A network station sorting and tracking method based on RB-Stream algorithm

Through the network station sorting and tracking method based on the RB-Stream algorithm, the frequency hopping signal parameter data flow is clustered using an efficient clustering model representing the point structure, and the existing technology of the network station sorting algorithm is solved in a complex environment. The problem of poor computing complexity and poor real-time performance realizes real-time tracking and sorting of frequency hopping communication systems, and improves the visualization level.

CN115940993BActive Publication Date: 2024-11-05THE 54TH RESEARCH INSTITUTE OF CHINA ELECTRONICS TECHNOLOGY GROUP CORPORATION
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202211546755.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-12-05
Publication Date
2024-11-05
Estimated Expiration
2042-12-05

AI Technical Summary

Technical Problem

The existing network station sorting algorithms have complex calculations and poor real-time performance in complex electromagnetic environments, and cannot realize real-time tracking of network stations, resulting in a degradation of performance.

Method used

The network station sorting and tracking method based on RB-Stream algorithm is adopted, and the frequency hopping signal parameter data stream is clustered through an efficient clustering model representing the point structure, real-time sorting and tracking is realized, and the traditional batch processing mode is abandoned.

Benefits of technology

It improves the real-time and visualization level of network and station sorting, realizes real-time tracking of frequency hopping communication systems, moderate calculation volume, and is easy to implement.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115940993B_ABST
    Figure CN115940993B_ABST
Patent Text Reader

Abstract

The present invention discloses a network station sorting and tracking method based on RB-Stream algorithm, and belongs to the field of communication countermeasure technology. The present invention includes the steps of caching frequency hopping signal parameter data stream, sorting the cached data in ascending order according to arrival time, dividing duration groups, estimating hopping speed, forming / updating frequency hopping period groups, estimating frequency hopping period, calculating clustering features, clustering data stream using RB-Stream algorithm, identifying network station status and calculating network station parameters. The present invention abandons the traditional batch processing mode, and realizes real-time sorting and tracking of network stations by clustering the frequency hopping signal parameter data stream using an efficient clustering model based on representative point structure, and improves the visualization level of network station sorting. The present invention has advanced technology, moderate calculation amount, and is easy to implement in engineering. It can realize real-time processing of frequency hopping data, and is an important improvement on the prior art.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of communication confrontation, and specifically refers to a network station sorting and tracking method based on RB-Stream algorithm. Background Art

[0002] Frequency hopping communication has been widely used in the field of military communications as an anti-interference communication method. As a third party in non-cooperative communication, if you want to conduct reconnaissance and interference on frequency hopping communication, an important prerequisite is to be able to correctly sort the network stations and obtain the frequency set, hopping speed and other parameters of the frequency hopping radio station.

[0003] Existing network station sorting algorithms include network station sorting algorithms based on blind source separation, network station sorting algorithms based on frequency hopping feature clustering, etc. Among them, the network station sorting algorithm based on blind source separation uses the blind source separation algorithm to process the time domain signal for sorting. Its calculation is complex, the real-time performance is poor, and it is difficult to adapt to complex electromagnetic environments. The network station sorting algorithm based on frequency hopping feature clustering sorts according to the arrival time, direction and other characteristics of the frequency hopping signal. If the clock consistency of different radio stations is poor, its arrival time will drift slowly, affecting the extraction of the arrival time characteristics of the frequency hopping radio station, making the performance of the network station sorting algorithm based on frequency hopping feature clustering degraded. The above-mentioned network station sorting methods are all batch processing modes, and the sorting results output at different times have no correlation and no network station tracking capability. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects existing in the above-mentioned prior art and propose a network station sorting and tracking method based on the RB-Stream algorithm. This method abandons the traditional batch processing mode and clusters the frequency hopping signal parameter data stream by using an efficient clustering model based on the representative point structure. It realizes real-time sorting and tracking of network stations and provides technical support for frequency hopping communication countermeasures.

