A false target elimination method based on active and passive radar information correlation decision

Abstract

The present application is directed to the problem of false target interference in active radar detection, and proposes a false target elimination method based on active and passive radar information correlation decision, which uses the active and passive correlation decision result to suppress the active false target. Firstly, the two-dimensional state estimation and two-dimensional covariance matrix of the passive target are obtained; secondly, the two-dimensional state estimation and two-dimensional covariance matrix of the active target are obtained; thirdly, active and passive target correlation decision is performed; fourthly, the passive observation and active detection targets are diversified; and fifthly, active false target elimination is performed according to the diversification result. The present application can effectively remove the active detection false target and improve the active radar detection effect.

Description

Technical Field

[0001] This invention relates to an active radar false target elimination method. Background Technology

[0002] When an active radar detects a target, if the target itself or a jammer simulates a real radar echo signal to mislead the radar's detection and tracking of the real target, the active radar is considered to be under false target interference. Jammers simulate the echo signal of a real target, causing the radar to mistakenly believe there is an additional target, leading to misjudgments and false alarms by the radar and equipment system. On the one hand, false target interference weakens the radar's judgment capability, affecting its reliability and accuracy. On the other hand, the false targets generated by false target interference consume the radar's processing resources and bandwidth, preventing the radar from effectively processing real target information, resulting in a decrease in radar detection range and accuracy, and affecting the radar's efficiency and effectiveness. Currently used methods for radar anti-false target interference include frequency Doppler processing, spatiotemporal domain processing, and adaptive processing. These methods have complex algorithms and processing procedures, placing high demands on the real-time performance and computational and storage resources of the equipment system. Existing methods for joint active and passive target discrimination require a geographically distributed active and passive radar network, obtaining discrimination statistics by calculating the correlation distances of each target between the active and passive radars at different times. This method requires a large number of devices and a long distance between the active and passive radars. Summary of the Invention

[0003] To address the issues of complex processing methods for traditional radar anti-false target interference, which place high demands on equipment real-time performance and computational and storage resources, and the need for geographically dispersed radar networks in existing combined active and passive radar anti-false target interference methods, this invention proposes a false target elimination method based on active and passive radar information association judgment. This method uses active-passive association judgment to compare the correlation degree between active radar targets and the same batch of passive targets. The comparison results are used to identify and eliminate active false targets. The process is simple, effective, and robust, requiring only one active radar and one passive radar. It has low requirements for equipment real-time performance and computational and storage resources, and has no requirements on equipment spacing or location.

[0004] To achieve the above-mentioned technical objectives, the present invention is implemented through the following technical solution:

[0005] Step 1: The passive reconnaissance equipment detected M in the kth observation. k For the m-th objective (m = 1, 2, ..., M), k There are 1 target, and its batch number is denoted as P. mk The batch number received from the passive reconnaissance equipment is P. mk Target two-dimensional state estimation Among them, P mk The azimuth estimate is denoted as P mk The pitch angle estimate is denoted as The batch number receiving the output from the passive reconnaissance equipment is P. mk Target two-dimensional covariance matrix Among them, P mk The variance of the azimuth angle measurement is denoted as P mk The azimuth, elevation, and angle covariance are denoted as P mk The variance of the pitch angle measurement is denoted as

[0006] Step 2: The active radar detected N in the kth observation. k For the nth objective (n = 1, 2, ..., N), k There are 1 target, and its batch number is denoted as R. nk The batch number for receiving active radar output is R. nk Target two-dimensional state estimation Among them, R nk The azimuth estimate is denoted as R nk The pitch angle estimate is denoted as The batch number for receiving active radar output is R. nk Target two-dimensional covariance matrix Among them, R nk The variance of the azimuth angle measurement is denoted as R nk The azimuth, elevation, and angle covariance are denoted as R nk The variance of the pitch angle measurement is denoted as

[0007] Step 3: Execute P mk With R nk The target-related judgment;

