Research On Anti-narrowband Interference Techniques For OFDM Systems

Orthogonal Frequency Division Multiplexing(OFDM)has been the major technology in modern broadband wireless communication systems due to its outstanding advantages in spectrum efficiency,inherent ability in performing dynamic power and carrier assignment,robustness against the multi-path impairment and trivial adoption to multi-antenna technology.OFDM systems share spectrum with other narrowband systems is a promising solution to deal with the scarce spectrum resource problem,which is inevitably subject to narrowband interference at the same time.The system performance dropped sharply when the interference intensity exceeds the threshold.How to suppress narrowband interference effectively is an important challenge for OFDM systems.This paper proposes an efficient frequency domain iterative interference cancellation algorithm to solve the narrowband interference problem for OFDM systems.The main work and innovation points of this thesis can be summarized as follows:1.Two typical interferes(single tone and multi-tone interference)are modeled,then the theory expression of bit error rate performance is given.Their influence on OFDM subcarriers and bit error rate performance is analyzed through simulation.Numerical results show that,non-aligned interference will cause spectrum leakage and impair system more seriously than the case of aligned interference.With the increase of interference to signal ratio(ISR),the system performance is declined.2.Four main frequency domain interference detection algorithms in OFDM systems,including Consecutive Mean Excision(CME)algorithm,Forward Consecutive Mean Excision(FCME)algorithm,double threshold algorithm and adaptive multi-threshold algorithm are studied in this paper.The performance of those four algorithms is analyzed in theory and simulation aspects.It can be found that the performance of FCME algorithm is better than CME algorithm.The double threshold detection method is suitable for broadband interference detection.The performance of adaptive multi-threshold algorithm is excellent under low signal to noise ratio.3.To solve the single-tone interference problem in OFDM systems,an effective frequency-domain iterative interference cancellation algorithm is proposed.The basic idea of the proposed algorithm is to make accurate estimation and reconstruction for the interference in frequency domain,and then the interference is subtracted from the raw data.To achieve a more accurate estimation of coarse interference frequency,a new method of spectrum peak search which combined with zero padding interpolation is proposed.The method effectively avoids the misjudgment of spectral peak position caused by spectrum leakage when interference located between the subcarriers.The accurate estimation of interference frequency is implemented by iteratively adopt a combination of frequency conversion,low pass filtering and weighted phase average(WPA)algorithm,the estimation error decreases at each iteration.Simulation results show that,for ISR of-20 dB above,the estimation accuracy of proposed algorithm is close to the Cramer Rao Lower Bound(CRLB).The proposed algorithm reduces the ISR threshold and improves the estimation accuracy significantly when compared with WPA algorithm and Rife algorithm respectively.It can ensure the accurate estimation of interference parameters,reserve the useful signal and remove interference to the utmost,make the system to achieve bit error rate performance without interference.4.To solve the multi-tone interference problem in OFDM systems,an effective frequency-domain interference cancellation algorithm is proposed,which is based on interference reconstruction method.Firstly,the interference detection algorithm is used to detect the single-tone interference with the strongest intensity among multi-tone signal.Then it is removed from the received signal by using the modified single tone estimation and cancellation algorithm.Repeating above process until all interferes are eliminated.Simulation results show that the proposed algorithm is robust to interference frequency interval,and achieve great performance even when interferes are near apart.The proposed algorithm can estimate and recover interference accurately when ISR is 0dB above,which greatly recovery the interferes and improve the system performance.