices BD, wherein Tx is connected to N reconfigurable holographic surfaces RHS via K feeds. s A reflective element is connected, RHS serves as the transmitting antenna for Tx, and SN is equipped with N. r One antenna is used to demodulate the BD signal;

[0017] The transmit power p and the digital beamforming matrix are jointly optimized using an alternating optimization algorithm. and holographic beamforming matrix Maximize the system energy efficiency EE, where:

[0018]

[0019] Among them, P R =N s P s +KP r +P f P represents the operating power of the RHS. s P represents the power of the converter. r For the power of the RF chain, P f To control the power of the RHS control panel and other electronic components; w1 and w2 are preset weighting coefficients, R j R is the achievable rate of the j-th MU. SN P is the achievable rate of the backscattering system at SN. R p is the RHS operating power, and p is the transmit power;

[0020] The optimization process satisfies the constraint: p ≤ P t ,Tr{V H V} = 1, Among them, P t For the transmit power budget, The holographic beamforming matrix is ​​applied to the coordinates (n) of the reflecting element. x ,n y The amplitude control coefficient at point ().

[0021] Furthermore, it also includes Strategy I, which employs minimum mean square error-continuous interference cancellation to demodulate backscattered system signals in MMSE-SIC receivers, including:

[0022] At SN, the ambient signal stream s(n) is demodulated by the beamforming matrix W0 received via MMSE:

[0023]

[0024] in For direct link equivalent channel, For the backscattering link equivalent channel of the l-th BD, σ 2 Noise power, α l For BD reflectance, For N r ×N r The identity matrix;

[0025] After eliminating ambient signals, the BD signal is demodulated using maximum ratio combining (MRC) and MMSE-SIC. l .

[0026] Furthermore, the digital beamforming matrix V is optimized, including:

[0027] Introducing the auxiliary variable β j and ζ j By using fractional programming, the non-convex problem is transformed into a convex optimization problem;

[0028] The penalty function method and the sequential parametric convex approximation SPCA are used to handle the constraint Tr{Φ-vv H}≤0, where Φ=vv H v = vec(V);

[0029] Solve the transformed convex problem using convex optimization tools.

[0030]

[0031] Where μ1>0 is the penalty factor, v (t) Let r be the result of the t-th iteration. m ,r s It is an auxiliary variable for rate.

[0032] Furthermore, Strategy II also includes avoiding environmental signal demodulation through zero-forcing (ZF) beamforming and transmission frame structure design, as detailed below:

[0033] Design environment signal flow s(n) = [s a (n),s b (n)] H , where s b (n) is the pilot signal for which SN is known;

[0034] ZF beamforming is used to eliminate interference between the MU and BD:

[0035]

[0036] in H eq =[q1,…,q L ,h1,…,h L ] T M;

[0037] The transmission frame is divided into T1 and T2 phases: the T1 phase transmits s b (n) = 0, SN stores the direct link signal; non-zero s is transmitted in stage T2. b (n), SN eliminates direct links through differential reception.

[0038] Furthermore, the steps for optimizing the power allocation matrix P include:

[0039] Define τ MU =[τ1,…,τ J ] T ,τ BD =[τ1′,…,τ′ L ] T ;

[0040] Introducing auxiliary variables through fractional programming and The problem of maximizing energy efficiency is transformed into a problem concerning τ. MU and τ BD The convex optimization problem.

[0041] Furthermore, the optimization of the holographic beamforming matrix M includes:

[0042] Reorganize M = M0Θ, where Θ is a fixed phase shift matrix. This is the amplitude control matrix;

[0043] Transform the optimization problem of m into a convex problem:

[0044]

[0045] in, For amplitude control vector,

[0046] μ2>0 is the penalty factor, m (t) This is the result of the t-th iteration.

