Estimation of carrier frequency offset for FBMC/OQAM by using modified Kalman filtering

• Published in: International journal of communication systems 2020-11, Vol.33 (16), p.n/a
• Main Authors: Rani, S.N. Nisha, Sridevi, B.
• Format: Article
• Language: English
• Subjects: Bit error rate; Carrier frequencies; carrier frequency offset; Codes; Compensation; Error analysis; filter bank multicarrier; Filter banks; Kalman filters; Mobile communication systems; Orthogonal Frequency Division Multiplexing; Quadrature amplitude modulation; Random noise; signal‐to‐noise ratio; Subcarriers
• ISSN: 1074-5351; 1099-1131
• DOI: 10.1002/dac.4579

Summary Filter bank multicarrier (FBMC) based on offset quadrature amplitude modulation (OQAM) is regarded as the prospective system for mobile communication. To utilize the potential of the system, the issues like carrier frequency offset have to be investigated extensively. The modulations in the multicarrier system draw a great interest among researchers and engineers in the telecommunication field. Despite the OFDM advantages, significant disadvantages in the system enable researches to approach into efficiently evolved FBMC system. In this paper, the detrimental effects caused by carrier frequency offset on the received signal is considered. Carrier Frequency Offset (CFO) compensation has been effectively undertaken by modified Kalman filtering. The compensation technique not only depends on the specific FBMC modification but also influenced by the selected receiver. A comprehensive analysis of the frequency offset response under three conditions (without offset, with offset, and the compensated samples) has been performed in the proposed system with performance measures as bit error rate (BER) and signal‐to‐noise ratio. 9 CFO compensation is analyzed by varying the channel characteristics (additive white Gaussian noise [AWGN] and Rayleigh Fading) and subcarrier filtering (prototype filter, Gaussian filter, and root‐raised coefficients). From the overall analysis, it is observed that the BER varies from 0.1 to 0.001. The graphical describes the overall process of the proposed framework. The data bits are initially mapped into QAM symbols and then modulated with OQAM preprocessing. After preprocessing, diversion from serial to parallel change have been accomplished. Then, the optimization of impulse response coefficients is done by prototype filtering. The resulting frequency domain of the signal is transferred to time domain signal followed by the addition of the carrier offset to the transmitter bits. Again, diversion of parallel to serial is done and then the signals are passed to noise channel AWGN or Rayleigh fading by varying their channel characteristics. The proposed system utilizes the benefits of FBMC/OQAM with modified Kalman filtering for the purpose of compensating carrier frequency offset and thereby reducing the bit error rate.