Intelligent reflector-assisted satellite-ground fusion network interference suppression method

Abstract

The invention relates to a satellite-ground fusion network interference suppression method assisted by an intelligent reflecting surface. The method comprises the following steps: respectively establishing channel models of a direct link from a satellite to a user, a direct link from a base station to the user, an intelligent reflecting link from the satellite to the user and an intelligent reflecting link from the base station to the user; according to the propagation path of the signal expected by the user and the propagation path of the interference signal, a function representation of the signal received by the user is established by adopting a corresponding channel model, and then a signal to interference plus noise ratio based on statistical channel expectation is calculated; and constructing an optimization problem meeting the condition of the minimum signal to interference plus noise ratio, solving the optimization problem to obtain an optimal base station beam forming matrix and an optimal intelligent reflecting surface shift matrix which enable the base station to provide services for all users to have the minimum total energy consumption, and dynamically adjusting the beam form and the intelligent reflecting surface phase of the base station. The time-varying interference caused by high-speed movement of the satellite can be effectively solved, and the power consumption of the base station is reduced on the premise of ensuring the user service quality.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a smart reflector-assisted method for suppressing interference in satellite-ground fusion networks. Background Technology

[0002] Low Earth orbit satellite communications offer unique advantages by providing wide-area coverage in areas where traditional terrestrial networks are too costly or geographically impractical, such as rural areas, ships at sea, and aircraft. However, inherent limitations, including intermittent coverage windows and the propagation characteristics of satellite links, make it impractical to rely solely on satellite networks to serve all end users. Furthermore, the high cost of satellites necessitates that they not be limited to serving only a few devices.

[0003] Given the strong complementarity between terrestrial and satellite communication systems in terms of coverage and service provision, space-ground integration, as a promising paradigm for leveraging their respective strengths, has attracted increasing research interest. However, the exponential growth in service demand has intensified spectrum competition between satellite and terrestrial systems, making spectrum scarcity a key bottleneck in the evolution of integrated space-ground networks. Spectrum sharing mechanisms based on exclusion zones (EZs) are a fundamental method for suppressing co-channel interference (CCI) in satellite-terrestrial integrated networks. This method ensures reliable satellite communication by dynamically isolating terrestrial transmission in critical areas, thereby reducing the interference-to-noise ratio (INR) below a threshold. However, due to trade-offs in its implementation, it tends to favor the normal operation of satellite terminals. To ensure this, the actual size of the isolation zone significantly reduces the available resources of the terrestrial system, thus lowering spectrum efficiency. This contradicts the goals of a space-terrestrial integrated network. Furthermore, since the exclusion zone needs to be set according to the satellite's elevation angle relative to the ground, the computational complexity is difficult to manage in real time due to the rapid movement of satellites, placing significant pressure on terrestrial network scheduling. Database-based approaches utilize ground stations within satellite systems to obtain beam coverage information from satellite broadcasts, forming a database. Ground cellular systems report their physical locations to this database to determine maximum available transmit power and frequency bands, optimizing terminal probability and cellular throughput for satellite ground stations. However, with large-scale LEO satellite networks, the high-speed movement of these systems necessitates frequent database updates, and base stations incur significant overhead to handle these database changes and their resulting policy adjustments. To further improve the flexibility of spectrum resource utilization, dynamic spectrum sharing schemes based on cognitive radio have gained increasing attention in recent years. However, in scenarios with large-area LEO satellite network coverage, this approach also faces challenges such as inaccurate channel state information and spectrum sensing results, and a lack of information exchange between the primary and secondary systems. These issues pose significant challenges to meeting the differentiated service needs of satellite-ground systems.

[0004] Smart reflectors (RIS) are an emerging technology capable of tunable anomalous scattering of incident electromagnetic waves. This characteristic allows for active control of the propagation environment, introducing an additional dimension of optimization to wireless communication networks. While RIS have demonstrated good performance across various communication domains, their application in satellite-to-ground interference coordination remains to be explored. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an intelligent reflector-assisted interference suppression method for satellite-ground fusion networks, which can effectively solve the time-varying interference caused by high-speed satellite movement and reduce base station power consumption while ensuring user service quality.

