A method and system for deploying a converged base station

Abstract

The present application relates to the field of edge network, more particularly, to a kind of sensing integration base station deployment method and system. Wherein the method comprises: system modeling is carried out, and base station deployment model is obtained;According to the positioning performance of base station deployment model, the positioning performance of base station deployment model is evaluated, and the objective function with the target of maximizing total rate and constraint are set according to the positioning performance of base station deployment model;By approximate convex relaxation, introducing auxiliary vector and semi relaxation technique, the objective function and constraint are processed, and the convex objective function and constraint are obtained;The convex objective function and constraint are solved, and base station deployment is completed.The present application sets the objective function with the target of maximizing total rate and constraint;Then by approximate convex relaxation, introducing auxiliary vector and semi relaxation technique are processed, and the convex objective function and constraint are obtained;Further, the non-convex problem that is more difficult to solve is converted into can be directly solved, so that the base station deployment scheme satisfying the maximum total rate is obtained, and deployment is completed.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of base station deployment, and more particularly, to a sensing and communication integrated base station deployment method and system. BACKGROUND

[0002] With the gradual maturity of 5G commercialization and the gradual freezing of related standards, its network performance and technical system are gradually stable. Following this, the concept of 6G opens a new chapter in the future development of communication technology. 6G aims to solve the more stringent demands in the social vision around 2030, which not only means that it needs to enhance the three typical scenarios of 5G, but also needs to expand new scenarios to meet the needs of emerging businesses. Among them, ISAC as a new scenario proposed by the International Telecommunication Union (ITU) for 6G has shown great development potential [1]. ISAC technology enables communication and sensing functions to be implemented simultaneously in the same system by sharing hardware resources, which has a wide application prospect in many fields. Therefore, researching and developing a sensing and communication integrated base station (ISAC) base station deployment strategy that can balance communication and sensing functions is of great significance to support the development of future 6G networks. Currently, the main problem of the deployment strategy of the ISAC base station is that the communication rate of the deployed base station is insufficient.

[0003] The prior art discloses a base station deployment method and device for high-precision positioning terminals. The method combines sensing in sensing and communication with environment sensing through multiple positioning methods such as AOA and TOA, and simultaneously performs positioning and map construction to achieve multi-dimensional mass data collection, and realizes: multi, that is, multiple mobile points of the second device cooperate with multiple first devices to obtain more data samples through beam scanning and using positioning technologies such as AOA and TOA; accurate, that is, environment sensing, RSRP / RSRPP, etc.; fast, that is, using sidelink communication technology to replace the original base station anchor point with the user equipment as a sampling point, so that fast station location planning can be realized. However, the communication rate of the base station deployed by the method is still insufficient. SUMMARY

[0004] The purpose of the present application is to disclose a sensing and communication integrated base station deployment method and system with better base station communication rate.

[0005] In order to achieve the above-mentioned purpose, the present application provides a sensing and communication integrated base station deployment method, comprising:

[0006] S1: system modeling of base station deployment is performed to obtain a base station deployment model;

[0007] S2: according to the base station deployment model, the positioning performance of the base station deployment model is evaluated, and a target function with the target of maximizing the total rate and constraints is set according to the positioning performance of the base station deployment model;

[0008] S3: The objective function and constraints are processed by approximate convex relaxation, introduction of auxiliary vectors and semi-relaxation technology to obtain convex objective function and constraints;

[0009] S4: The convex objective function and constraints are solved to complete base station deployment.

[0010] Further, in step S1, the base station deployment model includes: J candidate sites as candidate positions for base station deployment, the state of all candidate sites is stored by a deployment decision vector x j ∈{0,1}, the coordinates of the base station are denoted as q i =(x i ,y i ); the coordinates of the test point a are denoted as p=(x,y); the distance between the test point a and the base station is denoted as d i (p)=‖p-q i ‖; whether the test point a is associated with the base station is recorded by a matrix A tj ∈{0,1}; the average downlink signal-to-noise ratio experienced by a typical user terminal in a slow fading state at the coordinates of the base station is denoted as

[0011]

