Search

CN-121978634-A - Active-passive hybrid radar antenna selection and power distribution method based on graph theory

CN121978634ACN 121978634 ACN121978634 ACN 121978634ACN-121978634-A

Abstract

The invention discloses a method for selecting and distributing active and passive hybrid radar antennas based on graph theory, which relates to the technical field of radars and comprises the steps of establishing an active and passive hybrid radar system model, establishing a target detection problem, bringing in a log-likelihood ratio, determining an optimal detector by utilizing an NP criterion, calculating the output signal-to-noise ratio of the radar system as a replacement index of the detection probability, constructing a graph theory model of the active-passive hybrid radar system, and providing a joint optimization AGS algorithm based on a Gale-shape (GS) algorithm to maximize the detection probability and iterate a matching antenna selection and power distribution algorithm. The invention efficiently selects reasonable active antenna operation and passive antenna processing, distributes proper power for the active antenna, greatly reduces the calculation complexity and improves the feasibility in practical application.

Inventors

  • WANG ZHEN
  • DU HONGLIN
  • LIU YANG

Assignees

  • 西南石油大学

Dates

Publication Date
20260505
Application Date
20260323

Claims (8)

  1. 1. The active and passive hybrid radar antenna selection and power distribution method based on graph theory is characterized by comprising the following steps: S1, establishing an active-passive hybrid radar system model; S2, establishing a target detection problem according to a received signal vector in the active-passive hybrid radar system model; S3, bringing the target detection problem into a log-likelihood ratio, and converting a log-likelihood ratio expression; S4, the log likelihood ratio expression determines an optimal detector based on NP criterion and deduces detection statistics; s5, calculating the output signal-to-noise ratio of the active and passive hybrid radar system according to the detection statistic to serve as a substitute index of the detection probability; s6, establishing a graph theory model of the active-passive hybrid radar system according to the detection probability; And S7, based on the graph theory model, an active antenna selection algorithm, a power distribution algorithm and a passive antenna selection algorithm are provided, the maximization of the detection probability is realized under the condition of limited total transmission power and system processing capacity, reasonable active and passive antennas are selected to work, and reasonable power is distributed for the working active antennas.
  2. 2. The method for power allocation and antenna selection for active-passive hybrid radar target detection according to claim 1, wherein step S1 comprises: S11, setting the active and passive hybrid radar system to comprise Candidate active antennas, N receivers Candidate passive antennas, from which to select The radar transmitting antennas work and distribute power, and select Defining binary selection variables and determining constraint conditions of the binary selection variables, wherein the binary selection variables comprise active antenna selection variables, power distribution variables and passive antenna selection processing condition variables; In the formula, For the active antenna selection variable, a value of 1 indicates selection of the first Candidate active antennas as the first The active antenna works, otherwise, the value is 0; a value of 1 for the power allocation variable indicates the first The transmission power of each active antenna selection is Otherwise, the value is 0; for the selection processing condition variable of the passive antenna, the value of 1 is the first The processing antenna seat is selected by A passive antenna; indicating the total amount of active antennas required to operate; Representing the number of candidate active antennas; Active antenna transmit power, the power of which is from a finite power set Is selected from the group consisting of, Wherein Is a collection Representing a common power class number Selectable discrete power, power per active antenna Are all a finite set One element of (a) is provided; representing the total amount of passive antennas processed; representing the number of passive antennas to be processed; S12, assume that the target is located Obtain the first With multiple receiving antennas (receivers) at The received signal at the time is expressed as: In the formula, Watch (watch) The receiving antennas are at A time of day received signal, wherein Indicating the serial number of the receiving antenna, The number of the sampling points is given, Is the sampling period; Is shown in Time of day (time) Noise at the individual receive antennas; And Respectively representing the reflection coefficient of the working active antenna path and the reflection coefficient of the passive antenna path, and assuming that the reflection coefficients are mutually independent zero-mean complex Gaussian random variables, the variances are respectively And ; And Respectively representing corresponding time delays; represent the first The distance from the receiving antennas to the target position; Representing the selected first The distance from the active antenna to the target location; represent the first Transmitting signals of the working active antennas; Representing the first to be processed The distance from the passive antennas to the target location; represent the first The transmission signals of the processed passive antennae; represent the first The transmission power of the passive antennas to be processed; S13, repeating the step S12 to obtain the first The receiving antenna receives signals at different moments to obtain the first The vector of all received signals for the individual receive antennas is expressed as: Wherein, the In the formula, Represent the first Vectors of all received signals for the individual receive antennas; 、 、 represent the first The receiving antennas are at 、 、 Symbol "†" represents a transpose; represent the first Noise vectors for the individual receive antennas; 、 、 Respectively represent the first The receiving antennas are at 、 、 Noise at the moment; represent the first Channel vectors of active antennas at the receiving antennas; 、 、 Respectively represent the first A receiving antenna and 1 st, 2 nd, Channel gains for the active antennas; represent the first Channel vectors of passive antennas at the respective receiving antennas; 、 、 Respectively represent the first A receiving antenna and 1 st, 2 nd, Channel gains for the passive antennas; s14, repeating the step S13, and obtaining vectors of all received signals of each receiving antenna, wherein the vectors are expressed as: Wherein, the Dimension is Dimension is , In the formula, 、 、 Respectively represent 1 st, 2 nd, Vectors of all received signals for the individual receive antennas; representing a matrix that gathers all active antenna signals of operation; 、 、 Respectively represent 1 st, 2 