Dimension reduction method of target reconstruction problem of continuous wave system 1-bite radar

A radar target and bit technology, applied in the field of radar detection, can solve the problems of reducing the dimension of the reconstruction model and cannot be solved, and achieve the effect of reducing the dimension

CN110007282AActive Publication Date: 2019-07-12NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
7 Cites 7 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Publication Date
2019-07-12

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a dimension reduction method of the target reconstruction problem of continuous wave system 1-bite radar. The dimension reduction method the target reconstruction problem of the continuous wave system 1-bite radar comprises the following steps that, a single frame 1-bite data cube is subjected to three-dimensional fast Fourier transform (3D-FFT) through a receiving end, phase-coherent accumulation of target signals is completed in three domains of a slow time domain, a null domain and a fast time domain, and a three-dimensional frequency domain data cube is obtained; the three-dimensional frequency domain data cube is subjected to three-dimensional constant false-alarm pre-detection, target trace point information after being subjected to the pre-detection is obtained, and the target trace point information comprises serial numbers of slow time units, null domain units and fast time units where trace points located; and based on the serial numbers of the slow time units, the null domain units and the fast time units where the pre-detection trace points located, a dimension reduction measurement matrix is calculated, an original measurement matrix is replacedwith the dimension reduction measurement matrix, and a dimension reduction measurement model is obtained. According to the dimension reduction method the target reconstruction problem of the continuous wave system 1-bite radar, the dimensionality of a reconstruction model can be greatly reduced, and the problem that conventional nonlinear reconstruction algorithms cannot be solved due to too highdimensionality is solved.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the technical field of radar detection, in particular to a dimensionality reduction method for 1-bit radar target reconstruction in a continuous wave system. Background technique

[0002] With the emergence of new application requirements such as millimeter-wave radar high-precision detection and air formation cooperative detection, factors such as insufficient device capabilities and limited data transmission bandwidth in complex electromagnetic environments have gradually become challenges for conventional radar systems. For example, the virtual control system based on millimeter-wave radar mounts radar on watches, audio and other equipment, uses radar technology to capture human gestures, and operates machines in the air. The detection accuracy is required to reach the submillimeter level. In this case, the required detection bandwidth can reach several GHz or even more than 10 GHz. At the same time, high requirements are pl...

Examples

Embodiment Construction

[0037] A dimensionality reduction method for a 1-bit radar target reconstruction problem in a continuous wave system, comprising the following steps:

[0038] (1) At the receiving end, the three-dimensional fast Fourier transform 3D-FFT is performed on the single-frame 1-bit data cube, and the coherent accumulation of the target signal is completed in the three domains of the slow time domain, the spatial domain and the fast time domain, and the three-dimensional frequency domain is obtained. domain data cube;

[0039] (2) Carry out three-dimensional constant false alarm pre-detection to the three-dimensional frequency-domain data cube, and obtain the pre-detected target point track information, including the slow time unit, airspace unit and fast time unit number where each point track is located;

[0040] (3) Calculate the dimensionality reduction observation matrix based on the fast time, airspace and slow time unit numbers where the pre-detection points are located, and re...