[0008] Step 4: Perform diversity analysis on the target observed by the passive reconnaissance equipment and the active radar in the kth observation: For each R nk (n = 1, 2, ..., N) k If the batch number is P mk (m=1,2,...,M k The target was determined to be related to R. nk If there is a correlation, then P mk Belongs to set SP 1nk (n = 1, 2, ..., N) k Conversely, P mk Belongs to set SP 2nk (n = 1, 2, ..., N) k ); Set SP 1nkCorrelation statistic U nmk The largest P mk It also belongs to the set SP 3nk (n = 1, 2, ..., N) k For each SP 3nk (n = 1, 2, ..., N) k ) of P mk (m=1,2,...,M k If the batch number is R nk (n = 1, 2, ..., N) k The target was determined to be related to P. mk If there is a correlation, then R nk It belongs to set S1; otherwise, R nk It belongs to set S2; the correlation statistic U in set S1 nmk The largest R nk It also belongs to set S3;

[0009] Step 5: Perform active radar false target removal: R targets belonging to set S1 but not to set S3 nk For active radar false targets, the false targets are eliminated.

[0010] This invention identifies and eliminates active false targets based on the correlation between active radar targets and the same batch of passive targets. The process is simple and effective, enhancing the active radar's ability to resist false target interference and improving the overall detection effect of the equipment, while also having lower requirements for equipment quantity, real-time performance, and computing resources. Attached Figure Description

[0011] Figure 1 This is a diagram illustrating a specific implementation method. Detailed Implementation

[0012] The implementation process of this invention is as follows: Figure 1 As shown. The preferred implementation process includes the following steps:

[0013] Step 1: Obtain the two-dimensional state estimate and two-dimensional covariance matrix of the target passive observation:

[0014] The passive reconnaissance equipment detected M in the kth observation. k For the m-th objective (m = 1, 2, ..., M), k There are 1 target, and its batch number is denoted as P. mk The batch number received from the passive reconnaissance equipment is P. mk Target two-dimensional state estimation Among them, P mk The azimuth estimate is denoted as P mk The pitch angle estimate is denoted as The batch number receiving the output from the passive reconnaissance equipment is P. mk Target two-dimensional covariance matrix Among them, P mk The variance of the azimuth angle measurement is denoted as P mk The azimuth, elevation, and angle covariance are denoted as P mk The variance of the pitch angle measurement is denoted as

[0015] Step 2: Obtain the two-dimensional state estimate and two-dimensional covariance matrix of the target active detection:

[0016] The active radar detected N in the kth observation. k For the nth objective (n = 1, 2, ..., N), k There are 1 target, and its batch number is denoted as R. nk The batch number for receiving active radar output is R. nk Target two-dimensional state estimation Among them, R nk The azimuth estimate is denoted as R nk The pitch angle estimate is denoted as The batch number for receiving active radar output is R. nk Target two-dimensional covariance matrix Among them, R nk The variance of the azimuth angle measurement is denoted as R nk The azimuth, elevation, and angle covariance are denoted as R nk The variance of the pitch angle measurement is denoted as

[0017] Step 3: Execute P mk With R nk Target-related judgment:

[0018] Step 3 preferably implements the following process:

[0019] Step 3.1: Construct P mk With R nk Correlation statistics

[0020] Calculate the correlation covariance P nmk :

[0021] P nmk =P a (θ_R nk ,φ_R nk )+P p (θ_P mk ,φ_P mk )-P c-(P c ) T ,

[0022] Among them, P c It is the cross-covariance of target observations observed by the sensor, P c =0.

[0023] Correlation statistic U nmk The structure is as follows:

[0024]

[0025] Where N is the number of observations used for correlation, and its value is any positive integer not greater than k.

[0026] Step 3.2: Determine the correlation decision threshold

[0027] If the above The vector has dimension Sn, the rejection region significance level is set to pr = 0.05, and the objective function is P. mk With R nk The associated decision threshold λ nmk for:

[0028]

[0029] Step 3.3: Execute P mk With R nk Related judgments

[0030] If the following conditions are met:

[0031] U nmk ≤λ nmk ,

[0032] Determine P mk With R nk If there is no correlation, then determine P. mk With R nk Not related.