[0005] To achieve the above object, the technical solution adopted by the present invention is:

[0006] A network station sorting and tracking method based on RB-Stream algorithm, characterized by comprising the following steps:

[0007] (1) caching a frequency hopping signal parameter data stream; the frequency hopping signal parameter data stream is derived from a reconnaissance of a frequency hopping communication system, the frequency hopping communication system comprising a plurality of network stations, the stations in the network stations communicating via frequency hopping signals; the frequency hopping signal parameter data stream comprises a plurality of frequency hopping signal parameter vectors, the frequency hopping signal parameter vectors comprising a frequency, an arrival time and a duration of the frequency hopping signal;

[0008] (2) sorting the cached frequency hopping signal parameter vectors in ascending order according to arrival time;

[0009] (3) setting the current time to the sum of the arrival time and duration of the last frequency hopping signal in the frequency hopping signal parameter data stream;

[0010] (4) using the time resolution to normalize the arrival time and duration in the frequency hopping signal parameter vector;

[0011] (5) The frequency hopping signal parameter vectors are divided according to the duration. The frequency hopping signal parameter vectors with the same duration are divided into the same duration group. The frequency hopping signal parameter vectors in the duration group are still arranged in ascending order according to the arrival time.

[0012] (6) Screening the duration groups divided in step (5); if the number of frequency hopping signal parameter vectors of a duration group exceeds threshold 1 and the time interval between the current time and the estimated hopping speed of the duration group is greater than threshold 2, performing hopping speed estimation; if the hopping speed estimation succeeds, setting the hopping speed of the duration group to the hopping speed estimation value, and setting the estimated hopping speed time of the duration group to the current time; if the hopping speed estimation fails, setting the hopping speed of the duration group to an illegal value; the threshold 1 is a set value M; when the hopping speed of the duration group is an illegal value, the threshold 2 is a set time A; when the hopping speed of the duration group is a legal value, the threshold 2 is a set time B, wherein the set time A is less than the set time B;

[0013] (7) The frequency hopping signal parameter vectors in the duration group whose hopping speed is a legal value and whose hopping speed difference does not exceed the set value C are moved to the same frequency hopping period group, and the frequency hopping period of the newly formed frequency hopping period group is set to an illegal value. The frequency hopping signal parameter vectors in the frequency hopping period group are still arranged in ascending order according to the arrival time; if the hopping speed of the duration group is an illegal value, and the number of frequency hopping signal parameter vectors in the duration group exceeds threshold 3 or the time interval between the current time and the arrival time of the first frequency hopping signal parameter vector in the duration group exceeds threshold 4, the frequency hopping signal parameter vectors of the duration group are cleared and the estimated hopping speed time is set to the current time; the threshold 3 is the set value D, and the threshold 4 is the set value E;

[0014] (8) If the frequency hopping period of the frequency hopping period group is an illegal value and the number of the frequency hopping signal parameter vectors in the frequency hopping period group exceeds a threshold of 5, a frequency hopping period estimation value is calculated according to the arrival time of the frequency hopping signal parameter vectors in the frequency hopping period group and the frequency hopping period of the frequency hopping period group is set to the frequency hopping period estimation value; the threshold of 5 is a set value F;

[0015] (9) For each frequency hopping period group whose frequency hopping period is a legal value, the sequence number and relative arrival time of each frequency hopping signal parameter vector are calculated as clustering features;

[0016] (10) For each frequency hopping period group whose frequency hopping period is a legal value, frequency hopping signal parameter vectors with a time span of duration 1 are sequentially taken out in chronological order, and data stream clustering is performed on them using the RB-Stream algorithm. The frequency hopping signal parameter vectors in each cluster belong to the same network station; the duration 1 is greater than the product of the frequency hopping period of the frequency hopping period group and the set value H;