nd, The receiving antenna comprises a matrix of all active antenna transmitting signals; Representing a matrix that gathers all passive antenna signals; 、 、 Respectively represent 1 st, 2 nd, A matrix of processed passive antenna signals is included at the receive antenna; Representing a noise vector; 、 、 Respectively represent 1 st, 2 nd, Noise vectors at the receiving antennas; representing active antenna signal vectors of operation; 、 、 Respectively represent 1 st, 2 nd, The receiving antennas output the signal vectors of the received working active antennas; Representing the processed passive antenna channel vector; 、 、 Respectively represent 1 st, 2 nd, Receiving signal vectors of the processed passive antennas by the receiving antennas; Representation of Is a conjugate transpose of (2); Representation of Is a conjugate transpose of (2); Representing mathematical expectations; Representation of Is a covariance matrix of (a); Representation Is a covariance matrix of (a); 、 All obey zero-mean complex gaussian distribution and are mutually independent; 、 、 for intermediate variables, the general formula is calculated as: 、 、 for intermediate variables, the general formula is calculated as: 。
  3. 3. The graph-theory-based active-passive hybrid radar antenna selection and power distribution method according to claim 2, wherein in step S2, the object detection problem is: Wherein, the Indicating that the target is not present; indicating the presence of a target.
  4. 4. The graph-theory-based active-passive hybrid radar antenna selection and power allocation method according to claim 3, wherein in step S3, the log-likelihood ratio expression is as follows: Wherein, the In the formula, Representing a covariance matrix of the received vector when the object appears; representing a noise vector covariance matrix; 、 Represents an intermediate variable; representing vectors Is a conjugate transpose of (2); representing vectors Is a conjugate transpose of (2); Representing mathematical expectations; Representation of Is the inverse of (2); Representing a determinant.
  5. 5. The graph-theory-based active-passive hybrid radar antenna selection and power allocation method according to claim 4, wherein in step S4, the following operations are included: under the NP criterion, the expression of the optimal detector is as follows: representing a decision threshold obtained according to the false alarm probability constraint; The detection statistics are defined as The expression of the detection statistic obtained through deduction is as follows: In the formula, The representation is the variance of the noise; Is shown in the first At the receiving antenna, corresponding to the first Candidate antenna as the first The matched filters of the working antennas output signals; Is shown in the first At the receiving antenna, corresponding to the first Passive antenna to be processed as the first Processing the matched filter output signal of the passive antenna; Representing a complex number of modes.
  6. 6. The graph-theory-based active-passive hybrid radar antenna selection and power distribution method according to claim 5, wherein in step S5, the expression of the output signal-to-noise ratio of the detector is as follows: In the formula, Representing a mathematical expectation of the detection statistic in the presence of the target; Representing a mathematical expectation representing a detection statistic in the absence of the target; 、 are all intermediate variables.
  7. 7. The graph-theory-based active-passive hybrid radar antenna selection and power allocation method according to claim 6, wherein step S6 comprises: S61, the active radar part models the candidate radar transmitting antenna and the optional power into one vertex set of the bipartite graph in a combined way, models the working radar transmitting antenna into the other vertex set of the bipartite graph, and constructs a complete bipartite graph Wherein A set of joint vertices representing candidate active radar antennas and power, the set size being , 、 、 And Are all vertex sets Respectively represents that the power level of the 1 st candidate active radar antenna is 1, the power level of the 1 st candidate active radar antenna is 2, and the power level of the 1 st candidate active radar antenna is First, the The power class of each candidate active radar antenna is ; Representing a working radar transmitting antenna vertex set with the size of , 、 、 Respectively represent 1 st, 2 nd, An active radar section edge set is represented as The edges gather the elements of each edge 、 From a collection respectively Aggregation of Mapping the relation between the selection of radar transmitting antenna and power distribution, and defining the weight value of each side as ; S62, the passive radar part regards the passive antenna and the passive antenna processing seat as two vertexes of the bipartite graph respectively, and a complete bipartite graph of the passive antenna is constructed ; Representing a collection of passive antenna processing seats of a size of , Represents the 1 st to the 1 st A passive antenna processing seat; Representing a passive antenna vertex set with a set size of , Represents the 1 st to the 1 st Passive antennas to be processed; edge set Ensuring that the two vertices of each edge in the edge set are one from the set of passive antennas One from a collection of disposable passive antenna seats Mapping the relation between the total passive antenna and the passive antenna processing seat, and defining the weight of each side as 。
  8. 8. The method for power allocation and antenna selection for active-passive hybrid radar target detection according to claim 7, wherein step S7 specifically comprises: s71, setting initial parameters, namely the number of receiving antennas Sum distance Number of passive antennas Number of radar transmitting antennas Passive antenna position Active radar transmit antenna location Passive antenna transmit power Selectable power of radar transmitting antenna Variance of target reflection coefficient corresponding to passive antenna Variance of target reflection coefficient corresponding to active radar transmitting antenna ; S72, calculating a weight matrix according to the system parameters And And establish a preference list of passive antennas Preference list of radar transmitting antennas Preference list of processed passive antenna seats Preference list of radar transmit antenna selectable position and power combinations ; S73, in the passive part, processing the seat In its preference list Is selected to be the highest ranked and unmatched passive antenna If it is matched with And at Ranking of (3) Updating the matching result to be Otherwise, keeping the original matching, working seat in the active part In its list Selecting the highest ranked and unoccupied antenna-power combination if the combination is already on site Occupied and at Middle rank And updating the active radar antenna selection set to be smaller than the preset power after replacement Otherwise, the original matching is kept, and the iteration is carried out until the matching is completed.

