Middle repetition frequency PD radar time-frequency domain two-dimensional sliding window ambiguity resolution method and device

Abstract

The invention provides a medium repetition frequency PD radar time-frequency domain two-dimensional sliding window ambiguity resolution method and device, and the method comprises the steps: 1, carrying out the distance dimension and speed dimension extension of a distance trace point formed by a multi-frequency period radar echo in an observation range, and determining all possible position and speed information of a target; step 2, quantizing the observation area according to the set quantization unit parameters, and marking all targets in a nine-square manner at the same time; 3, sliding window detection is sequentially carried out on the observation areas after quantitative marking in a matts mode, and whether a target exists or not is judged according to an M / N detection criterion; and 4, detecting and judging the existing target through the step 3, extracting target information by taking the maximum amplitude as a criterion, and clearing the identifier of the observation area of the target. According to the method, the difficulty and the calculation amount of fuzzy understanding are greatly reduced, and meanwhile, the generation of false plots is also reduced by utilizing distance and speed information.

Description

Technical Field

[0001] The present application relates to the field of radar signal processing technology, and in particular to a two-dimensional sliding window deambiguation method and device for a medium-repetition-rate PD radar in the time-frequency domain. Background Art

[0002] Pulse Doppler (PD) technology is currently widely used in various radar equipment due to its excellent clutter suppression performance. Compared with low repetition rate and high repetition rate modes, the radar echo in medium repetition rate mode has both range and speed ambiguity in both range and Doppler dimensions, and is also affected by the obstruction of range blind spots and speed blind spots.

[0003] In order to eliminate the ambiguity and obstruction problems, radar systems generally adopt a multi-frequency staggered working mode. The difficulty of deambiguation in medium-frequency PD radar systems lies in how to calculate the real position and speed information of the target with high reliability and accuracy.

[0004] The traditional algorithm itself is limited to the number theory settlement in a single dimension, and does not use the most important characteristic information of PD radar - Doppler frequency. Moreover, the reliability, accuracy and calculation amount of the solution are greatly affected by the algorithm. Summary of the invention

[0005] The present application provides a two-dimensional sliding window deambiguation method and device in the time-frequency domain of a medium-repetition-frequency PD radar, which can be used to solve the technical problem that traditional algorithms do not utilize the most important characteristic information of the PD radar - the Doppler frequency.

[0006] The present application provides a two-dimensional sliding window deambiguation method in the time-frequency domain of a medium-repetition-frequency PD radar, the method comprising:

[0007] Step 1: Extend the distance point trace formed by the multi-frequency periodic radar echo in the observation range in the distance dimension and the speed dimension to determine all possible position and speed information of the target;

[0008] Step 2: quantize the observation area with the set quantization unit parameters, and mark all targets in a "nine-square grid" manner;

[0009] Step 3: Perform sliding window detection on the quantitatively marked observation area in a grid-like manner, and determine whether the target exists using the "M / N" detection criterion;

[0010] Step 4: Detect and determine the target in step 3, extract the target information based on the maximum amplitude, and clear the mark of the observation area of ​​the target.

[0011] Furthermore, in step 1, the distance trace formed by the multi-frequency periodic radar echo is extended in the distance dimension and the speed dimension within the observation range to determine all possible position and speed information of the target, including:

[0012] Step 11, the parameters of each period of radar multi-frequency difference are PRT_1, PRT_2...PRT_N, then the maximum unambiguous distance of each period is:

[0013] RanBlind_N=PRT_N*C / 2, where C is the speed of light;

[0014] The point distance is:

[0015] Plot_Ran=Plot_Ran_Vision+k*RanBlind_N,

[0016] Where Plot_Ran_Visio is the apparent distance of the point trace, k is the expansion multiple, which is an integer;

[0017] Step 12: According to the radar multi-frequency staggered period parameters, the maximum unambiguous speed of each period is:

[0018] VelBlind_N=λ / (2*PRT_N),

[0019] Where λ is the radar operating wavelength, and the point velocity is:

[0020] Plot_Vel=Plot_Vel_Vision+k*VelBlind_N

[0021] Where Plot_Vel_Vision is the apparent velocity of the point trace, and k is the expansion multiple, which is an integer.

