Multi-type precession warhead parameter estimation method based on double-layer double-channel convolutional neural network

Abstract

The invention discloses a multi-type precession warhead parameter estimation method based on a double-layer double-channel convolutional neural network. The method comprises the following steps: firstly, emitting single-frequency pulses to one type of bullets with different sizes at one viewing angle to obtain main polarization echo data of the same type of bullets with different sizes at the viewing angle, and then emitting single-frequency pulses from another viewing angle to obtain main polarization echo data of the same type of bullets with different sizes at the viewing angle; and then replacing another type of warhead and repeating the operation until all types of warheads have main polarization echoes under two visual angles; making time-frequency transformation on the obtained echodata to obtain a large number of time-frequency graphs; using four time-frequency graphs of the same warhead under two main polarizations at two visual angles as a group of data, and dividing the obtained data into a training set and a test set; building a double-channel double-layer convolutional neural network regression model, training the obtained training set, testing the test set, and obtaining the relative root-mean-square error of each parameter of each type of warhead. The invention can be used for carrying out parameter estimation on multiple types of bullets.

Description

technical field [0001] The invention belongs to the technical field of signal processing, in particular to a multi-type precessing warhead parameter estimation method based on a double-layer and double-channel convolutional neural network. Background technique [0002] When the ballistic missile is flying at high speed in the air, the spin motion maintains the stability of its attitude, and the lateral disturbance will convert the spin motion into a precession form, where the spin refers to the ballistic missile's rotational motion around its own symmetry axis, Motion refers to the rotation of the ballistic missile around the cone axis while spinning. [0003] Space target recognition is a crucial link in the ballistic missile defense system. The mid-stage flight is the longest in the flight process of a ballistic missile, and the space environment is relatively simple. At this time, the target is shown as a small rotation around the center of mass while the target is movin...

AI Technical Summary

Problems solved by technology

The traditional method gives only a method for estimating one kind of warhead, which is not universally applicable.

Images