Feature classification target detection method based on improved normalized Doppler spectrum peak-to-average ratio

Abstract

This paper proposes a feature-based target detection method based on improved normalized Doppler peak-to-average ratio (PPR). The authors found that when the number of target unit samples is small, the target's motion speed degrades the FPAR feature's ability to distinguish, thereby reducing target detection performance. Further research revealed that this degradation varies periodically with speed. Therefore, this performance degradation can be addressed by pre-multiplying the target unit by a Doppler steering vector.

Description

Technical Field

[0001] The present invention relates to the field of radar sea detection. Background Art

[0002] Currently, the most commonly used radar target detector is an incoherent energy detector. It uses only the difference in echo energy between sea clutter and the target for target detection, resulting in a sharp decline in detection performance at low signal-to-clutter ratios. Detection based on the characteristics of sea clutter and the target can reflect other differences between the two, allowing for target detection even at low signal-to-clutter ratios. The peak-to-average ratio (FPAR) of the Doppler spectrum of the echo signal is one feature that can be used to distinguish sea clutter from targets. However, in practice, using FPAR for target detection requires a large number of target unit samples, placing significant demands on radar resources. Summary of the Invention

[0003] In order to solve the problem that the number of correlated pulses in scanning mode is difficult to support FPAR feature target detection, the present invention proposes a feature classification target detection method based on improved normalized Doppler spectrum peak-to-average ratio.

[0004] The technical solution to implement the present invention is: after selecting the unit to be detected, use different Doppler steering vectors to multiply it, and then calculate the corresponding Doppler spectrum peak-to-mean ratio. The specific steps are as follows:

[0005] Step 1: Use radar to detect the sea surface and obtain radar echo video data including fast time (distance) and slow time (time or azimuth) dimensions.

[0006] Step 2: Use the sliding window method to select the target unit f(t) to be detected with a specific number of coherent processing units at a specific distance unit.

[0007] Step 3: Calculate the Doppler steering vector p corresponding to different target movement speeds with a certain step size k .

[0008] Step 4: Direct different Doppler vectors p k Multiply it by the corresponding element of the target unit to be detected to obtain a new set of units to be detected f k (t).

[0009] Step 5: For each f in the newly reorganized unit to be detected k (t), calculate its normalized Doppler spectrum, and calculate the corresponding peak-to-average ratio FPAR based on the obtained normalized Doppler spectrum k , and select the maximum value as the final corrected peak-to-average ratio FPAR.

[0010] Step 6: Calculate the corrected peak-to-average ratio (FPAR) of the pure sea clutter unit to be detected and the corrected peak-to-average ratio (FPAR) of the unit to be detected containing a target according to the above steps, and use them as the corrected FPAR feature datasets of sea clutter and target.

[0011] Step 7: Use the feature dataset from step 6 to train the classification algorithm model to obtain a classifier that can be used to classify new data and achieve target detection in sea clutter background.

[0012] Furthermore, the unit to be detected and the Doppler steering vector have the same dimension.

[0013] Furthermore, in step 3, the target velocity ranges of all Doppler steering vectors cover the Doppler resolvable interval.

[0014] Furthermore, in step 5, the normalized Doppler spectrum is defined as a new sequence obtained by comparing each element in the Doppler spectrum sequence with the maximum value of the Doppler spectrum.

[0015] Furthermore, the peak-to-average ratio is defined as the ratio of the maximum value to the average value of a function or sequence.

[0016] The present invention reduces the number of samples required for a unit to be detected, saves radar resources, and promotes the practical application of a feature-based detection algorithm that was originally difficult to apply in radar target detection.

[0017] The present invention will be described in further detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 Flowchart of the target detection method based on feature classification of improved normalized Doppler spectrum peak-to-average ratio DETAILED DESCRIPTION

[0019] The present invention will be further described below with reference to the accompanying drawings, but the protection scope of the present invention is not limited by the implementing regulations.