[0022] Furthermore, the observation area is quantized with the set quantization unit parameters, and all targets are marked in a "nine-square grid" manner, including:

[0023] Step 21, set the quantization unit parameters according to the radar performance requirements. Assuming that the maximum observation distance of the radar observation area is Detect_Ran_Max, the maximum observation speed is Detect_Vel_Max, the radar beam dwell time is Time_Dwell, and the sampling rate is Fs, the distance quantization unit parameter Thelta_Ran satisfies:

[0024]

[0025] The velocity quantization unit parameter Thelta_Vel satisfies:

[0026]

[0027] Step 22, mark all target points calculated in step 1 in a "nine-square grid" manner:

[0028] Calculate each point trace distance mark index Plot_Ran_Index=round(Plot_Ran / Thelta_Ran), where Plot_Ran is the point trace distance, and the point trace velocity mark index Plot_Vel_Index=round(Plot_Vel / Thelta_Vel), where Plot_Vel is the point trace velocity;

[0029] Then mark the index area [Plot_Ran_Index-1:Plot_Ran_Index+1, Plot_Vel_Index-1:Plot_Vel_Index+1], and the index area is the "nine-square grid".

[0030] Furthermore, the M / N detection criterion means that N refers to the number of radar multiple frequency difference cycles, and M refers to the detection threshold of the number of target echoes in the radar difference cycle. N and M are both integers and M≤N; if M or more radar difference cycle echoes appear within N radar multiple frequency difference cycles, the target is determined to exist. For example, the "2 / 4" detection criterion is illustrated. The number of radar multiple frequency difference cycles is 4. When there are echoes in 2 or more radar difference cycles at the same time.

[0031] Furthermore, the observation area after quantitative marking is sequentially subjected to sliding window detection in a grid-like manner, and the existence of the target is determined by the "M / N" detection criterion, including:

[0032] The sliding window method of the grid means that if the coordinates of the current detection point are [m,n], [m+1,n], [m,n+1] and [m+1,n+1] are used as joint judgment conditions at the same time. Only when the four coordinate identifiers in the grid meet the M / N criterion, the corresponding position is judged to be the real target point.

[0033] The present application also provides a two-dimensional sliding window defuzzification device in the time-frequency domain of a medium-frequency PD radar, the device comprising:

[0034] A determination module is used to extend the distance point trace formed by the multi-frequency periodic radar echo within the observation range in the distance dimension and the speed dimension to determine all possible position and speed information of the target;

[0035] The marking module is used to quantify the observation area with the set quantization unit parameters and mark all targets in a "nine-square grid" manner;

[0036] The judgment module is used to perform sliding window detection on the quantitatively marked observation area in turn in a grid-like manner, and determine whether the target exists according to the "M / N" detection criterion;

[0037] The zeroing module is used to detect and determine the existence of a target, extract target information based on the maximum amplitude as the criterion, and clear the identification of the observation area of ​​this target.

[0038] Furthermore, the module is specifically used to:

[0039] The parameters of each period of radar multi-frequency staggered are PRT_1, PRT_2…PRT_N, and the maximum unambiguous distance of each period is:

[0040] RanBlind_N=PRT_N*C / 2, where C is the speed of light;

[0041] The point distance is:

[0042] Plot_Ran=Plot_Ran_Vision+k*RanBlind_N,

[0043] Where Plot_Ran_Visio is the apparent distance of the point trace, k is the expansion multiple, which is an integer;

[0044] According to the radar multi-frequency staggered period parameters, the maximum unambiguous speed of each period is:

[0045] VelBlind_N=λ / (2*PRT_N),

[0046] Where λ is the radar operating wavelength, and the point velocity is:

[0047] Plot_Vel=Plot_Vel_Vision+k*VelBlind_N

[0048] Where Plot_Vel_Vision is the apparent velocity of the point trace, and k is the expansion multiple, which is an integer.

[0049] Furthermore, the marking module is specifically used for:

[0050] The quantization unit parameters are set according to the radar performance requirements. Assuming that the maximum observation distance of the radar observation area is Detect_Ran_Max, the maximum observation speed is Detect_Vel_Max, the radar beam dwell time is Time_Dwell, and the sampling rate is Fs, the distance quantization unit parameter Thelta_Ran satisfies:

[0051]

[0052] The velocity quantization unit parameter Thelta_Vel satisfies:

[0053]

[0054] Mark all target points in a "nine-square grid" format:

[0055] Calculate each point trace distance mark index Plot_Ran_Index=round(Plot_Ran / Thelta_Ran), where Plot_Ran is the point trace distance, and the point trace velocity mark index Plot_Vel_Index=round(Plot_Vel / Thelta_Vel), where Plot_Vel is the point trace velocity;

[0056] Then mark the index area [Plot_Ran_Index-1:Plot_Ran_Index+1, Plot_Vel_Index-1:Plot_Vel_Index+1], and the index area is the "nine-square grid".