[0012] where p represents the transmit power, Pg′ i (p) represents the channel gain, g i (p)=(4πfd i (p) / v) -α exp(-σ 2 / 2ξ 2 ), f represents the center frequency, v represents the speed of light, α represents the propagation factor, σ represents the shadow standard deviation, ξ=10 / log(10), g i ′ (p) represents the power-normalized interference experienced by the UE,

[0013] represents the base station set, N0 represents the bidirectional noise power spectral density, and W represents the bandwidth;

[0014] For a communication target, the rate of the tth test point can be expressed as:

[0015]

[0016] where g tj (p)=g j (p t ), g′ tj (p)=∑ n∈J\{j} g tn x nrepresents the normalized interference experienced by a typical user terminal at the test point a given the deployment decision vector.

[0017] Further, in step S2, the setting, according to the base station deployment model, of an objective function aiming at maximizing the total rate to decide the deployment vector and the association matrix comprises evaluating the positioning performance of the base station deployment model by means of a lower bound of the positioning error:

[0018] The estimated user equipment position is defined as Based on the ranging vector The covariance matrix of

[0019]

[0020] where represents the expectation of represents is semi-definite and is the Fisher information matrix for given p and deployed base stations q = (q1, q2,... q J ) T The Fisher information matrix is defined as follows:

[0021]

[0022] Since the range measurements are assumed to be independent, it follows that Thus, it can be derived that

[0023]

[0024] The value of

[0025]

[0026] where v i is given by

[0027]

[0028] and d i = d i (p), θ i = θ i (p) is the angle between p and q i with respect to the horizontal axis, and h(y, d) is given by

[0029] ​

[0030] The formula to measure the positioning performance is as follows:

[0031]

[0032] where v t = [v 1t , v 2t ,..., v Jt ] T , v j (p t ) = v jt , {F t} i,j = v it v jt sin 2 (θ jt - θ it ).

[0033] Further, in step S2, the objective function includes:

[0034]

[0035] The constraints include:

[0036] Constraint one indicates that for the perception aspect of the positioning performance, b t should be less than a certain given threshold

[0037] b t ≤ ζ,

[0038] Constraint two indicates that when a test point is associated with a certain candidate site, at this time the candidate site must be deployed with a base station:

[0039] A ≤ 1 T x T ,

[0040] Constraint three indicates that each test point can only be associated with at most one built station:

[0041] A1 J ≤ 1 T ,

[0042] Constraint four indicates that the deployment is to be less than M:

[0043] x T 1 J ≤ M,

[0044] Constraint five indicates that for the deployment vector x and the association matrix A, the value can only be 0 or 1:

[0045] x ∈ {0, 1} J , A ∈ {0, 1} T×J

[0046] Further, in step S3, the processing by the approximate convex relaxation includes: introducing a new deployment vector constraint and an associated matrix constraint to relax the constraint five into a standard convex problem, by replacing the constraint five with the following inequality constraint, approximate convex relaxation is performed:

[0047]

[0048] Further, in step S4, the processing by introducing the auxiliary vector includes:

[0049] Rewrite r t as r′ t , the expression is as follows:

[0050]

[0051] By decomposing it into a series of convex problems: first, define a minimization problem of a log function as Eq

[0052]

[0053] According to the formula, r′ t can be processed, r′ t is rewritten by introducing an intermediate vector y, denoted as r″ t , the expression is as follows

[0054]

[0055] At this time, the objective function is:

[0056] Further, in step S3, the processing by the semi-relaxation technique includes:

[0057] Restate constraint one as:

[0058]

[0059] Using a known matrix identity, the is expressed as Tr(F t X); at the same time, due to the binary nature of the vector x, can be expressed as Tr(X⊙V t ), where V t is a diagonal matrix, whose diagonal elements are v t , at this time constraint one: Tr(X⊙Vt)≤ζTr(F t X),

[0060] Further, in step S4, fully convex the objective function and constraints include:

[0061]

[0062] subject to Tr(X☉V t )≤ζTr(F t X),

[0063] A≤1 T diag(X) T ,

[0064] A1 J ≤1 T ,

[0065] Tr(X)≤M,

[0066]

[0067] Further, at step S5, comprising: using the cvx toolbox to solve the completely convex objective function and constraint, completing the base station deployment.