[0017] (11) For each cluster, the state of the network station is determined according to the density of the data flow in the cluster. When the cluster changes from non-existence to existence, the network station is in the appearance state; when the cluster changes from existence to non-existence, the network station disappears; when the data flow density is not less than a threshold of 6, the network station is in the active state; when the data flow density is less than a threshold of 6, the network station is in the silent state; the threshold 6 is a set value N;

[0018] (12) For each cluster, the reciprocal of the frequency hopping period of the frequency hopping period group to which the cluster belongs is the hopping speed of the corresponding network station. The number of occurrences of the frequency of each frequency hopping signal parameter vector in the cluster is counted at a set time interval, and the frequency with the highest number of occurrences is selected to form the frequency set of the network station corresponding to the cluster;

[0019] (13) Repeat steps (1) to (12) to achieve real-time sorting and tracking of the network stations.

[0020] Furthermore, in step (6), the calculation method of the hopping rate estimation is the PRI pulse repetition interval transformation method.

[0021] Furthermore, in step (8), the frequency hopping period estimation value is calculated using a frequency hopping period estimation algorithm based on a cost function.

[0022] Furthermore, in step (9), the serial number of each frequency hopping signal parameter vector is the integer part of the quotient of the arrival time of the frequency hopping signal parameter vector and the frequency hopping period of the frequency hopping period group to which it belongs, and the relative arrival time of each frequency hopping signal parameter vector is the remainder of the division of the arrival time of the frequency hopping signal parameter vector and the frequency hopping period of the frequency hopping period group to which it belongs.

[0023] Furthermore, in step (12), the specific method of selecting the frequencies with high occurrence times to form the frequency set of the network station corresponding to the cluster is to calculate the mean of the number of occurrences of each vector frequency, and select the frequencies with an occurrence number greater than one-half of the mean to form the frequency set of the network station corresponding to the cluster.

[0024] Compared with the prior art, the present invention has the following advantages:

[0025] 1. The present invention abandons the traditional batch processing mode and clusters the frequency hopping signal parameter data stream by using an efficient clustering model based on the representative point structure, thereby realizing real-time sorting and tracking of network stations and improving the visualization level of network station sorting.

[0026] 2. The present invention has advanced technology, moderate computational complexity, is easy to implement in engineering, and can realize real-time processing of frequency hopping data. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flow chart of the network station sorting and tracking method in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The following is combined with Figure 1 The present invention is described in further detail.

[0029] A network station sorting and tracking method based on RB-Stream algorithm is used to analyze the frequency hopping signal parameter data stream obtained by reconnaissance, so as to sort and track the network stations. The frequency hopping signal parameter data stream comes from the reconnaissance of the frequency hopping communication system, and the frequency hopping communication system includes a plurality of network stations, and the radio stations in the network stations communicate with each other through frequency hopping signals. The method includes the following steps:

[0030] (1) caching a frequency hopping signal parameter data stream; the frequency hopping signal parameter data stream includes a plurality of frequency hopping signal parameter vectors, wherein the frequency hopping signal parameter vectors include components of a signal frequency, an arrival time, and a duration;

[0031] (2) sorting the cached frequency hopping signal parameter vectors in ascending order according to arrival time;

[0032] (3) setting the current time to the sum of the arrival time and duration of the last frequency hopping signal in the frequency hopping signal parameter data stream;

[0033] (4) using the time resolution to normalize the arrival time and duration in the frequency hopping signal parameter vector, the calculation method is: normalized arrival time = arrival time / time resolution, normalized duration = duration / time resolution;

[0034] (5) The frequency hopping signal parameter vectors are divided according to the duration. The frequency hopping signal parameter vectors with the same duration are divided into the same duration group. The frequency hopping signal parameter vectors in the duration group are still arranged in ascending order according to the arrival time.