[0047] Furthermore, the channel modeling adopts a spherical wavefront model:

[0048] Line-of-sight channel:

[0049]

[0050] Where G is the gain of the receiving antenna. The coordinates are (n) x ,n y The distance d from the reflecting element to the receiving antenna. s Indicates the distance between RHS reflective elements. This indicates that the signal leaves the coordinate (n) x,n y The angle of departure (AoD) of the reflecting element;

[0051] Combined Channels:

[0052]

[0053] Where κ is the view distance weight. This refers to the non-line-of-sight channel component.

[0054] The present invention also discloses a communication system for implementing the method of any one of claims 1-7, comprising:

[0055] Reconfigurable holographic surface RHS, containing K feed sources and N s One reflective element;

[0056] Transmitter Tx, whose K RF chains are connected one-to-one with the RHS feed;

[0057] Settlement node SN, equipped with N r One antenna is used to demodulate the BD signal;

[0058] J single-antenna mobile user units (MU) and L single-antenna mobile user units (BD).

[0059] The present invention also discloses a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method as described in any one of claims 1-7.

[0060] Compared with the prior art, the advantages of the present invention are as follows:

[0061] 1. Significantly improve system energy efficiency: By jointly optimizing transmit power, digital beamforming and holographic beamforming, the system energy efficiency is improved by 0.6 bit / s / Hz / W when the number of RHS reflectors increases from 50 to 150 (compared to only 0.2 bit / s / Hz / W for the baseline solution);

[0062] 2. Overcomes the limitations of multiple users and multiple devices: Supports complex scenarios with J>3 users and L>2 BDs, solving the problem that existing technologies only support single users;

[0063] 3. Eliminate double path loss: Use the RHS directly as the transmitting antenna (non-RIS repeater) to avoid the signal experiencing two attenuations;

[0064] 4. Solving the problem of signal demodulation error propagation: Strategy II separates BD signals without demodulating environmental signals through innovative frame structure and zero-forcing beamforming;

[0065] 5. Reduced hardware power consumption and cost: RHS replaces the traditional RIS solution, eliminating the high power consumption problem of phase shifters;

[0066] 6. Improve optimization computation efficiency: Based on an alternating optimization algorithm of gradient descent, fractional programming, and SPCA;

[0067] 7. Precise channel modeling: The spherical wavefront model supports simultaneous far-field / near-field deployment, accurately characterizing channel features;

[0068] 8. Unleashing the degrees of freedom of holographic beams: By treating the RHS amplitude control matrix M0 as an independent optimization dimension, performance is improved by 0.4-0.6 bit / s / Hz / W compared to the fixed amplitude scheme;

[0069] 9. Enhanced practical deployment capabilities: Two demodulation strategies (Strategy I / II) are adapted to different scenario requirements. Strategy I is suitable for scenarios with a large number of BDs accessing the network, while the demodulation method of Strategy II is easier to implement than Strategy I, but is suitable for scenarios with fewer BDs. Attached Figure Description

[0070] Figure 1 This is a schematic diagram of the system model structure of an embodiment of the present invention;

[0071] Figure 2 This is a flowchart of the MMSE-SIC receiver according to an embodiment of the present invention;

[0072] Figure 3 This is a schematic diagram of the transmission frame structure of the environmental signal stream in Strategy II of this invention.

[0073] Figure 4 This is a graph showing the relationship between energy efficiency and RHS element number in Strategy I of this invention.

[0074] Figure 5 This is a graph showing the relationship between energy efficiency and the number of RHS components in embodiment II of the present invention. Detailed Implementation

[0075] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and examples.

