Real value rooting multiple signal classification method and system, computer equipment and application

Abstract

The invention belongs to the technical field of radars, and discloses an actual value rooting multiple signal classification method and system, computer equipment and application. The method comprises the following steps: carrying out dimension reducing treatment on received data; splicing a real part and an imaginary part of the received data in a low-dimensional space by using the non-circular characteristic of the signal to obtain real value received data with the number of virtual array elements doubled; constructing a new steering vector and a new noise subspace, and constructing a real coefficient polynomial according to the orthogonality of the new steering vector and the new noise subspace; and obtaining the estimation of the target orientation by solving the root of the real coefficient polynomial. The angle estimation precision of a subspace estimation algorithm can be greatly improved, the problem that the estimation precision of the subspace algorithm is insufficient under the condition that the signal-to-noise ratio is low and the number of snapshots is limited is solved, and a better reality scheme is provided for MIMO radar target angle estimation on occasions with low snapshots and the low signal-to-noise ratio. Meanwhile, the problem of calculation complexity is considered, and the real-time processing pressure of a radar system is relieved by jointly adopting a dimension reduction method and a real-valued method.

Description

technical field [0001] The invention belongs to the technical field of radar, and in particular relates to a real-valued root-finding multiple signal classification method, system, computer equipment and application. Background technique [0002] At present: multiple input multiple output (MIMO) radars use multiple transmitting antennas to transmit different waveforms, and simultaneously use multiple receiving antennas to simultaneously receive target reflected signals. Compared with traditional phased array radar, MIMO radar has more potential advantages in target detection, anti-jamming, target parameter estimation and target identification, etc., so it has received extensive attention from the academic community. [0003] Direction of arrival (DOA) estimation is an important content of MIMO radar parameter estimation. ESPRIT algorithm and MUSIC algorithm are classic high-resolution estimation methods based on subspace, and obtain DOA estimation of the target based on the...

AI Technical Summary

Problems solved by technology

Since the estimation accuracy of the ESPRIR algorithm itself is not as good as the root-finding MUSIC algorithm, this type of algorithm is also difficult to meet the needs of practical applications under the conditions of low signal-to-noise ratio and low beat speed.

In addition, the angular resolution of the ESPRIR algorithm is not enough, and it cannot be resolved or fails to resolve targets with close azimuths, which also limits the application range of the ESPRIR algorithm

[0005] To sum up, the problems existing in the prior art are: the MUSIC algorithm and the ESPRIT algorithm are classic subspace DOA estimation algorithms, they both need to perform subspace estimation on the autocorrelation matrix of the received signal, and use the characteristics of the subspace to obtain the target The DOA estimate of

[0007] (1) The estimation accuracy and angular resolution of the ESPRIT algorithm itself is not as good as that of the root-seeking MUSIC method. Under the conditions of low SNR and low number of snapshots, the estimation performance of the algorithm will be severely degraded due to inaccurate subspace estimation, and even may cause the estimate to fail

Moreover, the ESPRIT algorithm cannot distinguish between two targets with close azimuths, and its application range is limited.

Some algorithms use the non-circular characteristics of the signal to improve the estimation accuracy of the ESPRIT algorithm, but in many cases it still cannot meet the needs of practical applications, and there is still room for improvement in the estimation performance

[0008] (2) Although the estimation performance of the root-finding MUSIC algorithm is better than that of the ESPRIT algorithm, it is also limited by the signal-to-noise ratio and the number of snapshots.

Although the root-finding MUSIC algorithm uses polynomial root-finding instead of spectral peak search, it still needs to perform the eigenvalue decomposition of the high-dimensional complex covariance matrix and complex coefficient polynomial root-finding operations, so its computational complexity is still very high, which will increase the radar system. Real-time processing pressure, which brings difficulties to the practical application of the algorithm

[0009] The difficulty in solving the above problems and defects is: At present, there are few literatures that combine the non-circular characteristics of the signal with the root-finding MUSIC algorithm. The noise subspace and the new steering vector, it is difficult to effectively integrate the non-circular characteristics with the MUSIC algorithm

In addition, the expanded received data matrix will bring additional calculations, which will increase the original calculation cost. Therefore, on the basis of ensuring the improvement of estimation accuracy, how to take effective measures to optimize the complexity of the algorithm is also a difficulty in solving the above problems.

Images