[0068] In addition, the application also provides a sensing integration base station deployment system comprising:

[0069] The model construction module: the base station deployment is modeled to obtain a base station deployment model;

[0070] The objective function module: according to the base station deployment model, the positioning performance of the base station deployment model is evaluated, and the objective function and constraint with the target of maximizing the total rate are set according to the positioning performance of the base station deployment model;

[0071] The convex module: the objective function and constraint are processed through approximate convex relaxation, introduction of auxiliary vectors and semi-relaxation technology to obtain a convex objective function and constraint;

[0072] The solving module: the completely convex objective function and constraint are solved to complete the base station deployment.

[0073] Compared with the prior art, the beneficial effects of the technical scheme of the application are:

[0074] The application sets the objective function and constraint with the target of maximizing the total rate, then processes the objective function and constraint through approximate convex relaxation, introduction of auxiliary vectors and semi-relaxation technology to obtain a convex objective function and constraint, and further converts the non-convex problem which is difficult to solve into a problem that can be directly solved, so that the base station deployment scheme satisfying the maximum total rate is obtained, and the deployment is completed. BRIEF DESCRIPTION OF DRAWINGS

[0075] Figure 1 A flow chart of a sensing integration base station deployment method according to the first embodiment;

[0076] Figure 2 A base station deployment model schematic diagram as described in Embodiment Two;

[0077] Figure 3 A system block diagram of a sensing-integrated base station deployment system as described in Embodiment Three; DETAILED DESCRIPTION

[0078] The accompanying drawings are only intended to illustrate the embodiments of the present application, and should not be construed as limiting the present application;

[0079] The technical solutions of the present application will be further described in combination with the accompanying drawings and embodiments.

[0080] Embodiment One:

[0081] The present embodiment provides a sensing-integrated base station deployment method as shown in Figure 1 , comprising:

[0082] S1: System modeling is performed on base station deployment to obtain a base station deployment model;

[0083] S2: According to the base station deployment model, the positioning performance of the base station deployment model is evaluated, and a target function and constraints with the goal of maximizing the total rate are set according to the positioning performance of the base station deployment model;

[0084] S3: The target function and constraints are processed through approximate convex relaxation, introduction of auxiliary vectors, and semi-relaxation technology to obtain a convex target function and constraints;

[0085] S4: The convex target function and constraints are solved to complete the base station deployment.

[0086] The present embodiment sets a target function and constraints with the goal of maximizing the total rate, and then processes the target function and constraints through approximate convex relaxation, introduction of auxiliary vectors, and semi-relaxation technology to obtain a convex target function and constraints. Further, the non-convex problem which is difficult to solve is converted into a problem that can be directly solved, so that the base station deployment scheme that satisfies the goal of maximizing the total rate is obtained, and the deployment is completed.

[0087] Embodiment Two:

[0088] The present embodiment is further disclosed on the basis of Embodiment One:

[0089] In step S1, the base station deployment model comprises, as shown in Figure 2 : J candidate sites as candidate positions for base station deployment, and the states of all candidate sites are stored through a deployment decision vector x j ∈{0,1}, and the coordinates of the base station are denoted as q i =(x i ,y iThe coordinates of test point a are denoted as p = (x, y); the distance between test point a and the base station is denoted as d. i (p)=‖pq i ||; Matrix A is used to record whether there is a correlation between the base stations corresponding to test point a. tj ∈{0,1}; The average downlink signal-to-noise ratio experienced by a typical user terminal under slow decay at the base station-related coordinates is expressed as:

[0090]

[0091] Where p represents the transmission power, Pg′ i (p) represents the channel gain, g i (p)=(4πfd i (p) / v) -α exp(-σ 2 / 2ξ 2 ), f represents the center frequency, v represents the speed of light, α represents the propagation factor, σ represents the standard deviation of the shadow, ξ = 10 / log(10), g i ′ (p) represents the power-normalized interference experienced by the UE.

[0092] N represents the set of base stations, N0 represents the bidirectional noise power spectral density, and W represents the bandwidth.