[0035] (6) Screening the duration groups divided in step (5); if the number of frequency hopping signal parameter vectors of a duration group exceeds threshold 1 and the time interval between the current time and the estimated hopping speed of the duration group is greater than threshold 2, performing hopping speed estimation; if the hopping speed estimation is successful, setting the hopping speed of the duration group to the hopping speed estimation value, and setting the estimated hopping speed time of the duration group to the current time; otherwise, setting the hopping speed of the duration group to an illegal value; the threshold 1 is greater than 150; when the hopping speed of the duration group is an illegal value, the threshold 2 is greater than 1 second, otherwise, the threshold 2 is greater than 8 seconds; the calculation method of the hopping speed estimation is the PRI pulse repetition interval conversion method;

[0036] (7) Processing the data in the duration group to form or update the frequency hopping period group; specifically:

[0037] (7a) moving the frequency hopping signal parameter vectors in the duration group whose hopping rates are legal values ​​and similar to each other to the same frequency hopping period group;

[0038] (7b) If the frequency hopping period group is newly formed, its frequency hopping period is set to an illegal value;

[0039] (7c) The frequency hopping signal parameter vectors in the frequency hopping cycle group are arranged in ascending order according to the arrival time;

[0040] (7d) if the frequency hopping rate of the duration group is an illegal value, and the number of frequency hopping signal parameter vectors in the duration group exceeds threshold 3 or the time interval between the current time and the arrival time of the first frequency hopping signal parameter vector in the duration group exceeds threshold 4, then clear the frequency hopping signal parameter vector of the duration group and set its estimated frequency hopping rate time to the current time; the threshold 3 is greater than 15000, and the threshold 4 is greater than 30 seconds;

[0041] (8) if the frequency hopping period of the frequency hopping period group is an illegal value and the number of the frequency hopping signal parameter vectors in the frequency hopping period group exceeds a threshold of 5, a frequency hopping period estimation value is calculated according to the arrival time of the frequency hopping signal parameter vectors in the frequency hopping period group and the frequency hopping period of the frequency hopping period group is set to the frequency hopping period estimation value; the threshold of 5 is greater than 900; the frequency hopping period estimation value is calculated using a frequency hopping period estimation algorithm based on a cost function;

[0042] (9) For each frequency hopping period group whose frequency hopping period is a legal value, the sequence number and relative arrival time of each frequency hopping signal parameter vector are calculated as clustering features; the sequence number and relative arrival time are calculated as follows: RT i =mod(T i ,P), where N iis the serial number of the ith frequency hopping signal, RT i is the relative arrival time of the ith frequency hopping signal, means round down, mod means remainder, T i is the arrival time of the ith frequency hopping signal, and P is the frequency hopping period of the frequency hopping period group;

[0043] (10) For each frequency hopping period group whose frequency hopping period is a legal value, frequency hopping signal parameter vectors with a time span of duration 1 are sequentially taken out in chronological order, and data stream clustering is performed on them using the RB-Stream algorithm. The frequency hopping signal parameter vectors in each cluster belong to the same network station; the duration 1 is greater than the product of the frequency hopping period of the frequency hopping period group and 50;

[0044] (11) Identify network status and calculate network parameters; specifically:

[0045] (11a) For each cluster, the state of the network station is determined according to the density of the data flow in the cluster. The network station appears when the cluster changes from non-existence to existence, disappears when the cluster changes from existence to non-existence, is in an active state when the network station is active, and is in a silent state when the data flow density is lower than the threshold 6; the threshold 6 is greater than 0.01;

[0046] (11b) For each cluster, the reciprocal of the frequency hopping period of the frequency hopping period group to which the cluster belongs is the hopping rate of the corresponding network station;

[0047] (11c) Counting the number of occurrences of the frequency of each frequency hopping signal parameter vector in the cluster at a certain time interval, and selecting the frequencies with the highest number of occurrences to form the frequency set of the network station corresponding to the cluster; the time interval is greater than 1 second.

[0048] Repeat steps (1) to (11) to achieve real-time sorting and tracking of the network stations.

