Optimal Design Of FTN Wireless Communication System

In the 5G information age that has arrived,the quality of people's communication continues to improve,and the wide application of the new generation of information and communication technologies will produce huge data flow.Therefore,more spectrum resources are needed to increase the capacity of the system.Now people's attention has been focused on the millimeter wave band,and in the future,they will even prepare to communicate in the terahertz band to seek wider frequency resources.However,there are many difficulties in high-band communication,so it is particularly important to improve the transmission efficiency as much as possible in the existing spectrum resources.In the1960 s,researchers proposed the Faster-Than-Nyquist(FTN),which improved the spectral efficiency by destroying the orthogonality between the symbol waveforms.However,at that time,the detection algorithm of FTN system was very complex,and the digital signal processing and chip technology were not mature,so it was not widely used.But in recent years,with the proposal of many new signal detection algorithms,FTN has begun to receive attention again.This thesis first introduces the traditional Nyquist system,and leads to the principle and model of FTN system,then deduces the capacity and frequency band utilization of FTN system,and points out that FTN system can obtain higher spectral efficiency than traditional Nyquist system.Through the direct demodulation of the received signal of FTN system,it is pointed out that due to the existence of serious intersymbol interference,its bit error rate performance is very poor,so a corresponding detection algorithm is needed to improve the system performance.This thesis mainly studies the optimal design scheme of FTN system.In recent years,detection algorithm based on frequency domain equalization have attracted attention due to its low complexity,but its performance is not ideal.In this thesis,we consider the detection algorithm to eliminate the intersymbol interference matrix of the FTN system,such as the scheme based on SVD decomposition,but its performance is still unsatisfactory.Therefore,this thesis proposes a scheme to further improve the system performance by using the alternating direction method of multipliers after multiplying the inverse matrix of the intersymbol interference matrix at the receiver,by converting the signal detection problem of the FTN system into a maximum likelihood sequence estimation problem,and inserting pilots as a reference for iterative convergence,the ADMM-based signal detection scheme of the FTN system is completed.The simulation results show that this scheme has excellent performance when the compression factor is high,and when the system uses ultra-high-order modulation,it can still achieve good performance with a small increase in complexity.In this thesis,the forming process of the FTN system is equivalent to the superposition of waveforms,and it is found that when the compression factor is low,the total power at the transmitter is large.Therefore,this thesis proposes a system optimization scheme based on scrambling.By adding a scrambling code at the transmitter,the power is reduced as much as possible and the performance of the system is improved.The simulation results show that by adding a small amount of extra bits as a scrambling indication,the system can achieve good performance gains.In the FTN system with low compression factor,the signal detection algorithm based on trellis has excellent performance,but the complexity in hardware implementation is high.This thesis proposes a scheme that combines probability calculation in the maximum a posteriori probability detection algorithm,and optimizes the implementation steps in a targeted manner.It can adjust the length of the random number sequence to change the complexity of the detection algorithm of the system.The simulation results show that this scheme can significantly reduce resource consumption,and the performance degradation is small.