| Orthogonal frequency division multiplexing (OFDM) is the primary choice of modulation technique for several recent broadband communication standards such as 3GPP-LTE,WiMAX, etc. In theory, OFDM enjoys low implementation complexity and high performance which are the key properties for its ubiquitousness. However, there are several practical hardware implementation issues which are potentially limiting the performance of OFDM. In reality, radio frequency (RF) front-end analog circuit components are manufactured with certain linearity and accuracy limit, which can cause distortion to the received signal during the up-or down-conversion process. This distortion can be characterized in terms of inter-carrier interference (ICI) in OFDM systems. These imperfections among which are phase noise (PHN) and in-phase quadrature-phase (I/Q) imbalance could be performance bottleneck especially at high transmission rates i.e. as the signal constellation size increases. Channel estimation and data detection in the presence of PHN and I/Q imbalance become quit challenging for the OFDM systems due to high interference levels from the adjacent and/or image subcarriers. In this proposal, we present a joint maximum-likelihood (ML) channel and PHN estimator during full-pilot OFDM symbols. Since only scattered pilots are available during data OFDM symbols, a PHN tracking method using linear interpolation is proposed which is shown to achieve significant bit error rate (BER) improvement compared to the existing PHN mitigation techniques. Moreover, the effect of I/Q imbalance on the OFDM signal is studied and the joint ML estimator for channel and PHN is generalized to consider I/Q imbalance as well. The proposed algorithms is generalized to multiple input multiple output (MIMO) OFDM systems. Finally, the effect of PHN on the performance of relay systems is studied.;Keywords : Channel estimation, Phase noise, I/Q imbalance, ICI, OFDM, Linear interpolation, Relay. |