[0033] Step 4: Perform diversity analysis on the target observed by the passive reconnaissance equipment and the active radar in the kth observation:

[0034] For each R nk (n = 1, 2, ..., N) k If the batch number is P mk (m=1,2,...,M k The target was determined to be related to R. nk If there is a correlation, then P mk Belongs to set SP 1nk (n = 1, 2, ..., N) k Conversely, P mk Belongs to set SP 2nk(n = 1, 2, ..., N) k Set SP 1nk Correlation statistic U nmk The largest P mk It also belongs to the set SP 3nk (n = 1, 2, ..., N) k For each SP 3nk (n = 1, 2, ..., N) k ) of P mk (m=1,2,...,M k If the batch number is R nk (n = 1, 2, ..., N) k The target was determined to be related to P. mk If there is a correlation, then R nk It belongs to set S1; otherwise, R nk It belongs to set S2; the correlation statistic U in set S1 nmk The largest R nk It also belongs to set S3.

[0035] Step 5: Perform active radar false target removal:

[0036] R belongs to set S1 but not to set S3 nk This is a false target for active radar. The false target will be removed.

Claims

1. A method for false target elimination based on the correlation decision of active and passive radar information, characterized in that: Step 1: The passive reconnaissance equipment detected M in the kth observation. k For the m-th objective (m = 1, 2, ..., M), k There are 1 target, and its batch number is denoted as P. mk The batch number received from the passive reconnaissance equipment is P. mk Target two-dimensional state estimation Among them, P mk The azimuth estimate is denoted as P mk The pitch angle estimate is denoted as The batch number receiving the output from the passive reconnaissance equipment is P. mk Target two-dimensional covariance matrix Among them, P mk The variance of the azimuth angle measurement is denoted as P mk The azimuth, elevation, and angle covariance are denoted as P mk The variance of the pitch angle measurement is denoted as Step 2: The active radar detected N in the kth observation. k For the nth objective (n = 1, 2, ..., N), k There are 1 target, and its batch number is denoted as R. nk The batch number for receiving active radar output is R. nk Target two-dimensional state estimation Among them, R nk The azimuth estimate is denoted as R nk The pitch angle estimate is denoted as The batch number for receiving active radar output is R. nk Target two-dimensional covariance matrix Among them, R nk The variance of the azimuth angle measurement is denoted as R nk The azimuth, elevation, and angle covariance are denoted as R nk The variance of the pitch angle measurement is denoted as Step 3: Execute P mk With R nk The target-related judgment; Step 4: Perform diversity analysis on the target observed by the passive reconnaissance equipment and the active radar in the kth observation: For each R nk (n = 1, 2, ..., N) k If the batch number is P mk (m=1,2,...,M k The target was determined to be related to R. nk If there is a correlation, then P mk Belongs to set SP 1nk (n = 1, 2, ..., N) k Conversely, P mk Belongs to set SP 2nk (n = 1, 2, ..., N) k ); Set SP 1nk Correlation statistic U nmk The largest P mk It also belongs to the set SP 3nk (n = 1, 2, ..., N) k For each SP 3nk (n = 1, 2, ..., N) k ) of P mk (m=1,2,...,M k If the batch number is R nk (n = 1, 2, ..., N) k The target was determined to be related to P. mk If related, then R nk It belongs to set S1; otherwise, R nk It belongs to set S2; the correlation statistic U in set S1 nmk The largest R nk It also belongs to set S3; Step 5: Perform active radar false target removal: R targets belonging to set S1 but not to set S3 nk For active radar false targets, the false targets are eliminated.

2. The false target elimination method based on the correlation decision of active and passive radar information according to claim 1, characterized in that: The target association decision method in step 3 includes: Step 3-1: Construct P mk With R nk Correlation statistics; calculate correlation covariance P. nmk : P nmk =P a (θ_R nk ,φ_R nk )+P p (θ_P mk ,φ_P mk )-P c -(P c ) T , Among them, P c It is the cross-covariance of target observations observed by the sensor, P c =0; Correlation statistic U nmk The structure is as follows: Where N is the number of observations used for correlation, and its value is any positive integer not greater than k; Step 3-2: Determine the association decision threshold: If the above The vector has dimension Sn, the rejection region significance level is set to pr = 0.05, and the objective function is P. mk With R nk The associated decision threshold λ nmk for: Step 3-3: Execute P mk With R nk The associated judgment; if the following conditions are met: IN nmk ≤λ nmk ; Determine P mk With R nk If there is no correlation, then determine P. mk With R nk Not related.

Images