[0076] This invention provides a hybrid beamforming method for symbiotic radio networks based on reconfigurable holographic surfaces, characterized by comprising:

[0077] Construct a coexisting radio network system comprising a transmitter Tx, a settling node SN, J single-antenna mobile users MU, and L single-antenna backscattering devices BD, wherein Tx is connected to N reconfigurable holographic surfaces RHS via K feeds. s A reflective element is connected, RHS serves as the transmitting antenna for Tx, and SN is equipped with N. r One antenna is used to demodulate the BD signal;

[0078] The transmit power p and the digital beamforming matrix are jointly optimized using an alternating optimization algorithm. and holographic beamforming matrix Maximize the system energy efficiency EE, where:

[0079]

[0080] Among them, P R =N s P s +KP r +P f P represents the operating power of the RHS. s P represents the power of the converter. r For the power of the RF chain, P f To control the power of the RHS control panel and other electronic components; w1 and w2 are preset weighting coefficients, R j R is the achievable rate of the j-th MU. SN P is the achievable rate of the backscattering system at SN. R p is the RHS operating power, and p is the transmit power;

[0081] The optimization process satisfies the constraint: p ≤ P t ,Tr{V H V} = 1, Among them, P t For the transmit power budget, The holographic beamforming matrix is ​​applied to the coordinates (n) of the reflecting element. x ,n y The amplitude control coefficient at point ().

[0082] The present invention will now be described in detail.

[0083] (1) System Model

[0084] The communication system model studied in this invention is as follows: Figure 1 As shown, the system includes a transmitter (Tx) and an N-type transmitter. r There are J sink nodes (SN) for the antennas, J single-antenna mobile users (MU), and L single-antenna BDs. Among them, there are K feeds and N... s The backscattered signal receiver (RHS) of each reflector element serves as the transmitting antenna for the signal transmission (Tx). Each feed of the RHS is wired to an RF chain within the Tx. Since the distance between the RF chain and the feed is extremely short, path loss in the cable between the RF chain and the feed can be ignored. The receiver (MU) needs to receive the information transmitted by the Tx, while the signal receiver (SN) is responsible for demodulating and storing the backscattered signal information, i.e., all the backscattered signals (BDs).

[0085] (2) Channel Model

[0086] Assume the channel coefficients between RHS and the j-th user, the l-th BD and SN are respectively and The channel coefficients between the l-th BD and SN and MU are respectively and Because of the high energy efficiency of RHS, large-scale RHS deployment is feasible, allowing receivers in the SRN to be located in either the far-field or near-field regions. Therefore, introducing a spherical wavefront channel model is essential for accurately characterizing the channel features. In this case, all line-of-sight channels related to RHS can be represented as...

[0087]

[0088] Where G is the gain of the receiving antenna. The coordinates are (n) x ,n y The distance d from the reflecting element to the receiving antenna. s Indicates the distance between RHS reflective elements. This indicates that the signal leaves the coordinate (n) x ,n y The angle of departure (AoD) of the reflecting element. Non-line-of-sight channels can be accurately characterized using the Rayleigh channel model, i.e. In summary, the coordinates are (n) x ,n y The channel from the reflector element to the receiver antenna can be represented as:

[0089]

[0090] Where κ is the weight, which is a constant. Other channels are characterized using the Rayleigh channel model, i.e. Among them, L p Indicates path loss. Path loss L p It can be modeled as L p =C0(d / D0) -υ Where C0 is the reference path loss coefficient, D0 is the reference distance, d is the distance between the transmitting antenna and the receiving antenna, and υ is the path loss exponent.

[0091] (3) Signal Model

[0092] Let the information transmitted by Tx to MU be s(n) = [s1(n), ..., s J (n)] H The information transmitted by the l-th BD is c lLet c = [c1, ..., c L ] H , and Therefore, the signal received by the l-th BD is in Indicates that the element The holographic beamforming matrix is ​​formed by V = [v1, ..., v J ] represents the digital beamforming matrix, where p satisfies p≤P t For the transmit power of Tx, P t Let SRN be the transmit power budget. At this point, the signal received at SN can be represented as...

[0093]

[0094] Where α l Let be the reflection coefficient of the l-th BD. This represents additive white Gaussian noise.

[0095] The received signal at the j-th MU can be expressed as

[0096]

[0097] in This represents additive white Gaussian noise. Here, let's assume...