[0049] Here is a more specific example:

[0050] Reference Figure 1 , a network station sorting and tracking method based on RB-Stream algorithm, comprising the following steps:

[0051] Step 1) caching a frequency hopping signal parameter data stream: the frequency hopping signal parameter data stream includes a plurality of frequency hopping signal parameter vectors, wherein the frequency hopping signal parameter vectors include components of the frequency, arrival time and duration of the signal; in this embodiment, the parameters of the frequency hopping signal are obtained by a frequency hopping signal detection method based on short-time Fourier transform, and the time resolution in the detection method is 10 microseconds;

[0052] Step 2) sorting the cached frequency hopping signal parameter vectors in ascending order according to arrival time;

[0053] Step 3) setting the current time to the sum of the arrival time and duration of the last frequency hopping signal in the frequency hopping signal parameter data stream;

[0054] Step 4) normalize the arrival time and duration in the frequency hopping signal parameter vector using the time resolution, and the calculation method is: normalized arrival time = arrival time / time resolution, normalized duration = duration / time resolution;

[0055] Step 5) dividing the frequency hopping signal parameter vectors according to the duration, the frequency hopping signal parameter vectors with the same duration are divided into the same duration group, and the frequency hopping signal parameter vectors in the duration group are still arranged in ascending order according to the arrival time;

[0056] Step 6) Screen the duration groups divided in step 5). If the number of frequency hopping signal parameter vectors of a duration group exceeds threshold 1 and the time interval between the current time and the estimated hopping speed of the duration group is greater than threshold 2, the PRI pulse repetition interval transformation method is used to estimate the hopping speed; if the hopping speed estimation is successful, the hopping speed of the duration group is set to the hopping speed estimation value, and the time of the estimated hopping speed of the duration group is set to the current time, otherwise, the hopping speed of the duration group is set to an illegal value; in this embodiment, threshold 1 is 200; when the hopping speed of the duration group is an illegal value, threshold 2 is 3 seconds, otherwise, threshold 2 is 8 seconds;

[0057] Step 7) Process the data in the duration group to form or update the frequency hopping period group; specifically:

[0058] Step 7a) moving the frequency hopping signal parameter vectors in the duration group whose hopping rates are legal values ​​and similar to each other to the same frequency hopping period group;

[0059] Step 7b) if the frequency hopping period group is newly formed, then set its frequency hopping period to an illegal value;

[0060] Step 7c) The frequency hopping signal parameter vectors in the frequency hopping cycle group are arranged in ascending order according to arrival time;

[0061] Step 7d) if the hopping rate of the duration group is an illegal value, and the number of the frequency hopping signal parameter vectors in the duration group exceeds threshold 3 or the time interval between the current time and the arrival time of the first frequency hopping signal parameter vector in the duration group exceeds threshold 4, then the frequency hopping signal parameter vector of the duration group is cleared and the estimated hopping rate time is set to the current time; in this embodiment, threshold 3 is 20000 and threshold 4 is 32 seconds;

[0062] Step 8) If the frequency hopping period of the frequency hopping period group is an illegal value and the number of the frequency hopping signal parameter vectors in the frequency hopping period group exceeds a threshold of 5, a frequency hopping period estimation algorithm based on a cost function is used to calculate an estimated value of the frequency hopping period according to the arrival time of the frequency hopping signal parameter vectors in the frequency hopping period group, and the frequency hopping period of the frequency hopping period group is set to the estimated value of the frequency hopping period; in this embodiment, the threshold 5 is 1000;

[0063] Step 9) For each frequency hopping period group whose frequency hopping period is a legal value, the sequence number and relative arrival time of each frequency hopping signal parameter vector are calculated as clustering features; the sequence number and relative arrival time are calculated as follows: RT i =mod(T i ,P), where N i is the serial number of the ith frequency hopping signal, RT i is the relative arrival time of the ith frequency hopping signal, means round down, mod means remainder, T i is the arrival time of the ith frequency hopping signal, and P is the frequency hopping period of the frequency hopping period group;