[0006] The technical solution adopted by this invention to solve its technical problem is: to provide a smart reflector-assisted satellite-ground fusion network interference suppression method, applied to a communication network including satellites, base stations, smart reflectors, and several users, comprising the following steps:

[0007] Channel models for the direct link from satellite to user, the direct link from base station to user, the intelligent reflection link from satellite to user, and the intelligent reflection link from base station to user are established respectively.

[0008] Based on the propagation path of the user's desired signal and the propagation path of the interference signal, a functional representation of the user's received signal is established using the corresponding channel model, and then the signal-to-interference-plus-noise ratio based on the statistical channel expectation is calculated.

[0009] An optimization problem is constructed that satisfies the minimum signal-to-interference-plus-noise ratio constraint. Solving this optimization problem yields the optimal base station beamforming matrix and the optimal smart reflector phase shift matrix that minimize the beamforming power consumption of the base station in providing services to all its users.

[0010] Based on the optimal base station beamforming matrix and the optimal smart reflector phase shift matrix, the base station beam shape and the smart reflector phase are dynamically adjusted.

[0011] Furthermore, the minimum signal-to-interference-plus-noise ratio constraint includes:

[0012] Signal-to-interference-plus-noise ratio (SINR) of base station user received signals k Not less than the set threshold γ UE The desired signal for the base station user comes from the base station, while the interference signal comes from the satellite and other base station users;

[0013] Signal-to-interference-plus-noise ratio (SINR) of satellite user received signals su Not less than the set threshold γ SU The satellite user's desired signal comes from the satellite, while the interference signal comes from the base station.

[0014] Furthermore, the signal-to-interference-plus-noise ratio (SINR) of the signal received by the user at the base station k Represented as

[0015]

[0016] Among them, G S This represents the equivalent baseband channel from the satellite to the smart reflector. G represents the equivalent baseband channel from the satellite to the k-th base station user. B Represents the equivalent baseband channel from the base station to the smart reflector, h k This represents the equivalent baseband channel from the base station to the base station user. This represents the equivalent baseband channel from the smart reflector to the base station user, where P1 and P2 represent the signal power transmitted by the base station and the satellite, respectively. k w represents the beamforming vector of the k-th base station user served by the base station. m W represents the beamforming vector of the m-th base station user served by the base station. sat The beamforming matrix of the satellite is represented by Φ, the phase shift matrix of the smart reflector is represented by a. k The value σ indicates whether the satellite's direct path is blocked, taking values ​​of {0,1}. 2 Indicates noise.

[0017] Furthermore, the signal-to-interference-plus-noise ratio (SINR) of the satellite user received signal. su Represented as

[0018]

[0019] Among them, G S G represents the equivalent baseband channel from the satellite to the smart reflector. B Represents the equivalent baseband channel from the base station to the smart reflector, h su This represents the equivalent baseband channel from the base station to the satellite user. This represents the equivalent baseband channel from the smart reflector to the satellite user, where P1 and P2 represent the signal power transmitted by the base station and the satellite, respectively. k w represents the beamforming vector of the k-th base station user served by the base station. m W represents the beamforming vector of the m-th base station user served by the base station. sat Φ represents the beamforming matrix of the satellite, Φ represents the phase shift matrix of the smart reflector, g represents the direct channel from the satellite to the satellite user, and σ represents the beamforming matrix of the satellite. 2 Indicates noise.

[0020] Furthermore, solving the optimization problem includes:

[0021] Keeping the phase shift matrix of the intelligent reflector unchanged, and based on the duality of the constraints and the phase rotation invariance of the absolute value of the inner product of the beamforming vectors, the optimization problem is transformed into a second-order cone optimization problem.