[0098] (4) Strategy I

[0099] Strategy I aims to demodulate BD signals using a traditional SIC combined with MRC method. Since the direct link energy from the RHS is stronger than the backscattered signal in the signal received at SN, a Minimum Mean Square Error (MMSE)-SIC receiver is introduced. First, the environmental information stream s(n) is demodulated using multiple receiving antennas via MMSE beamforming. Then, the environmental signal is subtracted from the received signal at SN using SIC. At this point, the received signal at SN is the sum of all BD scattered signals and additive white Gaussian noise. Next, the BD signals corresponding to different environmental signal streams are combined using MRC, then demodulated using the environmental signal stream s(n), and finally, the MMSE-SIC receiver demodulates each BD signal individually.

[0100] make Based on the definition of MMSE and through calculation, the receiving beamforming matrix W0 of the demodulated ambient signal stream s(n) can be expressed as:

[0101]

[0102] After subtracting the direct link signal, the signal at SN can be represented as:

[0103]

[0104] Assuming the symbol period of BD is N times the ambient signal flow, combine all y(n), that is, let Y =

[0105] [y(1),…,y(N)], and through MRC, we have

[0106]

[0107] in make If z = vec(Z1), then the output signal can be expressed as:

[0108]

[0109] After the signal of the (l-1)th BD is demodulated and subtracted, the signal at SN can be expressed as:

[0110]

[0111] Let c demodulate the l-th BD signal. l The receiving beamforming vector is w l Then demodulate c l The signal-to-interference-plus-noise ratio (SINR) is expressed as:

[0112]

[0113] make This represents the interference and noise term. First, zero-phase component analysis (ZCA) is applied to separate the color noise. Whitening, then receiving the l-th BD signal c via beamforming demodulation using MRC. l At this point, according to MRC, the received beamforming vector can be expressed as:

[0114]

[0115] Substituting (11) into (10), the achievable rate of the backscattering system at SN can be expressed as:

[0116]

[0117] The complete flowchart of the MMSE-SIC receiver is shown in Figure 2.

[0118] In this invention, considering that the SRN operates in a symbiotic mode, i.e., the backscattering system provides an additional transmission path for the transmission of the environmental system, the received signal at the j-th MU can be rewritten as follows:

[0119]

[0120] in The achievable rate of receiving the signal at the j-th MU can be expressed as:

[0121]

[0122] Based on the above derivation, the energy efficiency of the SRN in this invention can be expressed as:

[0123]

[0124] Where P R =N s P s +KP r +P f P represents the operating power of the RHS. s P represents the power of the converter (diode, etc.). r For the power of the RF chain, P f To control the power of the control panel and other electronic components of the RHS; w1 and w2 are weights. Therefore, the energy efficiency optimization problem of the system can be expressed as:

[0125]

[0126] The problem is then solved by alternately optimizing the transmit power p, the digital beamforming matrix V, and the holographic beamforming matrix M (P1).

[0127] A) Optimize transmit power p

[0128] Given a digital beamforming matrix V and a holographic beamforming matrix M, let The optimization problem concerning the transmit power p can then be expressed as:

[0129]

[0130] in, λ j for The eigenvalues ​​of are given. It can be proven that the function F(p) is a unimodal function of the variable p in the above graph. Therefore, the global optimal solution to problem (PA.1) can be found using gradient descent. The specific algorithm steps are as follows:

[0131]

[0132] B) Optimize the digital beamforming matrix V

[0133] Let v = vec(V), in

[0134] Given the optimal transmit power p and the holographic beamforming matrix M, the optimization problem concerning the digital beamforming matrix V, transformed using fractional programming (P1), can be expressed as follows:

[0135]

[0136] in, The optimal values ​​of the two auxiliary vector variables introduced by fractional programming are

[0137]

[0138] Due to constraints and constraint v H v = 1 is a non-convex constraint, which we will now transform. It has been proven that the constraint Φ = vv H and Tr{Φ-vv H}≤0. Equivalent, therefore problem (PB.1) can be transformed into

[0139]

[0140] Then, the non-convex constraint Tr{Φ-vv} is handled using the penalty function method and SPCA. H Let μ1>0 represent the penalty factor, which is a constant, and v ≤ 0. (t) Let v be the result obtained after optimization in the t-th iteration. Then problem (PB.2) can be transformed into

[0141]

[0142] Problem (PB.3) is a convex optimization problem, which can be effectively solved using convex optimization tools such as CVX.