[0057] Furthermore, the M / N detection criterion indicates that N refers to the number of radar multiple frequency variation cycles, and M refers to the detection threshold of the number of times the target appears in the radar variation cycle echo, N and M are both integers and M≤N; if M or more radar variation cycle echoes appear within N radar multiple frequency variation cycles, the target is determined to exist; taking the "2 / 4" detection criterion as an example, the number of radar multiple frequency variation cycles is 4, and when it appears in 2 or more radar variation cycle echoes at the same time, the target is determined to exist.

[0058] Furthermore, the judgment module is specifically used for:

[0059] The sliding window method of the grid means that if the coordinates of the current detection point are [m,n], [m+1,n], [m,n+1] and [m+1,n+1] are used as joint judgment conditions at the same time. Only when the four coordinate identifiers in the grid meet the M / N criterion, it is judged as the real target point.

[0060] Compared with the prior art, the present invention has the following significant advantages: (1) It adopts the two-dimensional quantization method of the distance and speed in the radar detection area, which can simultaneously resolve the distance and speed ambiguity, with small amount of calculation and universality and practicality. (2) The targets are clustered and identified in the manner of "nine-square grid", which increases the margin of target deambiguation and reduces the difficulty of deambiguation. (3) It adopts the "M / N" detection criterion and uses the "grid" sliding window detection method to effectively reduce the generation of false targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A flowchart of a two-dimensional sliding window deambiguation method in the time-frequency domain of a medium-repetition-frequency PD radar provided in an embodiment of the present application;

[0062] Figure 2 A blind area coverage diagram under the 2 / 4 detection criterion under the 4-fold frequency cycle provided in the embodiment of the present application;

[0063] Figure 3 A schematic diagram of the "Nine-square grid" marking provided in an embodiment of the present application;

[0064] Figure 4 A schematic diagram of the “grid” sliding window detection provided in an embodiment of the present application. DETAILED DESCRIPTION

[0065] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0066] The present invention proposes a two-dimensional sliding window deambiguation method in the time-frequency domain of a medium repetition frequency PD radar. When the radar adopts a medium repetition frequency staggered working mode, the method can simultaneously perform deambiguation calculations on the target distance and speed, while ensuring high reliability and low calculation amount of deambiguation, which has certain reference value for the engineering implementation of pulse Doppler technology.

[0067] The present application is further described below through specific embodiments.

[0068] The radar parameters are simulated by matlab software, with wavelength λ = 0.1, pulse width Tr = 10us, and 4-fold frequency variation period PRT_N = [287121203164]us. The maximum distance of the radar observation area is set to Detect_Ran_Max as 120km, and the maximum speed is set to Detect_Vel_Max = 500m / s. The simulation target parameters are set to a distance of 90km and a speed of 300m / s.

[0069] In step 1, the distance point trace formed by the multi-frequency radar echo is extended in the distance dimension and the speed dimension within the observation range to calculate the position and speed information of all possible target points. The specific implementation method is:

[0070] Step 11, according to the PRT_N parameter and the speed of light parameter, the maximum unambiguous distance RanBlind_N corresponding to each period is calculated as [86100, 36300, 60900, 49200] m. The apparent distance of the simulation target in each staggered period is:

[0071] Plot_Ran_Visio_N = [3900, 17400, 0, 40800] m, where the target is in the speed blind spot in the fourth cycle, so there is no target in this cycle. The point traces are extended in the detection range in three staggered cycles in turn. The point trace distances formed in the four staggered cycles are: Plot_Ran_Tmp1 = [3900, 90000] m, Plot_Ran_Tmp2 = [17400, 53700, 90000] m, and Plot_Ran_Tmp4 = [40800, 90000] m.

[0072] Step 12, according to the simulation wavelength λ and the staggered period PRT_N parameters, the maximum unambiguous speed is calculated as VelBlind_N is [174.2, 413.2, 246.3, 304.9] m / s, the apparent speed of the simulated target in each staggered period is Plot_Vel_Visio_N is [125.78, -113.22, 53.7, 0] m / s, where the target is in the speed blind zone in the 4th period, so there is no target in this period, and the target is in the 3rd period. The point trace is extended at a speed within the detection range at different periods, and the point trace velocities formed are: Plot_Vel_Tmp1 = [-396.87, -222.65, -48.44, 125.78, 300.0, 474.21] m / s, Plot_Vel_Tmp2 = [-113.22, 300.0] m / s, Plot_Vel_Tmp3 = [-438.9, -192.6, 53.7, 300] m / s.