[0022] Solving the second-order cone optimization problem yields the optimal estimate of the base station beamforming matrix;

[0023] Using the optimal estimate of the calculated base station beamforming matrix, the optimization problem is transformed into a first optimization sub-problem for the phase shift matrix of the smart reflector.

[0024] Solving the first optimization subproblem yields the optimal estimate of the phase shift matrix of the intelligent reflector.

[0025] The beamforming power consumption of the base station is calculated using the optimal estimation of the base station beamforming matrix and the optimal estimation of the phase shift matrix of the intelligent reflector.

[0026] If the decrease in beamforming power consumption in the current round compared to the previous round is lower than a set threshold, then the optimal estimate of the current base station beamforming matrix and the optimal estimate of the smart reflector phase shift matrix are output as the optimal base station beamforming matrix and the optimal smart reflector phase shift matrix. Otherwise, the smart reflector phase shift matrix is ​​set to the optimal estimate of the current smart reflector phase shift matrix, and the process returns to the step of setting the smart reflector phase shift matrix unchanged.

[0027] Furthermore, solving the first optimization sub-problem includes:

[0028] By introducing the difference between the signal-to-interference-plus-noise ratio (SIR) of the user's received signal and its set threshold as an auxiliary variable, the first optimization subproblem is transformed into a second optimization subproblem that maximizes the sum of the residuals between the SIR of all user received signals and their set thresholds.

[0029] By employing the SDP relaxation method, the objective function and constraints of the second optimization subproblem are equivalently replaced to obtain the third optimization subproblem;

[0030] Solving the third optimization subproblem yields the optimal estimate of the phase shift matrix of the intelligent reflector.

[0031] Furthermore, when solving the third optimization subproblem, a method is adopted to mix the eigenvector of the largest eigenvalue with Gaussian random noise vector of the same dimension, so as to restore the calculated augmented matrix to the optimal estimate of the phase shift matrix of the intelligent reflector.

[0032] Furthermore, when modeling the channel for the direct link from satellite to user, the direct link from base station to user, and the direct link from smart reflector to user, Ricean fading is used to establish the corresponding models.

[0033] Furthermore, when modeling the channel for the direct link from the satellite to the smart reflector and the direct link from the base station to the smart reflector, the multipath effect is ignored.

[0034] Furthermore, the gain of the reflection amplitude of all reflective elements in the smart reflective surface remains unchanged.

[0035] Beneficial effects

[0036] By adopting the above-mentioned technical solutions, this invention has the following advantages and positive effects compared with the prior art: This invention achieves dynamic suppression of time-varying interference by introducing intelligent reflector technology, solving the problem of time-varying interference characteristics caused by satellite orbital motion and compensating for the shortcomings of traditional static interference coordination schemes in adapting to rapid topology changes; Furthermore, this invention reduces base station power consumption by 6-8 dB while ensuring user service quality by jointly optimizing base station beamforming and RIS phase offset, solving the drawback of existing schemes requiring significant increases in base station transmission power to address the near-far effect caused by power asymmetry between satellites and ground terminals; Simultaneously, this invention uses a threshold-based method to ensure generalization while maintaining performance by maximizing residual SINR in sub-problems, solving the generalization problem of existing processing methods requiring frequent changes in parameter selection based on the environment; This invention ensures rapid performance improvement during alternating optimization by employing an alternating optimization approach combined with residual SINR sub-problem design. Attached Figure Description

[0037] Figure 1 This is a flowchart of an embodiment of the present invention;

[0038] Figure 2 This is a schematic diagram of the communication network structure according to an embodiment of the present invention;

[0039] Figure 3 This is a flowchart of the optimization problem solving process according to an embodiment of the present invention. Detailed Implementation

[0040] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined by the appended claims.