[0143] C) Optimize the holographic beamforming matrix M

[0144] To simplify the optimization process, firstly, the holographic beamforming matrix M is reorganized into M = M0Θ, where Let M0 represent the phase shift of the RHS reflector, which is a constant, and M0 represent the amplitude control matrix of the RHS reflector, which is the matrix variable to be optimized. make

[0145]

[0146] in, Similar to optimizing the digital beamforming matrix V, after sequentially applying fractional programming, a penalty function, and SPCA, the optimization problem for the holographic beamforming matrix M can be expressed as follows:

[0147]

[0148] in, ,m (t) Let m be the result obtained after the t-th iteration, and μ2>0 represent the penalty factor. Problem (PC.1) is a convex optimization problem, which can be effectively solved using convex optimization tools such as CVX.

[0149] The complete algorithm for Strategy I is summarized below:

[0150]

[0151] (5) Strategy II

[0152] Strategy II, by designing the information flow s(n) and its transmission frame structure of the environmental system, and combining ZF beamforming to eliminate interference between MU and BD, achieves the subtraction of the intelligent link from the received signal at SN using SIC without demodulating the environmental signal flow s(n). Let s a (n)=[s1(n),…,s J (n)] H s b (n)=[s′1(n),…,s′ L (n)] H s(n)=[s a (n); s b (n)] H , where the signal stream s a (n) Transmitted to MU, signal stream s b (n) is transmitted to BD, and its codebook is known to SN, much like the pilot signal used in channel estimation. Let the equivalent channel H... eq =[q1,…,q L ,h1,…,h L ] T M, then there is the definition of ZF beamforming, and digital beamforming can be expressed as

[0153]

[0154] in and P = diag{τ1,…,τ J ,τ′1,…,τ′ L} represents the power allocation matrix. The transmission frame structure design for the environmental signal stream s(n) is as follows:

[0155] like Figure 3 As shown, since ZF beamforming eliminates all interference between MU and BD, and the signal transmitted to BD is 0 in stage T1, only the direct link from RHS is received at SN. At this time, SN stores the received signal. In stage T2... b (n) When non-zero symbols begin to be transmitted, the information c of the l-th BD at this time l This will be superimposed on the environmental signal s′ l (n) is transmitted at SN. Let V = [V a V b ],in At this point, SN subtracts the signal stored in stage T1 from the currently received signal to obtain the following signal:

[0156]

[0157] Because of s b (n) is a known signal, and the channel coefficients are all known during the channel estimation stage. Therefore, the information of the backscatter system can be demodulated one by one by the MMSE-SIC receiver.

[0158] In phase T1, the reachable rate of the j-th MU can be expressed as:

[0159]

[0160] In phase T2, the reachable rates of the j-th MU and SN can be expressed as follows:

[0161]

[0162] Therefore, the energy efficiency of SRN can be expressed as

[0163]

[0164] Where δ represents the percentage of the transmission time of s(n) occupied by stage T1, i.e., T1 = δT s s. Since the duration of phase T1 can be very short, and the simulation results show... Therefore, the energy efficiency can be approximated as:

[0165]

[0166] Therefore, the energy efficiency maximization problem of the system can be expressed as:

[0167]

[0168] Next, the problem is solved by alternately optimizing the transmit power, digital beamforming matrix, and holographic beamforming matrix (P2).

[0169] A) Optimize transmit power p

[0170] make Given the digital beamforming matrix and the holographic beamforming matrix, problem (P2) can be transformed into the following optimization problem regarding the transmit power p:

[0171]

[0172] in λ j for The eigenvalues. Problem (P2A) is similar to problem (PA.1) in Strategy I and can be solved using Algorithm I.

