Unlock instant, AI-driven research and patent intelligence for your innovation.

Adaptive digital beam forming method based on deep learning

A technology of digital beam and synthesis method, which is applied in the field of signal processing, can solve problems such as difficulty in adapting to the electromagnetic environment, inability to know the location of signal arrival, interference source and noise, etc., and achieve good suppression effect

Active Publication Date: 2020-07-24
TSINGHUA UNIV
View PDF5 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Ideally, the weighting vector can be directly set as the direction vector, but since it is usually impossible to know the arrival position of the desired signal and the interference sources and noises in the environment, it is necessary to use an adaptive digital beamforming method so that the desired signal can be received At the same time, it has an inhibitory effect on the signal in the direction of interference
Traditional adaptive beamforming algorithms such as the LCMV method require accurate incoming wave azimuth information, while some other methods such as Bayesian methods also need to give the prior probability distribution of the incoming wave azimuth in advance, and assume that the direction of the interference source is far away from the signal Source direction, it is difficult to adapt to the complex electromagnetic environment

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Adaptive digital beam forming method based on deep learning
  • Adaptive digital beam forming method based on deep learning
  • Adaptive digital beam forming method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

preparation example Construction

[0016] The flow chart of the digital beamforming method of the present invention is attached figure 1 , the specific steps are:

[0017] 1) Collect a large number of array signals containing interference and noise as training data, perform beamforming on the noisy array signals in the training set, and use the mean square error between the calculated result and the real desired signal as a loss function to train the deep neural network model, Complete the training steps;

[0018] 2) The antenna array receiving module receives and samples the signal in real time to obtain a section of signal, regards the signal received by each array element at the same time as a component, and arranges a section of array signal vector according to the spatial sequence;

[0019] 3) Take the complex envelope of the signal vector obtained in step 2), and obtain a section of complex envelope vector, denoted as x(1), x(2), x(3),...;

[0020] 4) Treat the real part and the imaginary part of each c...

Embodiment

[0025] Take a phased array antenna with 12 elements as an example. First, a large amount of signal data is generated through simulation. The direction of arrival of the desired signal and the interference signal in each group of signals is random, and contains interference of different strengths. These data are used as training data to complete the training of the deep neural network. In the test or use link, the antenna array receives a signal with a length of 100, and arranges the signal of each array element as a component at the same time into vectors in spatial order to obtain a signal vector sequence with a length of 100. Each vector in the sequence Both are 12-dimensional vectors. Take the complex envelope for each signal vector to obtain a sequence of complex envelope vectors with a length of 100, denoted as {x(1),x(2),...,x(100)}. The real part and the imaginary part of each component of the complex envelope vector are regarded as one channel respectively, then the a...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an adaptive digital beam forming method based on deep learning. The method comprises the steps of: arranging narrowband signals received by each array element of the antenna array in sequence to obtain an array signal vector, using Hilbert transform of the array signal vector as an imaginary part, using an original narrowband signal as a real part, and obtaining the analytic signal vector of the array signal vector; regarding the real part and the imaginary part of each element of the analytic signal vector as a signal channel respectively, and inputting the signal channels into a pre-trained deep neural network model to obtain a beamforming weighted vector; and obtaining an expected signal by calculating the inner product of the weighted vector and the array signalvector. The method can be suitable for the condition that the arrival directions and the intensity of the expected signal and the interference signal are unknown, adaptively adjusts the beam pointingdirection, can effectively enhance the expected signal and suppress the interference signal, and has very high robustness.

Description

technical field [0001] The present invention relates to the field of signal processing, in particular, to an adaptive digital beamforming method based on deep learning. Background technique [0002] In electronic communication systems, receivers often send and receive signals through phased array antennas to improve the flexibility of beam pointing. Ideally, the weighting vector can be directly set as the direction vector, but since it is usually impossible to know the arrival position of the desired signal and the interference sources and noises in the environment, it is necessary to use an adaptive digital beamforming method so that the desired signal can be received At the same time, it has an inhibitory effect on the signal in the interference direction. Traditional adaptive beamforming algorithms such as the LCMV method require accurate incoming wave azimuth information, while some other methods such as Bayesian methods also need to give the prior probability distribut...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): H04B7/08
CPCH04B7/088
Inventor 罗东琦司宾强朱纪洪
Owner TSINGHUA UNIV