[0093] For the communication target, the rate at the t-th test point can be expressed as:

[0094]

[0095] Where g tj (p)=g j (p t ), g′ tj (p)=∑ n∈J\{j} g tn x n This represents the standardized disturbance experienced by a typical user terminal at test point a given deployment decision vector.

[0096] In step S2, based on the base station deployment model, an objective function is set to determine the deployment vector and correlation matrix with the goal of maximizing the total rate. This includes evaluating the positioning performance of the base station deployment model through a lower bound on the positioning error.

[0097] Define the estimated user equipment location as Based on ranging vector The covariance matrix satisfies

[0098]

[0099] where represents the expectation of , represents is positive semi-definite, and is the Fisher information matrix for given p and deployed base station q = (q1, q2,... q J ) T , defined as follows:

[0100]

[0101] Since the range measurements are assumed to be independent, we have so that we can deduce

[0102]

[0103] The value of PEB is called the position error bound, which represents a lower bound on the mean square error of the distance between the UE position and its corresponding estimate. The PEB for ToA measurements with uniformly distributed measurement bias is expressed as follows:

[0104]

[0105] where v i is expressed as follows:

[0106]

[0107] and d i = d i (p), θ i = θ i (p) is the angle between p and q i with respect to the horizontal axis, and h(y, d) is expressed as follows:

[0108]

[0109] The formula to measure the positioning performance is as follows:

[0110]

[0111] where v t = [v 1t , v 2t ,..., v Jt ] T , v j (p t ) = v jt , {F t} i,j = v it vjt sin 2 (θ jt -θ it )。

[0112] In step S2, the objective function includes:

[0113]

[0114] The constraints include:

[0115] Constraint one indicates that for the positioning performance in the perception aspect, b t should be less than a given threshold ζ:

[0116] b t ≤ ζ,

[0117] Constraint two indicates that when a test point is associated with a candidate station, the candidate station must be deployed with a base station at this time:

[0118] A ≤ 1 T x T ,

[0119] Constraint three indicates that each test point can be associated with at most one station:

[0120] A1 J ≤ 1 T ,

[0121] Constraint four indicates that the deployment is less than M:

[0122] x T 1 J ≤ M,

[0123] Constraint five indicates that for the deployment vector x and the association matrix A, the value can only be 0 or 1:

[0124] x ∈ {0, 1} J , A ∈ {0, 1} T×J

[0125] In step S3, the processing by approximate convex relaxation includes: introducing a new deployment vector constraint and association matrix constraint to relax constraint five to a standard convex problem, by replacing constraint five with the following inequality constraint, approximate convex relaxation is performed:

[0126]

[0127] In step S4, the processing by introducing an auxiliary vector includes:

[0128] Rewrite r t as r′ t , the expression is as follows:

[0129]

[0130] By decomposing it into a series of convex problems: first, define a minimization problem of a log function as Eq

[0131]

[0132] According to the formula, r' t can be processed, by introducing an intermediate vector y to r' t Rewrite, denoted as r" t , the expression is as follows

[0133]

[0134] At this time the objective function is:

[0135] In step S3, by semi-relaxation technology processing includes:

[0136] Restate the constraint one as:

[0137]

[0138] Using a known matrix identity, the can be expressed as Tr(F t X); at the same time, due to the binary nature of the vector x, can be expressed as Tr(X⊙V t ), where V t is a diagonal matrix, whose diagonal elements are v t , at this time the constraint one: Tr(X⊙V t )≤ζTr(F t X),

[0139] In step S4, the objective function and the constraint are completely convex, including:

[0140]

[0141] subject toTr(X☉V t )≤ζTr(F t X),

[0142] A≤1 T diag(X) T ,

[0143] A1 J ≤1 T ,

[0144] Tr(X)≤M,

[0145]

[0146] At step S5, the base station deployment is completed by solving the completely convex target function and constraint using the cvx toolbox.

[0147] The embodiment sets a target function and constraint with the target of maximizing the total rate, then processes the target function and constraint by approximate convex relaxation, introduction of auxiliary vectors and semi-relaxation technology to obtain a convex target function and constraint, and further converts a non-convex problem which is difficult to solve into a problem which can be directly solved, so that a base station deployment scheme satisfying the maximum total rate is obtained, and the deployment is completed.

