RF impairments compensation schemes for direct conversion OFDM systems

Abstract

Orthogonal frequency division multiplexing (OFDM) has been widely recommended to the most promising baseband modulation and multiplexing technology for 3G and 4G communication systems because of its spectral efficiency and robustness against frequency selective fading channel. Furthermore, direct conversion based RF transceiver has had more advantages in aspects of power consumption, area and complexity than conventional heterodyne transceiver. In order to maintain all merits in both direct conversion structure and OFDM, many algorithms have been studied to compensate inevitable RF impairments such as I/Q imbalance and carrier frequency offset (CFO) in digital domain. In this respect, we firstly deal with a new baseband signal processing technique to compensate I/Q imbalance for OFDM system by jointly developing preamble and preprocessor. Based on weighted average concept, the main function of preprocessor is to separate desired signal and interference signal from received preamble even though all subcarriers carry non-zero values. Using the separated signals, we can efficiently estimate channel and imbalance parameter under least square (LS) criterion. Numerical comparisons show the superiority of proposed scheme compared with previous method. Secondly, the synchronization is also the main problem before de-modulating the received signal. Especially, CFO brings severe interferences from the other subcarriers in OFDM signal. On that account, we address a joint estimation method for CFO and I/Q imbalance using constrained preambles. Because existing methods can not guarantee correct solutions within all range of CFO, we derive two constraints about phase rotations which should be differently adjusted to each repeated pattern. Using these constraints, proposed estimators can find the optimal estimate of CFO and low complexity I/Q imbalance estimation are also possible in all range of CFO with no exceptions. Numerical results show the importance of these const...