[0020] The present invention proposes a feature classification target detection method based on improved normalized Doppler spectrum peak-to-average ratio. The preferred embodiment steps are as follows:

[0021] Step 1: Use radar to detect the sea surface and obtain radar echo video data including fast time (distance) and slow time (time or azimuth) dimensions.

[0022] Step 2: Use the sliding window method to select the target unit f(t) to be detected with a specific number of coherent processing units at a specific distance unit.

[0023] Step 3: Calculate the Doppler steering vector p corresponding to different target movement speeds with a certain step size k :

[0024] p(n)=exp(j2πf d nT r ),n=0,1,2,…,M-1 (1)

[0025] Step 4: Direct different Doppler vectors p k Multiply it by the corresponding element of the target unit to be detected to obtain a new set of units to be detected f k (t).

[0026] Step 5: For each f in the newly reorganized unit to be detected k (t), calculate its normalized Doppler spectrum, and calculate the corresponding peak-to-average ratio FPAR based on the obtained normalized Doppler spectrum k , select the maximum value as the final corrected peak-to-average ratio FPAR:

[0027]

[0028] Where, f k (t) is the time series of the unit to be detected, F k (f) is its frequency spectrum, abs[˙] represents the absolute value of the elements in the sequence, max[˙] represents the maximum value in the sequence, and mean[˙] represents the arithmetic mean of the sequence.

[0029] Step 6: Calculate the corrected peak-to-average ratio (FPAR) of the pure sea clutter unit to be detected and the corrected peak-to-average ratio (FPAR) of the unit to be detected containing a target according to the above steps, and use them as the corrected FPAR feature datasets of sea clutter and target.

[0030] Step 7: Use the feature dataset from step 6 to train the classification algorithm model to obtain a classifier that can be used to classify new data and achieve target detection in sea clutter background.

Claims

1. A feature classification target detection method based on improved normalized Doppler spectrum peak-to-average ratio, characterized by: Step 1: Use radar to detect the sea surface and obtain radar echo video data including fast time and slow time dimensions; Step 2: Use the sliding window method to select the target unit f(t) to be detected with a specific number of coherent processing units at a specific distance unit; Step 3: Calculate the Doppler steering vector p corresponding to different target movement speeds with a certain step size k ; Step 4: Direct different Doppler vectors p k Multiply it by the corresponding element of the target unit to be detected to obtain a new set of units to be detected f k (t); Step 5: For each f in the newly reorganized unit to be detected k (t), calculate its normalized Doppler spectrum, and calculate the corresponding peak-to-average ratio FPAR based on the obtained normalized Doppler spectrum k , select the maximum value as the final corrected peak-to-average ratio FPAR; Step 6: Calculate the corrected peak-to-average ratio (FPAR) of the pure sea clutter unit to be detected and the corrected peak-to-average ratio (FPAR) of the unit to be detected containing the target according to the above steps, and use them as the corrected FPAR feature data sets of sea clutter and target. Step 7: Use the feature dataset from step 6 to train the classification algorithm model to obtain a classifier that can be used to classify new data and achieve target detection in sea clutter background.

2. The feature classification target detection method based on improved normalized Doppler spectrum peak-to-average ratio according to claim 1, characterized in that: In step 2 and step 3, the unit to be detected and the Doppler steering vector have the same dimension.

3. The feature classification target detection method based on improved normalized Doppler spectrum peak-to-average ratio according to claim 1, characterized in that: In step 3, the target velocity ranges of all Doppler steering vectors cover the Doppler resolvable interval.

4. The feature classification target detection method based on improved normalized Doppler spectrum peak-to-average ratio according to claim 1, characterized in that: In step 5, the normalized Doppler spectrum is defined as a new sequence obtained by comparing each element in the Doppler spectrum sequence with the maximum value of the Doppler spectrum.

5. The feature classification target detection method based on improved normalized Doppler spectrum peak-to-average ratio according to claim 1, characterized in that: The peak-to-average ratio is defined as the ratio of the maximum value to the average value of a function or series.

Images