[0148] Embodiment three:

[0149] The embodiment provides a kind of general sense integration base station deployment system as shown in Figure 3 The embodiment provides a kind of general sense integration base station deployment system as shown in

[0150] Model construction module: the base station deployment is modeled, and the base station deployment model is obtained;

[0151] Target function module: according to the base station deployment model, the positioning performance of the base station deployment model is evaluated, and the target function and constraint with the target of maximizing the total rate are set according to the positioning performance of the base station deployment model;

[0152] Convex module: the target function and constraint are processed by approximate convex relaxation, introduction of auxiliary vectors and semi-relaxation technology to obtain a convex target function and constraint;

[0153] Solving module: the completely convex target function and constraint are solved, and the base station deployment is completed.

[0154] The embodiment sets a target function and constraint with the target of maximizing the total rate, then processes the target function and constraint by approximate convex relaxation, introduction of auxiliary vectors and semi-relaxation technology to obtain a convex target function and constraint, and further converts a non-convex problem which is difficult to solve into a problem which can be directly solved, so that a base station deployment scheme satisfying the maximum total rate is obtained, and the deployment is completed.

[0155] Obviously, the above embodiments of the present application are only examples for clearly illustrating the present application, and are not intended to limit the implementation modes of the present application. For ordinary skilled in the art, other different forms of changes or variations can be made on the basis of the above description. Here, all implementation modes are not required or can not be exhausted. Any modification, equivalent replacement and improvement made within the spirit and principle of the present application should be included in the protection scope of the claims of the present application.

Claims

1. A method for deploying an integrated sensing and communication base station, characterized in that, include: S1: Perform system modeling on base station deployment to obtain the base station deployment model; S2: Based on the base station deployment model, evaluate the positioning performance of the base station deployment model, and set the objective function and constraints with the goal of maximizing the total rate based on the positioning performance of the base station deployment model; S3: By processing the objective function and constraints through approximate convex relaxation, the introduction of auxiliary vectors, and semi-relaxation techniques, a convex objective function and constraints are obtained; Approximate convex relaxation involves: introducing a new deployment vector constraint and an incidence matrix constraint to relax constraint five into a standard convex problem; and performing approximate convex relaxation by replacing constraint five with the following inequality constraints: Processing by introducing auxiliary vectors includes: r t Rewritten as r′ t The expression is as follows: By decomposing it into a series of convex problems: First, we define a problem of minimizing a log function as Eq. According to this formula, we can determine r′ t This is done by introducing an intermediate vector y to adjust r′. t Rewrite, denoted as r″ t The expression is as follows The objective function is now: Treatment using semi-relaxation techniques includes: Restate constraint one as follows: Using a known matrix identity, Represented as Tr(F) t X); at the same time, due to the binary nature of vector x, This can be represented as Tr(X⊙V) t ), where V t It is a diagonal matrix with diagonal elements v t At this point, constraint one is: Tr(X⊙Vt)≤ζTr(FtX); The fully convex objective function and constraints include: subject toTr(X☉V t )≤ζTr(F t X), A1 J ≤1 T , Tr(X)≤M, S4: Solve the convex objective function and constraints to complete the base station deployment.

2. The method for deploying an integrated sensing base station according to claim 1, characterized in that, In step S1, the base station deployment model includes: J candidate sites as candidate locations for base station deployment, and the state x of all candidate sites is stored through a deployment decision vector. j ∈{0,1}, the coordinates of the base station are q i =(x i ,y i The coordinates of test point a are denoted as p = (x, y); the distance between test point a and the base station is denoted as d. i (p)=‖pq i ||; Matrix A is used to record whether there is a correlation between the base stations corresponding to test point a. tj ∈{0,1}; The average downlink signal-to-noise ratio experienced by a typical user terminal under slow decay at the base station-related coordinates is expressed as: Where p represents the transmission power, Pg′ i (p) represents the channel gain, g i (p)=(4πfd i (p) / v) -α exp(-σ 2 / 2ξ 2 ), f represents the center frequency, v represents the speed of light, α represents the propagation factor, σ represents the standard deviation of the shadow, ξ = 10 / log(10), g i ′ (p) represents the power-normalized interference experienced by the UE. N represents the set of base stations, N0 represents the bidirectional noise power spectral density, and W represents the bandwidth. For the communication target, the rate at the m-th test point can be expressed as: Where g tj (p)=g j (p t ), g′ tj (p)=∑ n∈J\{j} g tn x n This represents the standardized disturbance experienced by a typical user terminal at test point a given deployment decision vector.

