IQ imbalance compensation: A semi-blind method for OFDM systems in fast fading channels

Abstract

In this paper, an orthogonal frequency division multiplexing (OFDM) system operating in a fast fading environment modeled by a doubly selective channel (DSC) is considered. The paper first reformulates a commonly adopted system model using the generalized complex exponential basis expansion technique. The resulting model enables the IQ imbalance and DSC to be estimated in the time domain with a small number of scattered pilots within an OFDM symbol. A joint estimation and compensation scheme is then proposed which compensates all the inter-carrier interference terms. Simulation results show that the proposed compensation method achieves better symbol error rate performance than previous proposed methods.

Low complexity estimation of fast fading radio channels for higher order modulation

“…For lower fading rates of up to 0.1% of the sample rate, low complexity and near optimal techniques are available for orthogonal frequency division multiplexing (OFDM) and frequency-domain equalization schemes [5]- [7]. However, these methods rely on the fundamental assumption that the channel is nearly static over the duration of long blocks of symbols, so they fail to work at the fading rate addressed in this paper.…”

Pilot-Aided IQ Imbalance Compensation for OFDM Systems Operating Over Doubly Selective Channels

“…Since is a Hermitian matrix, based on eigen-decomposition, we have (33) where is the th eigenvalue of with being the corresponding eigenvector. On the other hand, comparing (23) and (30), we have where the vector in square brackets of and is not confined to but is applicable to any vector having compatible dimensions with and . Due to this fact and the eigen-decomposition on in (37) can be rewritten as (38) Dropping all the terms irrelevant to and grouping those terms with common factors of together, (38) is further rewritten as (39), shown at the bottom of the page.…”

“…However, the more important reason that the proposed LS scheme has better performance is due to more accurate IQ imbalance and channel parameter estimation. In fact, further to this work, a low complexity scheme has been proposed in [23] by considering only several neighboring subcarriers. The scheme has a complexity of which is lower than that in [15] but performs better.…”

“…In order to compensate for the IQ imbalance and channel so that the unknown data can be recovered, we substitute and into (17), and the expression becomes (23), shown at the bottom of the page, where the unknown data are in the first two terms only.…”

“…In this section, based on the EM algorithm [16], an iterative algorithm is developed to further improve the compensation performance. Recalling (23), if the noise is treated as being zero-mean white Gaussian with variance , the unknown variables are and . Using the EM terminology, we take as the complete data with being the underlying variable, and as parameters of interest.…”