Description

Active-passive hybrid radar antenna selection and power distribution method based on graph theory Technical Field The invention relates to the technical field of radars, in particular to a method for selecting and distributing active and passive hybrid radar antennas, and specifically relates to a method for selecting and distributing active and passive hybrid radar antennas based on graph theory. Background In an active-passive hybrid radar system, the number of active antennas and passive antennas that can actually operate due to system resource and cost limitations is limited, while the number of passive antennas that can be processed in the environment is far greater than the upper limit. Therefore, how to realize overall performance optimization by selecting reasonable active antenna operation and passive antenna processing under the premise of system processing capacity limitation becomes a key problem. Meanwhile, the transmission power distribution of the active antenna in the system also directly influences the performance. However, conventional power allocation and antenna selection methods are generally based on idealized assumptions, which are difficult to effectively address discrete power adjustment levels and limited processing power constraints in a real environment. Disclosure of Invention The invention aims to provide a method for selecting an active and passive hybrid radar antenna and distributing power based on graph theory, which aims to solve the problem of poor adaptability of the traditional method in an actual environment. In order to achieve the above object, the present invention provides the following technical solutions: a method for active and passive hybrid radar antenna selection and power distribution based on graph theory comprises the following steps: S1, establishing an active-passive hybrid radar system model; S2, establishing a target detection problem according to a received signal vector in the active-passive hybrid radar system model; S3, bringing the target detection problem into a log-likelihood ratio, and converting a log-likelihood ratio expression; S4, the log likelihood ratio expression determines an optimal detector based on NP criterion and deduces detection statistics; s5, calculating the output signal-to-noise ratio of the active and passive hybrid radar system according to the detection statistic to serve as a substitute index of the detection probability; s6, establishing a graph theory model of the active-passive hybrid radar system according to the detection probability; And S7, based on the graph theory model, an active antenna selection algorithm, a power distribution algorithm and a passive antenna selection algorithm, an AGS algorithm are provided, the maximization of the detection probability is realized under the condition of limited total transmission power and system processing capacity, reasonable active and passive antennas are selected to work, and reasonable power is distributed for the working active antennas. As a specific embodiment of the present invention, step S1 includes: S11, defining binary selection variables and determining constraint conditions of the binary selection variables, wherein the binary selection variables comprise active antenna selection variables, power distribution variables and passive antenna selection processing condition variables; In the formula, For the active antenna selection variable, a value of 1 indicates selection of the firstCandidate active antennas as the firstThe active antenna works, otherwise, the value is 0; a value of 1 for the power allocation variable indicates the first The transmission power of each active antenna selection isOtherwise, the value is 0; for the selection processing condition variable of the passive antenna, the value of 1 is the first The processing antenna seat is selected byA passive antenna; indicating the total amount of active antennas required to operate; Representing the number of candidate active antennas; Active antenna transmit power, the power of which is from a finite power set Is selected from the group consisting of,WhereinIs a collectionRepresenting a common power class numberSelectable discrete power, power per active antennaAre all a finite setOne element of (a) is provided; representing the total amount of passive antennas processed; representing the number of passive antennas to be processed; Setting up an active-passive hybrid radar system includes Candidate active radar transmitters (number of candidate active antennas), N receivers andCandidate passive antennas (passive antennas to be processed) are selected fromThe radar transmitting antennas work and distribute power, and selectProcessing by passive antennae; S12, assume that the target is located Obtain the firstWith multiple receiving antennas (receivers) atThe received signal at the time is expressed as: The second term is derived from a passive antenna; In the formula, Watch (watch)The receiving