Research On Channel Estimation Of Filter Bank Multicarrier System

Cognitive Radio (CR) is one of the most popular research area of wireless communication. The physical layer of current CR network is Orthogonal Frequency Division Multiplexing (OFDM). OFDM has advantages like high transmission rate and overcoming the inter-symbol interference (ISI). But the out of band interference is large in OFDM, leading limitation of apply in Dynamic Spectrum Access (DSA). Meanwhile, since OFDM needs protect band to separate subchannels, the spectral efficiency is also reduced. Based on this, some European universities and research organizations develop PHYDYAS project. PHYDYAS proposed Filter Band MultiCarrier (FBMC) technology as the physical layer of Cognitive Radio. FBMC can satisfy the need of DSA and spectrum sharing well. In this paper, we study the channel estimation technology in FBMC.In this paper, we first introduce the basic principle of FBMC, including the design of prototype filter, polyphase network technology, OQAM modulation method and basic structure of FBMC. We analyze the advantages of FBMC:lower interference and higher data rate.For the problem of pilot-based channel estimation algorithm, the paper propose two pilot design methods-auxiliary pilot (AP) method and MIAM method, AP method requires additional computations and MIAM method loss some spectral efficiency. Simulation results show that the performance of the AP superior MIAM, while MIAM is more flexible than AP in practical application. And we also present a proposed pilot design method according to tile-based sub-channel mode. The simulation results show that the combination of both AP and MIAM methods can achieve ideal performance.The paper also gives detailed analysis of computational complexity and spectral efficiency of FBMC. In addition to the comparison of the computational complexity of OFDM and FBMC, we also give analysis of the computational complexity and spectrum efficiency of different pilot design algorithm. Finally, we discuss some problems need to be solved in application of FBMC in wireless communication and future research directions.