3. The method for deploying an integrated sensing base station according to claim 2, characterized in that, In step S2, based on the base station deployment model, an objective function is set to determine the deployment vector and correlation matrix with the goal of maximizing the total rate. This includes evaluating the positioning performance of the base station deployment model through a lower bound on the positioning error. Define the estimated user equipment location as Based on ranging vector The covariance matrix satisfies in Representative to Expectations represent It is positive semidefinite, and For a given p and a deployed base station q = (q1, q2, ..., q... J ) T The Fisher information matrix is ​​defined as follows: Since the range measurements are assumed to be independent, then we have Therefore, it can be deduced that The value of is called the position error bound, which represents the lower bound of the mean square error of the distance between the UE position and its corresponding estimate. The PEB of a ToA measurement with a uniformly distributed measurement bias is expressed as follows: Where v i The expression is as follows: And d i =d i (p), θ i =θ i (p) refers to p and q about the horizontal axis. i The angle between them, h(y,d), is expressed as follows: The formula for measuring positioning performance is as follows: where v t = [v 1t , v 2t , …, v Jt T , v j (p t ) = v jt , {F t} i,j = v it v jt sin 2 (θ jt - θ it ).​ 4. The method for deploying an integrated sensing base station according to claim 3, characterized in that, In step S2, the objective function includes: The constraints include: Constraint 1 indicates that for the localization performance in terms of perception, b should be satisfied. t Less than a given threshold b t ≤≤ζ, Constraint 2 states that when a test point is associated with a candidate site, that candidate site must have a base station deployed at that site. A≤1 T x T , Constraint 3 specifies that each test point can be associated with at most one website: A1 J ≤1 T ， Constraint 4 indicates that the deployment must be less than M: x T 1 J ≤M, Constraint 5 states that for the deployment vector x and the incidence matrix A, the values ​​can only be 0 or 1: x∈{0,1} J ,A∈{0,1} T×J 。 5. The method for deploying an integrated sensing base station according to claim 1, characterized in that, Step S4 includes: using the cvx toolbox to solve the fully convex objective function and constraints to complete the base station deployment.

6. A sensor-integrated base station deployment system, characterized in that, include: Model building module: Performs system modeling of base station deployment to obtain the base station deployment model; Objective function module: Based on the base station deployment model, evaluate the positioning performance of the base station deployment model, and set the objective function and constraints with the goal of maximizing the total rate based on the positioning performance of the base station deployment model; Convexification module: The objective function and constraints are processed through approximate convex relaxation, the introduction of auxiliary vectors, and semi-relaxation techniques to obtain convex objective function and constraints; Approximate convex relaxation involves: introducing a new deployment vector constraint and an incidence matrix constraint to relax constraint five into a standard convex problem; and performing approximate convex relaxation by replacing constraint five with the following inequality constraints: Processing by introducing auxiliary vectors includes: r t Rewritten as r′ t The expression is as follows: By decomposing it into a series of convex problems: First, we define a problem of minimizing a log function as Eq. According to this formula, we can determine r′ t This is done by introducing an intermediate vector y to adjust r′. t Rewrite, denoted as r″ t The expression is as follows The objective function is now: Treatment using semi-relaxation techniques includes: Restate constraint one as follows: Using a known matrix identity, Represented as Tr(F) t X); at the same time, due to the binary nature of vector x, This can be represented as Tr(X⊙V) t ), where V t It is a diagonal matrix with diagonal elements v t At this point, constraint one is: Tr(X⊙Vt)≤ζTr(FtX); The fully convex objective function and constraints include: subject toTr(X☉V t )≤ζTr(F t X), A≤1 T diag(X) T , A1 J ≤1 T ， Tr(X)≤M, Solver module: Solve the fully convex objective function and constraints to complete the base station deployment.

Images