CN112198489A

CN112198489A - Improved maximum likelihood algorithm-based machine-swept radar angle super-resolution angle measurement method

Info

Publication number: CN112198489A
Application number: CN202010944824.XA
Authority: CN
Inventors: 龙腾; 王宏宇; 成芳蕾; 姚笛; 李阳
Original assignee: Beijing Institute of Technology BIT
Current assignee: Beijing Institute of Technology BIT
Priority date: 2020-09-10
Filing date: 2020-09-10
Publication date: 2021-01-08
Anticipated expiration: 2040-09-10
Also published as: CN112198489B

Abstract

The invention discloses a machine scanning radar angle super-resolution angle measurement method based on an improved maximum likelihood algorithm, which constructs a machine scanning radar signal model by analyzing the characteristics of machine scanning radar signals, constructs a maximum likelihood function based on the model according to the characteristics of the model and the properties of the maximum likelihood algorithm, and solves the problem that the information source number needs to be known in advance in the maximum likelihood function by assuming a known mode, so that the method can better meet the actual engineering requirements; meanwhile, the maximum likelihood solution is solved by using intelligent algorithms such as a genetic algorithm or an ant colony algorithm, and the problems of long time and large calculated amount in common space search are solved.

Description

Improved maximum likelihood algorithm-based machine-swept radar angle super-resolution angle measurement method

Technical Field

The invention belongs to the technical field of radar signal processing algorithms, and particularly relates to a machine scanning radar angle super-resolution angle measurement method based on an improved maximum likelihood algorithm.

Background

The common mechanical scanning system radar adopts the rotation of an antenna to change the direction so as to realize scanning, and has simple scanning mode and stable and reliable performance. One principle of the target detection performance of the mechanical scanning radar is as follows: the radar antenna continuously emits radar beams in the scanning process, the radar waves are reflected when meeting a target, the reflected echoes are received by the radar again, and information such as the shape, the size, the speed, the position and the like of the target can be solved through a series of signal processing. However, the above-mentioned detection of the target is based on the fact that there is only one target at the same distance from the same radar beam, and the above-mentioned method cannot accurately distinguish two or more equidistant targets in the same radar beam, and at this time, super-resolution technology is required.

The super-resolution of radar angle is to resolve two or more equidistant targets in one radar beam by using a digital signal processing method under the condition of not changing the working system and hardware system of the radar. For a long time, due to the limitations of the rayleigh criterion, two radiation sources have to be angularly separated by at least a certain angle to be resolved. In this case, to improve the angular resolution of the radar, the operating wavelength may be reduced or the size of the antenna aperture may be increased. In practice, however, reducing the operating wavelength is limited by the performance of the system itself, and increasing the antenna aperture necessarily increases the volume and weight of the radar, greatly reducing its maneuverability and adaptability. Therefore, the method for improving the radar angle resolution by using hardware is not suitable, and we must find another idea, namely, a method of using the "angle super-resolution". On the premise of not changing hardware conditions, the azimuth resolution of the radar is improved by utilizing a digital signal processing mode.

The traditional maximum likelihood algorithm has two defects in the angle super-resolution application: firstly, the essence of the algorithm is the problem of solving an extreme value of a nonlinear function, the solution is difficult and the calculation amount is large; secondly, the number of the information sources needs to be known in advance, and the method cannot be applied to the actual situation that the number of the information sources is unknown.

Disclosure of Invention

In view of the above, the invention provides an angle super-resolution angle measurement method for a mechanical scanning radar based on an improved maximum likelihood algorithm, which can overcome the defects of the traditional maximum likelihood algorithm in angle super-resolution application and improve the angle super-resolution of a mechanical scanning system radar.

The technical scheme for realizing the invention is as follows:

a machine-swept radar angle super-resolution angle measurement method based on an improved maximum likelihood algorithm comprises the following steps:

step one, establishing a signal model of a mechanical scanning radar:

X＝A(θ)S+N

wherein, X is the received echo, A (theta) is a steering matrix, namely the attenuation multiple of the radar transmitted wave in the corresponding direction, S is the radar transmitted wave, and N is noise;

step two, assuming the number of signal sources, and constructing a maximum likelihood function;

and step three, solving the maximum likelihood solution by using an intelligent algorithm, namely determining the number of signal sources and the position of the target.

Further, the process of determining the number of signal sources specifically includes:

if a targets are assumed, b targets actually exist, and a is not less than b, b different results can be obtained through the maximum likelihood function solution.

Further, the intelligent algorithm is a genetic algorithm or an ant colony algorithm.

Has the advantages that:

the method solves the problem that the information source number needs to be known in advance in the maximum likelihood algorithm in a hypothesis mode, so that the method can better meet the actual engineering requirement; meanwhile, the maximum likelihood solution is solved by using the intelligent algorithm of the genetic algorithm and the ant colony algorithm, so that the problems of long time and large calculated amount in common space search are solved.