[0173] B) Optimize the digital beamforming matrix V

[0174] Substituting (21) into problem (P2), given the transmit power and the holographic beamforming matrix, the problem of optimizing the digital beamforming matrix V can be transformed into optimizing the power allocation matrix P. Then, using fractional programming, problem (P2) can be transformed into...

[0175]

[0176] Among them, auxiliary variables introduced in fractional programming and The optimal value can be expressed as

[0177]

[0178] The remaining parameters in question (P2B) are shown below.

[0179]

[0180] C) Optimize the holographic beamforming matrix M

[0181] By transforming problem (P2) through fractional programming and then applying the penalty function method and SPCA, problem (P2) is transformed into...

[0182]

[0183] Where μ3>0 represents the penalty factor. The optimal values ​​of the two auxiliary variables introduced by the fractional programming are respectively

[0184]

[0185] The remaining parameters for the question (P2C) are as follows:

[0186]

[0187] The complete algorithm for Strategy II is summarized below:

[0188]

[0189]

[0190] Figure 4 Matlab simulations were used to obtain the relationship between the energy efficiency of Strategy I and the number of RHS elements under SRN with 3 users and different numbers of BDs. The graph shows the SRN energy efficiency on the left and the number of RHS reflectors on the right. In the legend, 'Proposed' represents the result obtained using Algorithm II, and 'Benchmark' represents the result of setting the RHS amplitude control matrix M0 to an identity matrix and optimizing only transmit power and digital beamforming, i.e., the RHS is used as a traditional antenna array. The graph shows that the energy efficiency obtained by the proposed Algorithm II is always higher than the benchmark energy efficiency. When the number of RHS reflectors increases from 50 to 150, the system energy efficiency improves by 0.6 bit / s / Hz / W for different numbers of BDs, while the benchmark only increases by about 0.2 bit / s / Hz / W. Furthermore, the gap between them gradually widens as the number of beamforming elements (BDs) increases. When the SRN contains two BDs, the energy efficiency obtained by Algorithm 2 is approximately 0.4 bit / s / Hz / W higher than the baseline, while when the number of BDs is four, the difference reaches approximately 0.6 bit / s / Hz / W. This proves the effectiveness of Algorithm 2 and also illustrates the necessity of optimizing RHS holographic beamforming.

[0191] Figure 5 The diagram illustrates the relationship between system energy efficiency and RHS element number obtained using Strategy II under different numbers of MUs and BDs in the SRN. Figure 4 Similarly, in the figure, the SRN energy efficiency is on the left and the number of RHS reflective elements on the right. In the legend, 'Proposed' represents the result obtained using Algorithm 3, and 'Benchmark' represents the result of setting the RHS amplitude control matrix M0 to an identity matrix and optimizing only transmit power and digital beamforming, i.e., the RHS is used as a traditional antenna array. The figure shows that the result obtained using Algorithm 3 is always inferior to the benchmark result, and the difference between them increases with the number of RHS elements, indicating that optimizing holographic beamforming is essential.

[0192] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory). This computer-readable storage medium is a memory device in a terminal device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and extended storage media supported by the terminal device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, this storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device.

[0193] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the hybrid beamforming method for symbiotic radio networks in the above embodiments; one or more instructions in the computer-readable storage medium are loaded and executed by a processor.

[0194] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0195] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, systems, and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0196] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0197] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0198] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the implementation methods of the present invention, and should be understood that the scope of protection of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of the present invention.

