Key Technology Research On The Filter Bank Muticarrier Technology

Orthogonal frequency-division multiplexing(OFDM) is the most widely used muticarrier technology around the world,with simple struction and away from interference both in frequency and time domain assuming perfect frequency and time synchronization.Howerer,in the development of the fifth generation wireless communication,two point restrict its application.The first is OFDM has large side slobe,which means it is poor in frequency localization,and if there exists frequency or time offset,the perfomance of OFDM degrades severely.The second is that OFDM needs cyclic prefix(CP) added to each OFDM symbol to mitigate the mutipath impact,which leads to a lower frequency efficiency.Based on this,pepole propose a better muticarrier scheme,filter bank muticarrier(FBMC).FBMC has advantage over OFDM both in frequency focusability and efficiency.,but it needs more complicate computation and algorithm.In this article,we first carry out a thorough study on FBMC system.Then point out the difference between OFDM and FBMC,especially the intrinsic interference in FBMC.Due to the interference,traditional pilot method in OFDM can not be used in FBMC for channel estimation.After studying and simulating the current Pair of Pilot(POP) and Interference Approximation Method(IAM),we propose a novel Auxiliary Pilot(AP) method to mitigate the intrinsic interference to achieve accurate channel information,which turns out to outperform even OFDM in the simulation.Then we propose a simple least mean square(LMS) adaptive equlization method in FBMC,showing that even simple adaptive equlization can boost its perfomance.The intrinsic interference has also impact the combination of FBMC and MIMO. First,we point out that the traditional Maximum likelihood detection is not applicable in MIMO-FBMC,then we propose a novel MMSE-ML method to make use of ML detection method.