A modified FFT and biquadratic interpolation based algorithm for carrier frequency offset estimation in MC-CDMA based ad hoc networks

The future wireless communication systems demand very high data rates, anti-jamming ability and multiuser support. People want large amount of data to be continuously accessible in their personal devices. Direct Sequence (DS) spread spectrum based techniques such as Code Division Multiple Access (CDMA) fulfil these requirements but, at the same time, suffer from the Intersymbol Interference (ISI). Multicarrier CDMA (MC-CDMA) is an emerging technology to be used in mobile devices operating in an ad hoc setting due to its immunity towards ISI and having all the advantages of spread spectrum communication. One of the major problems with MC-CDMA is the high sensitivity towards carrier frequency offsets caused due to the inherent inaccuracy of crystal oscillators. This carrier frequency offset destroys the orthogonality of the subcarriers resulting in Intercarrier Interference (ICI). In this paper, we propose a computationally efficient algorithm based on Fast Fourier Transform (FFT) and biquadratic Lagrange interpolation. The FFT is based on the use of overlapping windows for each frame of the data instead of non-overlapping windows. This gives a coarse estimate of the frequency offset which is refined by the successive application of Lagrange quadratic interpolation to the samples in the vicinity of FFT peak. The proposed algorithm has been applied to the multiuser ad hoc network and simulated in Stanford University Interim (SUI) channels. It has been shown by simulations that the proposed algorithm provides better performance of almost 4~5 dB as compared to the conventional FFT based algorithms.