[0041] The embodiments of this invention relate to a smart reflector (RIS)-assisted interference suppression method for satellite-to-ground fusion networks. This method involves introducing RIS to construct multiple reflection links and dynamically adjusting the RIS phase to achieve interference cancellation. For example... Figure 1 As shown, the specific steps include:

[0042] Channel models for the direct link from satellite to user, the direct link from base station to user, the intelligent reflection link from satellite to user, and the intelligent reflection link from base station to user are established respectively.

[0043] Based on the propagation path of the user's desired signal and the propagation path of the interference signal, a functional representation of the user's received signal is established using the corresponding channel model, and then the signal-to-interference-plus-noise ratio based on the statistical channel expectation is calculated.

[0044] An optimization problem is constructed that satisfies the minimum signal-to-interference-plus-noise ratio (SINR). Solving this optimization problem yields the base station beamforming matrix and the smart reflector phase shift matrix that minimize the total energy consumption of the base station in providing services to all its users.

[0045] Based on the calculated base station beamforming matrix and smart reflector phase shift matrix, the base station beam shape and smart reflector phase are dynamically adjusted to suppress interference signals and reduce base station power consumption.

[0046] Users can be categorized into base station users (UEs) and satellite users (SUs). Base station users expect signals from the base station and interference signals from the satellite, while satellite users expect signals from the satellite and interference signals from the base station. The received signal models for base station and satellite users can be established as the superposition of the expected and interference signals for the corresponding links, and an optimization problem under constraints can be constructed. The specific constraints include:

[0047] Signal-to-interference-plus-noise ratio (SINR) of base station user received signals k Not less than the set threshold;

[0048] Signal-to-interference-plus-noise ratio (SINR) of satellite user received signals su Not less than the set threshold.

[0049] like Figure 2 The image shown is an embodiment 1 of this implementation, illustrating a space-ground converged network based on LEO satellites. More specifically, it considers a downlink spectrum sharing scenario between a terrestrial base station system and an LEO satellite communication system. The system architecture includes configuration N. t A multi-antenna base station with 1 transmit antenna (serving 1K single-antenna users within its coverage area), configured with N satA single-antenna LEO satellite system (serving users over a wide coverage area) can also have satellite users as single-antenna devices. In this configuration, terrestrial base station users receive desired signals from the base station and interference signals from the satellite, just as satellite users receive desired signals from the satellite and interference signals from the base station. This mutual interference fundamentally reduces the signal-to-interference-plus-noise ratio (SINR) for both types of terminals, impairing QoS metrics in spectrum-sharing systems. Furthermore, the interaction between urban structures and specific satellite elevation / azimuth geometry induces multipath effects through signal scattering mechanisms, causing compound channel impairments that require joint mitigation strategies.

[0050] In this embodiment, channel models for three key propagation paths are established, namely:

[0051] Direct links from LEO satellites to users;

[0052] A direct link from the browser / server (BS) to the user;

[0053] RIS-assisted reflection path (including smart reflection links from LEO satellites to users and smart reflection links from BS to users).

[0054] The equivalent baseband channels from LEO satellite to RIS, from LEO satellite to the k-th UE, from BS to RIS, UE and SU, and from RIS to base station user UE and satellite user SU are represented as follows:

[0055] Satellite-to-ground channels are primarily modeled using free-space path loss (FSPL) and additional atmospheric attenuation. However, due to multipath effects induced by signal scattering mechanisms, they are modeled using Ricean fading, similar to terrestrial channels, to account for the dominant LOS component.

[0056]

[0057] The parameters involved are defined as follows: β1, β2, β3 represent Ricean channel factors, μ k , This represents the large-scale path loss coefficient for each channel. The normalized line-of-sight component phase information for the corresponding channel can be directly obtained through spatial angle and position information. This represents the non-line-of-sight Rayleigh fading component, which is an independent and identically distributed complex Gaussian random variable with zero mean and unit variance.

