Research And VSA Implementation Of Frequency Offset And Channel Estimation Algorithm In 5G NR

As a new generation of mobile communication system,New Radio（NR）of 5th Generation Mobile Communication（5G）has been developing vigorously in recent years,and has been officially commercialized and become a research hotspot.On the one hand,the Orthogonal Frequency Division Multiplexing（OFDM）technology used in 5G downlink is very sensitive to frequency offset,which can easily cause distortion of the received signal;On the other hand,signals will fade after being transmitted through wireless channels,which leads to a serious decline in signal analysis ability.Therefore,frequency offset estimation and channel estimation module play a vital role in improving signal analysis ability.For communication test instruments,the frequency offset estimation range is required to be larger than that of communication equipment.Traditional frequency offset estimation algorithms have their limitations.For example,the frequency offset estimation algorithm based on Cyclic Prefix（CP）has poor estimation performance,while the frequency offset estimation algorithm based on Demodulation Reference Signal（DMRS）has high accuracy,mean square error less than 10^-5（SNR 30dB）,but the estimation range is small.At the same time,in order to make the communication test instrument for different channel environment have a higher signal analytical performance,need to adopt better performance of channel estimation algorithm,and the traditional Linear Minimum Mean Square Error（LMMSE）channel estimation algorithm has good performance but complexity is too high,which seriously affects the analytic speed of the system in the application process.In view of the problems existing in the above algorithms,this paper studies the frequency offset and channel estimation algorithms based on the characteristics of the 5G NR.The main work and innovation points are as follows:1.A practical DMRS-CP joint frequency offset estimation algorithm is proposed.The algorithm takes advantage of the high accuracy of frequency offset estimation based on DMRS and the large range of frequency offset estimation based on CP.When the subcarrier interval is30k Hz,the estimation range is increased from(-1.557k Hz,1.557k Hz]to(-15k Hz,15k Hz],and the estimation range of the algorithm is increased while still having a high estimation accuracy.2.An improved LMMSE channel estimation algorithm based on sliding window is proposed.Aiming at the high complexity of the LMMSE algorithm caused by the operation of the channel correlation matrix,the channel autocorrelation matrix is constructed by sliding window method,and performs order reduction processing on the original matrix.The computational complexity shows that compared with the traditional LMMSE algorithm,the algorithm greatly reduces the number of complex multiplication and addition,the complexity of the algorithm is reduced by 99.74%,and has higher engineering practicability.3.The DMRS-CP frequency offset estimation algorithm and the channel estimation algorithm based on sliding window proposed in this paper are verified and applied in the Vector Signal Analysis（VSA）software platform.The experimental results show that the estimation accuracy of frequency offset reaches±0.1Hz at-100Hz and-14.99k Hz,the Error Vector Magnitude（EVM）better than 1%（center frequency 1GHz,bandwidth 100MHz,SNR 40d B,256QAM）,and it improves the analytical performance of the software.