Research On Distributed UAV-borne Cooperative Target Detection And Parameter Estimation Technology

The system design and signal processing technology of distributed MIMO radar is a novel research direction in the field of radar.MIMO radar can utilize different diversity methods such as space diversity and waveform diversity,it has certain advantages over traditional phased array radars in some application scenarios.For example,space diversity enables distributed MIMO radars to fight well against target rcs flicker,thus achieving better target detection performance;waveform diversity makes MIMO radar to obtain higher target parameter estimation accuracy easily.Combining distributed MIMO radar and unmanned aerial vehicle(UAV)to form a distributed airborne MIMO radar system can achieve better maneuverability.At the same time,the increasing radar altitude also increase the detection range of low-altitude and ground targets effectively.How to effectively use the diversity advantages of distributed airborne MIMO radar to process echo signals more efficientily,and improve target detection performance and parameter estimation performance is of great significance to the development of MIMO radar.Based on this background,this paper focuses on the signal processing of the cooperative target detection and parameter estimation of the distributed airborne MIMO radar system,and proposes the integrated method of non-coherent AMF fusion detection and estimation and the integration of DP-SA non-coherent fusion detection and estimation method.The method proposed in this paper not only improves the performance of target detection and parameter estimation,but also effectively reduces the amount of calculation for data processing.The main contents of the full work are summarized as follows:1.Established the echo model of the distributed airborne MIMO radar,the clutter suppression and non-stationary clutter compensation methods were introduced,and gave an overall signal processing flow.First of all,this paper eatablished clutter models of the system based on the geometric relationship of clutter observation.On this basis,the clutter characteristics of airborne single and double base radars are studied,and the clutter suppression methods suitable for the system are introduced.Then,in view of the non-uniform and non-stationary clutter problem of airborne MIMO radar,combined with generalized inner product method,Doppler compensation and angular Doppler compensation,the non-uniform and nonstationary clutter compensation was performed.Finally,combining the above analysis and research,a complete signal processing flow is given.2.An integrated method of non-coherent AMF fusion detection and estimation is proposed.In the non-coherent target detection of the distributed airborne MIMO radar systems,the existing methods generally only study the detection performance of the target,which leads to the problem that the system information cannot be fully utilized and the parameter estimation results cannot be given at the same time.This paper proposes an integrated method of non-coherent AMF fusion detection and estimation to solve the above problems.This method first grids the multi-dimensional search area composed of position and velocity,and uses the echo data of the MIMO radar pairwise combination to perform space-time processing to obtain the corresponding output of the target at different nodes,and then obtain the relevant data required by the AMF detector.Then,according to the relevant data that has been obtained,the AMF detector expression is used to calculate the fusion detection statistics of each grid point under a given number of receiving and sending nodes,and the position and speed of each grid point are recorded.Finally,the detection value is compared with the threshold value,and the target detection result and parameter estimation result are obtained at the same time.3.An integrated method of DP-SA non-coherent fusion detection and estimation is proposed.In order to further reduce the amount of calculation,this paper proposes an integrated method of DP-SA non-coherent fusion detection and estimation.Under the premise of ensuring that the detection performance is not reduced as much as possible,this method takes advantage of the distributed MIMO radar system to observe the target from multiple directions,by borrowing the DP algorithm,uses the number of receiving and sending pairs of nodes instead of time as the stage division indicator,classifies each receiving and sending node and the detection statistics are accumulated iteratively.This method avoids the process of traversal search and reduces the amount of calculation effectively.Specifically,the method first uses part of the transceiver pair nodes in the system to obtain the distance Doppler parameter information of the target at the corresponding node.Then,according to the three-point positioning principle,a predicted target point is roughly located by using the obtained range Doppler information,and within a given error range,the information of the predicted target point is used to dynamically iteratively accumulate the detection amount of each receiving and sending node in the space.Finally,the total detection value is compared with the threshold value,and the results of target detection and parameter estimation are obtained at the same time.