[0058] The multipath effects of the channels from ground base stations to the RIS and from satellites to the RIS are almost negligible due to altitude. Therefore, the modeling only considers the phase difference caused by the placement angle and azimuth, as follows:

[0059]

[0060] Where, μ GS μ represents the fading in the direct channel from the satellite to the RIS. GB This indicates the fading in the direct channel from the base station to the RIS. This represents the phase difference vector formed by the departure angle on the satellite antenna array. θ represents the phase difference vector formed by the departure angle on the base station antenna array. AOA θ represents the off-axis angle between the off-axis direction and the frontal reference direction. AOD φ represents the off-axis angle between the direction of incoming wave and the reference direction of the front surface. AOA φ represents the azimuth angle between the direction of departure from the wave and the reference direction of the front surface. AOD This indicates the azimuth angle between the direction of incoming wave and the reference direction of the front surface.

[0061] G S G B The phase matrix, obtained from the angle of arrival (AOA) and the angle of departure (AOD), along with large-scale fading parameters, is used, ignoring multipath effects. A similar method can be used to obtain the channel information for satellite users, which will not be elaborated upon in this implementation.

[0062] To represent the received signal in this situation, set... This represents the beamforming vector of the base station serving the k-th base station user. Therefore, the complex baseband transmitted signal from this base station can be expressed as: Where x k Let represent the expected signal of the k-th base station user with zero mean and specific power as independent variables. Let the reflection matrix of RIS be Φ = diag{φ1,φ2,...φ M}, where φ i Let |φ| represent the reflection coefficient of the RIS, where the gain of the reflection amplitude of all reflecting elements is set to the maximum value of 1 (i.e., only phase adjustment, no amplitude change), i.e., |φ| i |=1,i=1,2,...,M. Therefore, the superimposed signal received by the user at the k-th base station can be expressed as:

[0063]

[0064] in W represents the additive white Gaussian noise (AWGN) at the k-th UE. sat The beamforming of the LEO satellite is represented by s, which represents the transmitted signal of the LEO satellite with zero mean and specific power.

[0065] Similarly, the superimposed signal received by a satellite user can be represented as follows:

[0066]

[0067] The proposed system is driven by three requirements: 1) reducing parameter configuration sensitivity, 2) ensuring communication continuity under QoS degradation thresholds caused by interference, and 3) maintaining energy-efficient operation. We have established the following constrained optimization framework.

[0068]

[0069] Where SINR is the expectation based on the statistical channel representation, SINR k and γ UE SINR represents the signal-to-interference-plus-noise ratio (SINR) of the base station user UE and its threshold, respectively. su and γ SU These represent the signal-to-interference-plus-noise ratio (SIR) and its threshold for the base station user SU, respectively.

[0070] Specifically, it is expressed as follows:

[0071]

[0072]

[0073] Where P1 and P2 represent the signal power transmitted by the base station and the LEO satellite.

[0074] This optimization is constrained by three factors: 1) There are a large number of internal constraints, and the constraints are in the form of difficult-to-handle fractions. 2) Multiple optimization elements are explicitly and implicitly coupled in the expression, making direct solution impossible. 3) The modulus constraint characteristics of RIS make the solution difficult.

[0075] To address the three factors mentioned above, this implementation method employs an alternating optimization approach to divide the optimization problem into two sub-problems. Base station beamforming and RIS phase shift control are handled separately to decouple the original optimization. Different optimization objectives, both beneficial to system performance, are set for each sub-problem, maintaining their inherent coordination and consistency. This iterative refinement effectively resolves the variable coupling dilemma. Specifically, the following steps are included:

[0076] By setting the phase shift matrix of the intelligent reflector to remain unchanged, and based on the duality of the constraints and the phase rotation invariance of the absolute value of the inner product of the beamforming vectors, P1 is transformed into a second-order cone optimization problem P2.

[0077] Solving this second-order cone optimization problem yields the optimal estimate of the base station beamforming matrix;

[0078] Using the optimal estimate of the calculated base station beamforming matrix, P1 is transformed into the first optimization subproblem P3 for the phase shift matrix of the smart reflector.