[0064] Step 10) For each frequency hopping period group whose frequency hopping period is a legal value, frequency hopping signal parameter vectors with a time span of duration 1 are taken out in chronological order, and data stream clustering is performed on them using the RB-Stream algorithm. The frequency hopping signal parameter vectors in each cluster belong to the same network station; in this embodiment, duration 1 is the product of the frequency hopping period of the frequency hopping period group and 60;

[0065] Step 11) Identify the network station status and calculate the network station parameters; specifically:

[0066] Step 11a) For each cluster, the state of the network station is determined according to the density of the data flow in the cluster. The network station appears when the cluster changes from non-existence to existence, disappears when the cluster changes from existence to non-existence, is in the active state when the network station is active, and is in the silent state when the data flow density is lower than the threshold 6. In this embodiment, the threshold 6 is 0.05.

[0067] Step 11b) For each cluster, the reciprocal of the frequency hopping period of the frequency hopping period group to which the cluster belongs is the hopping speed of the corresponding network station;

[0068] Step 11c) Count the number of occurrences of the frequency of each frequency hopping signal parameter vector in the cluster at a certain time interval, calculate the mean of the number of occurrences, and select the frequencies whose number of occurrences is greater than half of the mean to form a frequency set; in this embodiment, the time interval is 3 seconds.

[0069] Repeat steps 1) to 11) to achieve real-time sorting and tracking of the network stations.

[0070] The present invention abandons the traditional batch processing mode and clusters the frequency hopping signal parameter data stream by using an efficient clustering model based on a representative point structure, thereby achieving real-time sorting and tracking of network stations and improving the visualization level of network station sorting. The present invention is technologically advanced and is an important improvement on the prior art.

[0071] It should be understood that the above description of the specific implementation methods of this patent is only an exemplary description listed to facilitate ordinary technicians in the field to understand the patent scheme, and does not imply that the scope of protection of this patent is limited to these individual examples. Ordinary technicians in this field can fully understand the technical scheme of this patent and, without any creative work, obtain more specific implementation methods by combining technical features, replacing some technical features, adding more technical features, etc. to the examples listed in this patent. All these specific implementation methods are within the scope of the claims of this patent. Therefore, these new specific implementation methods should also be within the scope of protection of this patent.