Claims

1. A hybrid beamforming method for symbiotic radio networks based on reconfigurable holographic surfaces, characterized in that, include: Construct a coexisting radio network system comprising a transmitter Tx, a settling node SN, J single-antenna mobile users MU, and L single-antenna backscattering devices BD, wherein Tx is connected to N reconfigurable holographic surfaces RHS via K feeds. s A reflective element is connected, RHS serves as the transmitting antenna for Tx, and SN is equipped with N. r One antenna is used to demodulate the BD signal; The transmit power is jointly optimized using an alternating optimization algorithm. Digital beamforming matrix and holographic beamforming matrix Maximize system energy efficiency ,in: ， in, Indicates the operating power of the RHS. Indicates the power of the converter. For the power of the RF chain, To control the power of the control panel and other electronic components of the RHS; and For preset weighting coefficients, For the first The achievable rate of a MU, The achievable rate of the backscattering system at SN is For RHS operating power, This refers to the transmission power. The optimization process satisfies the following constraints: , , ,in, For the transmit power budget, The holographic beamforming matrix is ​​applied to the coordinates of the reflecting element ( Amplitude control coefficient at () position; The hybrid beamforming method for symbiotic radio networks also includes Strategy I and Strategy II; Strategy I employs minimum mean square error-continuous interference cancellation to demodulate backscattered system signals in the MMSE-SIC receiver; Strategy II avoids environmental signal demodulation through zero-forcing ZF beamforming and transmission frame structure design.

2. The method according to claim 1, characterized in that, Strategy I employs minimum mean square error-continuous interference cancellation (MMSEC) to demodulate backscattered system signals in an MMSE-SIC receiver, including: Receive beamforming matrix via MMSE at SN. Demodulating ambient signal stream : ， in For direct link equivalent channel, For the backscattering link equivalent channel of the l-th BD, Noise power, For BD reflectance, For N r ×N r The identity matrix; After eliminating ambient signals, the BD signal is demodulated using maximum ratio combining (MRC) and MMSE-SIC. .

3. The method according to claim 2, characterized in that, Optimize digital beamforming matrix ,include: Introducing auxiliary variables and By using fractional programming, the non-convex problem is transformed into a convex optimization problem; Constraints are handled using the penalty function method and the sequential parameter convex approximation SPCA. ,in = , =vec(V); Solve the transformed convex problem using convex optimization tools: ， Where μ1>0 is the penalty factor, v (t) Let r be the result of the t-th iteration. m ,r s It is an auxiliary variable for rate.

4. The method according to claim 1, characterized in that, Strategy II avoids environmental signal demodulation through zero-forcing ZF beamforming and transmission frame structure design, as detailed below: Design environment signal flow ,in The pilot signal is known to SN; ZF beamforming is used to eliminate interference between the MU and BD: ， in , , ; The transmission frame is divided into T1 and T2 phases: T1 phase transmission =0, SN stores the direct link signal; Non-zero transmission during T2 phase The SN eliminates direct links through differential reception.

5. The method according to claim 4, characterized in that, The steps to optimize the power allocation matrix P include: definition ; Introducing auxiliary variables through fractional programming and This transforms the problem of maximizing energy efficiency into a question about and The convex optimization problem.

6. The method according to claim 1, characterized in that, The optimization of the holographic beamforming matrix M includes: Reorganization , For a fixed phase shift matrix, This is the amplitude control matrix; Will The optimization problem is transformed into a convex problem: ， in, For amplitude control vector, μ2>0 is the penalty factor, m (t) This is the result of the t-th iteration.

7. The method according to claim 1, characterized in that, Channel modeling uses a spherical wavefront model: Line-of-sight channel: ， in, For the gain of the receiving antenna, Indicates coordinates as The distance from the reflective element to the receiving antenna, Indicates the distance between RHS reflective elements. This indicates that the signal leaves the coordinates. The departure angle of the reflective element; Combined Channels: ， in, For line-of-sight weighting, This refers to the non-line-of-sight channel component.

8. A communication system implementing the method of any one of claims 1-7, characterized in that, include: Reconfigurable holographic surface RHS, containing K feed sources and N s One reflective element; Transmitter Tx, whose K RF chains are connected one-to-one with the RHS feed; Settlement node SN, equipped with N r One antenna is used to demodulate the BD signal; J single-antenna mobile user units (MU) and L single-antenna mobile user units (BD).

9. A computer-readable storage medium storing a computer program, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-7.

Images