[0079] Solve the first optimization subproblem P3 to obtain the optimal estimate of the phase shift matrix of the intelligent reflector;

[0080] The beamforming power consumption of the base station is calculated by using the optimal estimation of the current base station beamforming matrix and the optimal estimation of the phase shift matrix of the intelligent reflector.

[0081] If the decrease in beamforming power consumption in the current round compared to the previous round is less than a set threshold, then the optimal estimate of the current base station beamforming matrix and the optimal estimate of the smart reflector phase shift matrix are output as the optimal base station beamforming matrix and the optimal smart reflector phase shift matrix; otherwise, the process returns to the step of setting the smart reflector phase shift matrix unchanged.

[0082] More specifically, the following method is used to solve the first optimization subproblem P3:

[0083] By introducing the difference between the signal-to-interference-plus-noise ratio (SIR) of the user's received signal and its set threshold as an auxiliary variable, the first optimization subproblem P3 is transformed into a second optimization subproblem P4 that maximizes the sum of the residuals between the SIR of all user received signals and their set thresholds.

[0084] By using the SDP relaxation method, the objective function and constraints of the second optimization subproblem P4 are equivalently replaced to obtain the third optimization subproblem P5.

[0085] Solve the third optimization subproblem P5 to obtain the optimal estimate of the phase shift matrix of the intelligent reflector.

[0086] like Figure 3 The following is another specific embodiment 2 of this implementation method, which further describes the specific calculation process based on embodiment 1:

[0087] Step 1 (Transformation of the Second-Order Cone Programming Problem and Parameter Calculation): Statistical channel information was used as a parameter in Example 1 to improve the algorithm's generalization ability. During the calculation, its expected value needs to be calculated. Simultaneously, with the RIS phase shift matrix fixed, the following parameters can be obtained:

[0088]

[0089] For satellite user E su The expression is similar to the formula above, but for the sake of brevity, its specific expression will not be described in detail.

[0090] At the same time This parameter achieves the effect of an equivalent channel. Based on the duality of the constraints, the original optimization formula is transformed into a second-order cone programming problem. Through the reformulation of the norm-based objective function and constraints, combined with the phase rotation invariance of the absolute value of the inner product involving the beamforming vectors, the optimization formula is systematically transformed into an equivalent but more manageable form.

[0091]

[0092] in

[0093] Step 2 (Solving the second-order cone programming problem):

[0094] The optimized formula we transformed is a very standard second-order cone optimization problem, which can be solved quickly using the interior point method. There are also dedicated optimization toolkits available for its rapid solution.

[0095] Step 3 (Fixed base station beamforming, updating the solution):

[0096] After solving the base station beamforming matrix in the previous step, we need to optimize the remaining variables based on the solved matrix. The remaining optimization problem is:

[0097]

[0098] Since the previously solved formula is the same as this one, we cannot simply use the method of finding a feasible solution. This embodiment uses a method that maximizes the residual SINR to replace the optimization formula, thereby enabling mutual updates of subproblems and thus rapidly improving performance. K+1 auxiliary variables are introduced, representing the difference α between the SINR of K base station users and the threshold. k Let β be the difference between a satellite user's SINR and a threshold. All these variables must be greater than or equal to 0, while optimizing variable φ to maximize the sum of the auxiliary variables. The specific expression is as follows:

[0099]

[0100] Step 4 (SDP Relaxation and Parameter Calculation)

[0101] SINR, after fixing the base station beamforming matrix, can be simply expressed as follows:

[0102]

[0103] To simplify calculations, the following auxiliary parameters are introduced.

[0104]

[0105] These parameters are derived from the properties of diagonal matrices; the derivation is relatively easy but rather tedious. The remaining expressions and parameters can all be derived from A. k,m ,a k,m ,a k,m Similar forms and calculation processes are derived, but for the sake of brevity, their detailed derivations are omitted here.

