Blind estimation of channels for OFDM-based amplify-and-forward two-way relay network

• Main Authors: Wei-Hsuan Hsieh, 謝維軒
• Other Authors: See-May Phoong
• Format: Others
• Language: en_US
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/gct4mg

碩士 === 國立臺灣大學 === 電信工程學研究所 === 106 === Orthogonal frequency division multiplexing (OFDM) systems has recently received considerable interest for its advantages in high-bit-rate transmissions over frequency selective fading channels. In the last decade, wireless relay network becomes an active research due to its reliability, effective spectral and power efficiency. In particular, two-way relay network (TWRN) has drawn a lot of attention because its overall communication rate is approximately twice of that achieved in the conventional one-way relay network (OWRN). This thesis studies the blind channels estimation for amplify-and-forward (AF) TWRN in cyclic-prefixed OFDM (CP-OFDM) and in zero-padded OFDM (ZP-OFDM) systems. For CP-OFDM systems, we propose three methods to estimate the first cascaded channel resulting in self-interference. These methods are circular repetition method, incorporation of cyclic-prefixed into estimation method and subspace method. Then, we investigate the estimation accuracy of the first cascaded channel when there exist virtual carriers (VC) for different methods. Also we propose a least-squares carrier frequency offset (CFO) estimation method by using VC. After that, the second cascaded channel which is from the source to destination is estimated by nulling method. The algorithm for the second cascaded channel can use repetition method to handle the case of few received blocks. In addition, we also extend our channel estimation methods to the case of ZP-OFDM systems. Simulation results are carried out to verify the performance of the proposed methods.