Drawings

FIG. 1 is a block diagram of the improved maximum likelihood algorithm of the present invention.

FIG. 2 is a block flow diagram of a genetic algorithm.

Fig. 3 is a basic flowchart of the ant colony algorithm for solving the continuous optimization problem.

Detailed Description

The invention is described in detail below by way of example with reference to the accompanying drawings.

The invention provides a machine scanning radar angle super-resolution angle measurement method based on an improved maximum likelihood algorithm, as shown in figure 1, in a machine scanning radar model, an angle measurement task actually estimates an angle vector of a target according to an acquired echo signal. The improved maximum likelihood super-resolution angle measurement method is used, firstly, the number of signal sources is assumed, secondly, a likelihood function is constructed, and an intelligent algorithm is used for solving a maximum likelihood solution, namely, the number of the signal sources and the position of a target are determined.

The specific implementation process of the invention is as follows:

1. construction machine scanning radar signal model

The mechanical scanning radar rotates at a certain angular speed, and when the radar scans, a radar wave is transmitted at every other fixed angle, and a reflected echo is received, which is called a pulse. The machine-swept radar scans the echoes of all the pulses received in one beam width as echo vectors in this beam width.

Establishing a signal model of the mechanical scanning radar:

X＝A(θ)S+N

where X is the received echo, A (θ) is the steering matrix, S is the radar transmitted wave, and N is noise. Wherein, the guiding vector A (theta) is the attenuation multiple of the radar emission wave in the corresponding direction: if the radar wave is emitted to the target, the pulse echo value is maximum, and the element in the corresponding steering matrix is maximum.

2. Assuming the number of signal sources

The maximum likelihood angle super-resolution method needs to know the number of signal sources in advance, but in practical situations, the number of signal sources is generally an unknown number, and a certain method needs to be adopted to solve the problem. The present invention proposes a way to resolve this conflict, assuming known. Taking the maximum of three targets as an example, assuming three targets directly, the following results are obtained:

(1) assume that there are three targets, there is actually one target case: three same results can be obtained through maximum likelihood solving;

(2) assume that there are three targets, and there are actually two targets: two of three results obtained by maximum likelihood solution are the same;

(3) assume that there are three targets, and there are actually three target cases: three different results are obtained by maximum likelihood solving.

Thus, the number of targets can be directly judged.

3. Improved maximum likelihood solution using intelligent algorithm

Assuming that three targets exist, the noise in the swept radar signal model is assumed to be a mean vector which is a zero vector and a covariance matrix which is sigma²Gaussian white noise of M (M is unit array), so that an expression of a maximum likelihood function can be written

Then, the maximum likelihood solution of the three-dimensional target angle vector theta is obtained by means of derivation

Where F (θ) is an expression obtained by maximizing the likelihood function. For solving the problem of solving the extreme point of the nonlinear function F (theta), the invention provides a mode of replacing the traditional multidimensional space scanning by an intelligent algorithm, and solves the problems of large calculated amount and low speed in the maximum likelihood algorithm. Two intelligent algorithms are briefly introduced here:

genetic algorithm: the algorithm is a randomized search algorithm which uses natural selection and natural genetic mechanism in the biology world for reference, problem parameters are coded into chromosomes, and then operations such as selection, intersection, variation and the like are carried out in an iterative mode to exchange chromosome information in a population, so that the chromosomes meeting the required optimization target are finally generated, namely the optimal solution. Encoding the three-dimensional target angle vector theta into a chromosome, and finally obtaining the optimal solution of the theta through a genetic algorithm; the specific flow is shown in fig. 2.

Ant colony algorithm: the algorithm is a random search algorithm for simulating the foraging process of real ants in the nature, when one ant finds food (namely finds a large objective function value), pheromone is released, other ants are attracted to come over, iteration is carried out continuously, and finally a point with the highest pheromone concentration, namely an optimal solution, is formed. taking-F (theta) as a target function, and finally obtaining the optimal solution of theta through iterative search of an ant colony algorithm in an angle space; the specific flow is shown in fig. 3.

In summary, the above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A machine-swept radar angle super-resolution angle measurement method based on an improved maximum likelihood algorithm is characterized by comprising the following steps:

step one, establishing a signal model of a mechanical scanning radar:

X＝A(θ)S+N

2. The method for angle super-resolution angle measurement of the machine-swept radar based on the improved maximum likelihood algorithm according to claim 1, wherein the process for determining the number of signal sources specifically comprises:

3. The method for angle super-resolution angle measurement of machine-swept radar based on the improved maximum likelihood algorithm of claim 1, wherein the intelligent algorithm is a genetic algorithm or an ant colony algorithm.