Claims

1. A network station sorting and tracking method based on RB-Stream algorithm, It is characterized in that The following steps are involved: (1) caching a frequency hopping signal parameter data stream; the frequency hopping signal parameter data stream is derived from a reconnaissance of a frequency hopping communication system, the frequency hopping communication system comprising a plurality of network stations, the stations in the network stations communicating via frequency hopping signals; the frequency hopping signal parameter data stream comprises a plurality of frequency hopping signal parameter vectors, the frequency hopping signal parameter vectors comprising a frequency, an arrival time and a duration of the frequency hopping signal; (2) sorting the cached frequency hopping signal parameter vectors in ascending order according to arrival time; (3) setting the current time to the sum of the arrival time and duration of the last frequency hopping signal in the frequency hopping signal parameter data stream; (4) using the time resolution to normalize the arrival time and duration in the frequency hopping signal parameter vector; (5) The frequency hopping signal parameter vectors are divided according to the duration. The frequency hopping signal parameter vectors with the same duration are divided into the same duration group. The frequency hopping signal parameter vectors in the duration group are still arranged in ascending order according to the arrival time. (6) Screening the duration groups divided in step (5); if the number of frequency hopping signal parameter vectors of a duration group exceeds threshold 1 and the time interval between the current time and the estimated hopping speed of the duration group is greater than threshold 2, performing hopping speed estimation; if the hopping speed estimation is successful, setting the hopping speed of the duration group to the hopping speed estimation value, and setting the estimated hopping speed time of the duration group to the current time; if the hopping speed estimation fails, setting the hopping speed of the duration group to an illegal value; the threshold 1 is a set value M; When the jump rate of the duration group is an illegal value, the threshold 2 is the set time A, and when the jump rate of the duration group is a legal value, the threshold 2 is the set time B, wherein the set time A is less than the set time B; (7) Move the frequency hopping signal parameter vectors in the duration group whose hopping rates are legal values ​​and whose hopping rates differ by no more than the set value C to the same frequency hopping period group, set the frequency hopping period of the newly formed frequency hopping period group to an illegal value, and the frequency hopping signal parameter vectors in the frequency hopping period group are still arranged in ascending order according to the arrival time; If the hopping rate of the duration group is an illegal value, and the number of frequency hopping signal parameter vectors in the duration group exceeds threshold 3 or the time interval between the current time and the arrival time of the first frequency hopping signal parameter vector in the duration group exceeds threshold 4, the frequency hopping signal parameter vector of the duration group is cleared and the estimated hopping rate time is set to the current time; the threshold 3 is a set value D, and the threshold 4 is a set value E; (8) If the frequency hopping period of the frequency hopping period group is an illegal value and the number of the frequency hopping signal parameter vectors in the frequency hopping period group exceeds a threshold of 5, a frequency hopping period estimation value is calculated according to the arrival time of the frequency hopping signal parameter vectors in the frequency hopping period group and the frequency hopping period of the frequency hopping period group is set to the frequency hopping period estimation value; the threshold of 5 is a set value F; (9) For each frequency hopping period group whose frequency hopping period is a legal value, the sequence number and relative arrival time of each frequency hopping signal parameter vector are calculated as clustering features; (10) For each frequency hopping period group whose frequency hopping period is a legal value, frequency hopping signal parameter vectors with a time span of duration 1 are sequentially taken out in chronological order, and data stream clustering is performed on them using the RB-Stream algorithm. The frequency hopping signal parameter vectors in each cluster belong to the same network station; the duration 1 is greater than the product of the frequency hopping period of the frequency hopping period group and the set value H; (11) For each cluster, the state of the network station is determined according to the density of the data flow in the cluster. When the cluster changes from non-existence to existence, the network station is in the appearance state; when the cluster changes from existence to non-existence, the network station disappears; when the data flow density is not less than a threshold of 6, the network station is in the active state; when the data flow density is less than a threshold of 6, the network station is in the silent state; the threshold 6 is a set value N; (12) For each cluster, the reciprocal of the frequency hopping period of the frequency hopping period group to which the cluster belongs is the hopping speed of the corresponding network station. The number of occurrences of the frequency of each frequency hopping signal parameter vector in the cluster is counted at a set time interval, and the frequency with the highest number of occurrences is selected to form the frequency set of the network station corresponding to the cluster; (13) Repeat steps (1) to (12) to achieve real-time sorting and tracking of the network stations.

2. A network station sorting and tracking method based on RB-Stream algorithm according to claim 1, It is characterized in that In the step (6), the calculation method of the hopping rate estimation is the PRI pulse repetition interval transformation method.

3. A network station sorting and tracking method based on RB-Stream algorithm according to claim 1, It is characterized in that In the step (8), the frequency hopping period estimation value is calculated using a frequency hopping period estimation algorithm based on a cost function.

4. A method for sorting and tracking network stations based on RB-Stream algorithm according to claim 1, It is characterized in that In the step (9), the serial number of each frequency hopping signal parameter vector is the integer part of the quotient of the arrival time of the frequency hopping signal parameter vector and the frequency hopping period of the frequency hopping period group to which it belongs, and the relative arrival time of each frequency hopping signal parameter vector is the remainder of the division of the arrival time of the frequency hopping signal parameter vector and the frequency hopping period of the frequency hopping period group to which it belongs.

5. A network station sorting and tracking method based on RB-Stream algorithm according to claim 1, It is characterized in that In the step (12), the specific method of selecting the frequencies with high occurrence times to form the frequency set of the network station corresponding to the cluster is to calculate the mean of the number of occurrences of each vector frequency, and select the frequencies with occurrence times greater than half of the mean to form the frequency set of the network station corresponding to the cluster.