[0106] Although auxiliary parameters simplify the calculation, optimization is still impossible due to the mode balance constraint of the phase shift matrix. This embodiment employs the SDP relaxation method, introducing the augmented vector v and the augmented matrix V as auxiliary parameters, where V = vv H It satisfies rank(V) = 1 and V ≥ 0. Then, the original expression for SINR is further equivalently substituted based on the characteristics of the auxiliary variables and the matrix. Let... in

[0107]

[0108] The derivation is quite obvious and will not be elaborated here. Other similar expressions can be equivalently replaced in this way. The relaxed expression is as follows:

[0109]

[0110] Step 5: Solve for and reconstruct the phase shift matrix:

[0111] This form of optimization formula can approximate the optimal solution using the interior-point method and a path-following strategy, and then iteratively solve the KKT conditions using Newton's method. CVX can also be used for fast solving, but because the original variables are relaxed using V, the optimized variable V needs further processing to recover the required phase shift matrix. However, the commonly used eigenvector recovery method based on the largest eigenvalue suffers from performance degradation because the resulting V matrix does not numerically satisfy the rank-1 requirement. Ensuring the rank of the matrix is ​​constrained by the degrees of freedom and strictness of the constraints, which imposes certain requirements on parameter settings and does not meet performance requirements. Therefore, a Gaussian randomization-based recovery method is adopted, which involves adding Gaussian random noise vectors of the same dimension to the eigenvectors of the largest eigenvalue, and retaining the optimal result through sampling. Thus, the value of the RIS phase shift matrix is ​​obtained after fixing the base station beamforming matrix.

[0112] Step 6: Iterate through the loop:

[0113] After solving the above problems, the optimized RIS phase shift matrix and base station beamforming matrix were obtained. The beamforming power consumption of the base station was calculated and recorded. Φ was updated to the RIS phase shift matrix calculated in the current calculation round. The process was then repeated from the first step until the calculated beamforming power consumption of the base station was lower than the threshold.

Claims

1. A method for suppressing interference in a satellite-to-ground fusion network assisted by a smart reflector, characterized in that, Applied to communication networks including satellites, base stations, smart reflectors, and multiple users, the following steps are included: Channel models for the direct link from satellite to user, the direct link from base station to user, the intelligent reflection link from satellite to user, and the intelligent reflection link from base station to user are established respectively. Based on the propagation path of the user's desired signal and the propagation path of the interference signal, a functional representation of the user's received signal is established using the corresponding channel model, and then the signal-to-interference-plus-noise ratio based on the statistical channel expectation is calculated. An optimization problem is constructed that satisfies the minimum signal-to-interference-plus-noise ratio constraint. Solving this optimization problem yields the optimal base station beamforming matrix and the optimal smart reflector phase shift matrix that minimize the beamforming power consumption of the base station in providing services to all its users. Based on the optimal base station beamforming matrix and the optimal smart reflector phase shift matrix, the base station beam shape and the smart reflector phase are dynamically adjusted.

2. The method according to claim 1, characterized in that, The minimum signal-to-interference-plus-noise ratio constraint includes: Signal-to-interference-plus-noise ratio (SINR) of base station user received signals k Not less than the set threshold γ UE The desired signal for the base station user comes from the base station, while the interference signal comes from the satellite and other base station users; Signal-to-interference-plus-noise ratio (SINR) of satellite user received signals su Not less than the set threshold γ SU The satellite user's desired signal comes from the satellite, while the interference signal comes from the base station.

3. The method according to claim 2, characterized in that, The signal-to-interference-plus-noise ratio (SINR) of the signals received by the base station users k Represented as Among them, G S This represents the equivalent baseband channel from the satellite to the smart reflector. G represents the equivalent baseband channel from the satellite to the k-th base station user. B Represents the equivalent baseband channel from the base station to the smart reflector, h k This represents the equivalent baseband channel from the base station to the base station user. This represents the equivalent baseband channel from the smart reflector to the base station user, where P1 and P2 represent the signal power transmitted by the base station and the satellite, respectively. k w represents the beamforming vector of the k-th base station user served by the base station. m W represents the beamforming vector of the m-th base station user served by the base station. sat The beamforming matrix of the satellite is represented by Φ, the phase shift matrix of the smart reflector is represented by a. k The value σ indicates whether the satellite's direct path is blocked, taking values ​​of {0,1}. 2 Indicates noise.