[0073] In step 2, the observation area is quantized with the set quantization unit parameters, and all the targets calculated in step 1 are marked in a "nine-square grid" manner. The specific implementation steps are as follows:

[0074] Step 21, set the quantization unit parameters according to the radar performance requirements. The radar beam dwell time is Time_Dwell = 12400us. According to the constraints, the distance quantization parameters Thelta_Ran = 10m, Thelta_Vel = 5m / s can be set. The radar position and the minimum speed boundary are used as the quantization center. The observation area is the maximum distance dimension of 120km. The corresponding quantization area is [0, 12000], the speed range is [-1000, 1000], and the quantization area is [0, 400]. Therefore, the distance between the observation area and the speed quantization area is [12000, 400]. The horizontal axis is the distance and the vertical axis is the speed.

[0075] Step 22, according to the calculation in step 1, the number of traces plot_num1 formed by the target in the first cycle is: 2*6=12, the number of traces formed by the target in the second cycle is: 3*2=6, the number of traces formed by the target in the third cycle is: 2*4=8, the total number of traces is 26, and the index Plot_Ran_Index corresponding to each trace is calculated as ([390, -79], [9000, -79], [390, -45], [9000, -45], [390, -10], [9000, -10], [390, 25], [9000, 25], All the points are marked in the form of a "nine-square grid" in the observation area.

[0076] In step 3, the "M / N" criterion is set to 2 / 4. Only when all point marks in the "grid" meet the "M / N" criterion, it is determined that there is a target point here. Finally, the mark at the quantization area [8999:9001, 59:61] is 3, which meets the judgment criterion and is determined to be the target point.

[0077] Step 4: Determine the target existence area [8999:9001, 200:202] through step 3, extract the real information of the target based on the maximum amplitude of the point traces in the three cycles, clear the mark of the quantization area, and continue with step 3 to find the target point.

[0078] The present invention greatly reduces the difficulty and amount of calculation for deambiguation, and at the same time, by using distance and speed information, also reduces the generation of false traces. The present invention has the advantages of small amount of calculation, strong versatility and practicality, and can be well applied to the engineering implementation of medium-frequency PD radar deambiguation.

[0079] The above-described embodiments of the present application do not constitute a limitation on the protection scope of the present application.

Claims

1. A two-dimensional sliding window deambiguation method in time-frequency domain for medium repetition rate PD radar, characterized in that: The method comprises: Step 1: Extend the distance point trace formed by the multi-frequency periodic radar echo in the observation range in the distance dimension and the speed dimension to determine all possible position and speed information of the target; Step 2: quantize the observation area with the set quantization unit parameters, and mark all targets in a "nine-square grid" manner; Step 3: Perform sliding window detection on the quantitatively marked observation area in a grid-like manner, and determine whether the target exists based on the "M / N" detection criterion; Step 4: Detect and determine the target in step 3, extract the target information based on the maximum amplitude, and clear the mark of the observation area of ​​the target.

2. The method according to claim 1, characterized in that In step 1, the distance traces formed by the multi-frequency periodic radar echoes are extended in the distance dimension and the speed dimension within the observation range to determine all possible position and speed information of the target, including: Step 1-1, the parameters of each period of radar multi-frequency are PRT_1, PRT_2…PRT_N, then the maximum unambiguous distance of each period is: RanBlind_N=PRT_N*C / 2, where C is the speed of light; The point trace distance is: Plot_Ran=Plot_Ran_Vision+k*RanBlind_N, Where Plot_Ran_Visio is the apparent distance of the point trace, k is the expansion multiple, which is an integer; Step 1-2, according to the radar multi-frequency staggered period parameters, the maximum unambiguous speed of each period is: VelBlind_N=λ / (2*PRT_N), Where λ is the radar operating wavelength, and the point velocity is: Plot_Vel=Plot_Vel_Vision+k*VelBlind_N Where Plot_Vel_Vision is the apparent velocity of the point trace, and k is the expansion multiple, which is an integer.

3. The method according to claim 1, characterized in that The observation area is quantified with the set quantization unit parameters, and all targets are marked in a "nine-square grid" format, including: Step 2-1, set the quantization unit parameters according to the radar performance requirements. Assuming that the maximum observation distance of the radar observation area is Detect_Ran_Max, the maximum observation speed is Detect_Vel_Max, the radar beam dwell time is Time_Dwell, and the sampling rate is Fs, the distance quantization unit parameter Thelta_Ran satisfies: The velocity quantization unit parameter Thelta_Vel satisfies: Step 2-2, mark all target points calculated in step 1 in a "nine-square grid" format: Calculate each point trace distance mark index Plot_Ran_Index=round(Plot_Ran / Thelta_Ran), where Plot_Ran is the point trace distance, and the point trace velocity mark index Plot_Vel_Index=round(Plot_Vel / Thelta_Vel), where Plot_Vel is the point trace velocity; Then mark the index area [Plot_Ran_Index-1:Plot_Ran_Index+1, Plot_Vel_Index-1:Plot_Vel_Index+1], and the index area is the "nine-square grid".

