Research And System Construction Of Deep Cognitive Technology In 5G NR

After decades of development and evolution,public mobile communication technology has entered the fifth generation(5G)stage.Currently,China has built the world’s largest 5G network,providing reliable connectivity services for China’s industrial manufacturing,intelligent transportation,health care,and other fields.To enhance the quality of 5G services and drive the continuous and healthy development of the 5G commercial industry,regulatory and security agencies must approach the subject from a third-party standpoint.They should independently and objectively assess 5G signals,oversee user traffic conditions,and ensure frequency security.By doing so,they can promptly detect any problems or issues within 5G services.However,5G introduces signal parameter set,beam management and new coding methods in the air interface,which brings new challenges to the third-party air interface monitoring.The traditional frequency sweep monitoring equipment cannot identify the signal type and analyze the internal data transmission of the modulated signal;Although the 5G signal test instrument can perform protocol analysis and public channel decoding on the 5G signal,it cannot identify the users in the signal,let alone perceive the user’s frequency use behavior,and it is difficult to meet the needs of the regulatory and security departments.For this reason,this thesis studies 5G electromagnetic environment cognitive technology under non-cooperation conditions,and the main contributions include the following two parts:(1)In terms of user load assessment,a method for estimating the number of users in 5G cells is proposed.This method divides the user number estimation of the whole cell into the user number estimation of each beam in the space.Based on the physical downlink control channel(PDCCH)decoding algorithm under non-cooperative conditions,all downlink control information(DCI)in the beam is obtained for the beam signal at the spatial location of the monitoring equipment.The Cell-Radio Network Temporary Identity(C-RNTI)obtained in the decoding process is used as the user identification index to count the number of users in a single beam to obtain the cell user bearer.Finally,the feasibility of the user number estimation method in the beam is verified on the simulation platform,and the actual system is built to estimate the number of users in the commercial 5G signal beam.(2)In terms of frequency security monitoring,a 5G user frequency behavior awareness method is proposed.Based on(1),this method screens 5G users within the receiving range of the monitoring equipment and continuously tracks them.Based on the DCI analysis results and the location of the shared channel demodulation reference signal(DMRS)mapping,calculate the number of resource elements(RE)carrying service data in the target user’s shared channel data block,and calculate the transport block size(TBS)in combination with its modulation and coding scheme(MCS),Estimate the user’s frequency use duration and the physical layer’s uplink and downlink throughput,and then infer the user’s frequency use intention and perceive the user’s frequency use behavior.For users with abnormal frequency use behavior,decode the transmission bits of the physical uplink shared channel(PUSCH)of the user.Finally,the feasibility of the user frequency behavior perception method is verified on the simulation platform,and the actual system is built to perceive the commercial 5G user frequency behavior,in which the estimation error of the physical layer throughput rate is guaranteed to be within 7%.