Research On Key Technology Of Intelligent Sensing Of Wireless Signal Emitter

In modern warfare,weapons and equipment are increasingly relying on the radio spectrum,and the competition between the two sides for spectrum resources is becoming more and more fiercely.As an emerging technology that can improve the management and control of spectrum resources,cognitive electronic warfare technology has become an important research content to improve the military capability of each country.In the modern battlefield environment,wireless signal emitter has the characteristics of various types of frequency-using equipment,wide spectrum range,complicated signal types and different signal strength,especially in the condition of strong confrontation,which puts forward higher requirements on the cognitive ability of the cognitive electronic warfare.Emitter intelligent sensing system,one of the key subsystem of cognitive electronic warfare system,can realize intelligent sensing of wireless signal emitter situation information in battlefield condition.Faced with the requirements of multi-platform operations,Emitter intelligent sensing system needs to realize the dynamic reception of ultra-wideband wireless signals,the spectrum situational awareness of wideband wireless signals and the classification and identification of emitter under condition of low SNR(Signal to Noise Ratio).In this paper,the key technologies of wireless signal radiation source intelligent perception system are studied.The main tasks and innovations are as follows.(1)Design and modeling of emitter intelligent sensing system.At present,emitter sensing system is mostly built on a single platform.When traditional emitter sensing system is used to sensing specific emitter,task coordination ability,platform management ability and sensing efficiency of multi-platform are relatively low.In order to solve the wireless signal emitter intelligent sensing problem of complicated electromagnetic environment,this paper proposes a multi-platform oriented centralized-distributed emitter intelligent sensing system,designs the system solution and load scheme and analyzes the mathematical model of signal in the load.This study provides a theoretical basis for the design and implementation of emitter intelligent sensing system.(2)Design the reception strategy of full-band wireless signal.Although traditional digital channelized receiver can achieve channelized reception of specific frequency band and specific bandwidth,it is difficult to adapt to the reception of ultra-wideband wireless signals.In order to solve the full-band wireless reception problems of emitter intelligent sensing system,this paper adopts a reception scheme which divided the full-band signal into multiple sub-band,we design a super-heterodyne receiver based on multiple sets of adjustable antenna,and study the sub-band division and antenna status transfer problems of receiver.On this basis,the sub-band selection is studied.On the basis of modeling analysis,optimal discount upper confidence bound(OD-UCB)strategy is proposed,and the optimal parameter design method of the strategy is studied.The wireless signal reception strategy proposed in this paper realizes the high efficiency and dynamic receiving of the full-band wireless signal.(3)Wideband wireless signal cooperative spectrum sensing method.In the field of cognitive radio,many spectrum sensing algorithms have been proposed,but traditional spectrum sensing algorithms are difficult to solve the sensing problem of complex electromagnetic environment.Aiming at the spectrum sensing problem of wideband wireless signals in the emitter intelligent sensing system,this paper proposes a cluster-based cooperative spectrum sensing architecture which decompose the wideband spectrum sensing problem into the local wideband spectrum sensing problem with cluster headers collaborative spectrum sensing problem,and a local spectrum sensing algorithm based on multi-resolution singular spectrum entropy and a cooperative spectrum sensing algorithm based on k/N fusion rules are proposed.In this paper,the cooperative spectrum sensing of wideband wireless signal is realized under the condition of low SNR.(4)Emitter classification and identification algorithm under small sample condition.At present,classification and identification algorithms are generally developed from the perspective of pattern recognition,at the same time,classification and identification algorithms based on machine learning are proposed.It is difficult for emitter intelligent sensing system to meet the requirements of the two algorithms on sample size.In order to solve the problem of emitter classification and identification under the condition of small sample,a random forest classification and identification algorithm based on multi-domain characteristics of wireless signals is proposed.In this study,multi-domain characteristics are used to make up for the adverse effects of small sample size,and emitter classification and identification in low SNR situation with small sample size is realized.