4. The method according to claim 1, characterized in that: The M / N detection accuracy means that N refers to the number of radar multiple frequency difference cycles, and M refers to the detection threshold of the number of times the target echo appears in the radar difference cycle. N and M are both integers and M≤N; if M or more radar difference cycle echoes appear within N radar multiple frequency difference cycles, the target is determined to exist.

5. The method according to claim 4, characterized in that The observation area marked with quantitative information is subjected to sliding window detection in turn in a grid-like manner, and the "M / N" detection criteria are used to determine whether the target exists, including: The sliding window method of the grid means that if the coordinates of the current detection point are [m,n], [m+1,n], [m,n+1] and [m+1,n+1] are used as joint judgment conditions at the same time. Only when the four coordinate identifiers in the grid meet the M / N criterion, the corresponding position is judged to be the real target point.

6. A two-dimensional sliding window defuzzification device in time-frequency domain for medium repetition frequency PD radar, characterized in that: The device comprises: A determination module is used to extend the distance point trace formed by the multi-frequency periodic radar echo within the observation range in the distance dimension and the speed dimension to determine all possible position and speed information of the target; The marking module is used to quantify the observation area with the set quantization unit parameters and mark all targets in a "nine-square grid" manner; The judgment module is used to perform sliding window detection on the quantitatively marked observation area in a grid-like manner, and determine whether the target exists according to the "M / N" detection criterion; The zeroing module is used to detect and determine the existence of a target, extract target information based on the maximum amplitude as the criterion, and clear the identification of the observation area of ​​this target.

7. The device according to claim 6, characterized in that Determine the module specifically for: The parameters of each period of radar multi-frequency are PRT_1, PRT_2…PRT_N, and the maximum unambiguous distance of each period is: RanBlind_N=PRT_N*C / 2, where C is the speed of light; The point trace distance is: Plot_Ran=Plot_Ran_Vision+k*RanBlind_N, Where Plot_Ran_Visio is the apparent distance of the point trace, k is the expansion multiple, which is an integer; According to the radar multi-frequency staggered period parameters, the maximum unambiguous speed of each period is: VelBlind_N=λ / (2*PRT_N), Where λ is the radar operating wavelength, and the point velocity is: Plot_Vel=Plot_Vel_Vision+k*VelBlind_N Where Plot_Vel_Vision is the apparent velocity of the point trace, and k is the expansion multiple, which is an integer.

8. The device according to claim 6, characterized in that The tag module is specifically used for: The quantization unit parameters are set according to the radar performance requirements. Assuming that the maximum observation distance of the radar observation area is Detect_Ran_Max, the maximum observation speed is Detect_Vel_Max, the radar beam dwell time is Time_Dwell, and the sampling rate is Fs, the distance quantization unit parameter Thelta_Ran satisfies: The velocity quantization unit parameter Thelta_Vel satisfies: Mark all target points in a "nine-square grid" format: Calculate each point trace distance mark index Plot_Ran_Index=round(Plot_Ran / Thelta_Ran), where Plot_Ran is the point trace distance, and the point trace velocity mark index Plot_Vel_Index=round(Plot_Vel / Thelta_Vel), where Plot_Vel is the point trace velocity; Then mark the index area [Plot_Ran_Index-1:Plot_Ran_Index+1, Plot_Vel_Index-1:Plot_Vel_Index+1], and the index area is the "nine-square grid".

9. The device according to claim 6, characterized in that The M / N detection criterion means that N refers to the number of radar multiple frequency difference cycles, and M refers to the detection threshold of the number of times the target echo appears in the radar difference cycle. N and M are both integers and M≤N; if M or more radar difference cycle echoes appear within N radar multiple frequency difference cycles, the target is judged to exist.

10. The device according to claim 9, characterized in that The judgment module is specifically used for: The sliding window method of the grid means that if the coordinates of the current detection point are [m,n], [m+1,n], [m,n+1] and [m+1,n+1] are used as joint judgment conditions at the same time. Only when the four coordinate identifiers in the grid meet the M / N criterion, it is judged as the real target point.

Images