4. The method according to claim 2, characterized in that, The signal-to-interference-plus-noise ratio (SINR) of the satellite user received signal su Represented as Among them, G S G represents the equivalent baseband channel from the satellite to the smart reflector. B Represents the equivalent baseband channel from the base station to the smart reflector, h su This represents the equivalent baseband channel from the base station to the satellite user. This represents the equivalent baseband channel from the smart reflector to the satellite user, where P1 and P2 represent the signal power transmitted by the base station and the satellite, respectively. k w represents the beamforming vector of the k-th base station user served by the base station. m W represents the beamforming vector of the m-th base station user served by the base station. sat Let φ represent the beamforming matrix of the satellite, φ represent the phase shift matrix of the smart reflector, g represent the direct channel from the satellite to the satellite user, and σ represent the beamforming matrix of the satellite. 2 Indicates noise.

5. The method according to claim 1, characterized in that, Solving the optimization problem includes: Keeping the phase shift matrix of the intelligent reflector unchanged, and based on the duality of the constraints and the phase rotation invariance of the absolute value of the inner product of the beamforming vectors, the optimization problem is transformed into a second-order cone optimization problem. Solving the second-order cone optimization problem yields the optimal estimate of the base station beamforming matrix; Using the optimal estimate of the calculated base station beamforming matrix, the optimization problem is transformed into a first optimization sub-problem for the phase shift matrix of the smart reflector. Solving the first optimization subproblem yields the optimal estimate of the phase shift matrix of the intelligent reflector. The beamforming power consumption of the base station is calculated using the optimal estimation of the base station beamforming matrix and the optimal estimation of the phase shift matrix of the intelligent reflector. If the decrease in beamforming power consumption in the current round compared to the previous round is lower than a set threshold, then the optimal estimate of the current base station beamforming matrix and the optimal estimate of the smart reflector phase shift matrix are output as the optimal base station beamforming matrix and the optimal smart reflector phase shift matrix. Otherwise, the smart reflector phase shift matrix is ​​set to the optimal estimate of the current smart reflector phase shift matrix, and the process returns to the step of setting the smart reflector phase shift matrix unchanged.

6. The method according to claim 5, characterized in that, Solving the first optimization sub-problem includes: By introducing the difference between the signal-to-interference-plus-noise ratio (SIR) of the user's received signal and its set threshold as an auxiliary variable, the first optimization subproblem is transformed into a second optimization subproblem that maximizes the sum of the residuals between the SIR of all user received signals and their set thresholds. By employing the SDP relaxation method, the objective function and constraints of the second optimization subproblem are equivalently replaced to obtain the third optimization subproblem; Solving the third optimization subproblem yields the optimal estimate of the phase shift matrix of the intelligent reflector.

7. The method according to claim 6, characterized in that, When solving the third optimization subproblem, a method is used to mix the eigenvector of the largest eigenvalue with Gaussian random noise vectors of the same dimension to restore the calculated augmented matrix to the optimal estimate of the phase shift matrix of the intelligent reflector.

8. The method according to claim 1, characterized in that, When modeling the channel for direct links from satellite to user, direct links from base station to user, and direct links from smart reflector to user, Ricean fading is used to establish the corresponding models.

9. The method according to claim 1, characterized in that, When modeling the channel for the direct link from the satellite to the smart reflector and the direct link from the base station to the smart reflector, multipath effects are ignored.

10. The method according to claim 1, characterized in that, The gain of the reflection amplitude of all reflective elements in the intelligent reflective surface remains unchanged.

Images