Er Erik Fledderus

On the influence of phase noise induced ICI in MIMO OFDM systems

The influence of transmitter and receiver phase noise (PN) on the performance of a multiple-input multiple-output orthogonal frequency division multiplexing (MIMO OFDM) based communication system is analyzed. It is shown that in the case of frequency flat Rayleigh fading, the influence of receiver and transmitter PN is equal. In the case of independent Rayleigh fading, however, the impact of the receiver PN is shown to depend on the ratio between the number of transmit and receive branches.

Estimation and Compensation of Frequency Selective TX/RX IQ Imbalance in MIMO OFDM systems

The influence, digital estimation and compensation of IQ mismatch at both transmitter (TX) and receiver (RX) side of a direct-conversion based multiple antenna OFDM system are presented in this paper. The resulting IQ imbalance can be split into a frequency independent and a frequency selective part. For estimation and compensation a two-step approach is presented, combining a data-aided and decision directed approach. First the frequency independent IQ imbalance is estimated using a preamble and, subsequently, the frequency selective IQ imbalance is removed using adaptive filtering. A numerical performance analysis reveals that the presented approach effectively mitigates the influence of IQ mismatch in OFDM-based space division multiplexing systems experiencing both TX and RX IQ imbalance.

Estimation and compensation of TX and RX IQ imbalance in OFDM-based MIMO systems

This paper studies the influence, estimation and digital compensation of IQ imbalance at both transmitter and receiver side of a multiple-input multiple-output (MIMO) OFDM system. Hereto a preamble is designed, which enables simultaneous estimation of the channel and imbalance parameters. New estimation approaches for TX, RX and joint TX and RX IQ imbalance in MIMO OFDM systems are proposed and evaluated. Results from a numerical study show that all three proposed compensation approaches are able to significantly suppress the influence of IQ mismatch.

Influence and suppression of phase noise in multi-antenna OFDM

The influence of phase noise (PN) on the performance of a multiple-input multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) based communication system is analyzed. As part of this analysis, an estimator for the common phase error (CPE) is derived, based on the maximum likelihood estimation (MLE), which was chosen for its asymptotic optimality. Since the computational complexity of the MLE procedure clearly rules out a cost effective solution, a reduced-complexity algorithm. based on least squares estimation (LSE) is designed which yields suboptimal performance. Simulation results reveal that the difference in bit-error-rate performance between systems applying the LSE or MLE of the CPE is very small. This implies that the LSE is well applicable for 802.11a based MIMO systems.

Performance analysis of zero-IF MIMO OFDM transceivers with IQ imbalance

This contribution analytically studies the influence of IQ mismatch on the performance of multiple-antenna orthogonal frequency division multiplexing (OFDM) systems based on the zero-IF architecture. First a system model for such a multiple-input multiple-output (MIMO) OFDM system experiencing both transmitter (TX) and receiver (RX) IQ imbalance is derived. This is used to obtain analytical expressions for the probability of error in symbol detection for MIMO OFDM systems in Rayleigh faded multipath channels. The derived results can be used to find matching specifications for the TX- and RX-branches for different multicarrier MIMO systems. It is concluded that in fading channels RX IQ imbalance is on average more destructive than TX IQ imbalance. Additionally, it is concluded that the addition of extra RX antennas is beneficial to reduce the dependence on RX IQ imbalance, but increases at the same time the impact of TX IQ imbalance.

Distribution of the ICI Term in Phase Noise Impaired OFDM Systems

Orthogonality between the subcarriers of an orthogonal frequency division multiplexing (OFDM) system is affected by phase noise, which causes inter-carrier interference (ICI). The distribution of this interference term is studied in this paper. The distribution of the ICI for large number of carriers is derived and it is shown that the complex Gaussian approximation, generally applied in previous literature, is not valid and that the ICI term exhibits thicker tails. An analysis of the tail probabilities confirms these finding and shows that bit-error probabilities are severely underestimated when the Gaussian approximation for the ICI term is used, leading to too optimistic design criteria. Results from a simulation study confirm the analytical findings and show the validity of the limit distribution, obtained under the assumption of a large number of subcarriers, already for a modest number of subcarriers.

The Application of Spatial Shifting for Peak-to-Average Power Ratio Reduction in MIMO OFDM Systems

This paper investigates the application of spatial shifting (SS) of partial transmit sequences (PTSs) in multiple-input multiple-output OFDM. The technique rearranges the transmit (TX) vector in such a way that subparts are transmitted on those TX branches that result in the lowest overall peak-to-average power ratio. The application of different subcarrier grouping schemes and the combination of SS with phase shifting (PS) of the PTSs is investigated. Furthermore, a transparent extension of the techniques, without extra signalling overhead, is proposed. Numerical results prove the effectiveness of the SS and the combined SS/PS approach

Performance impact of IQ mismatch in direct-conversion MIMO OFDM transceivers

This contribution analytically studies the influence of transmitter (TX) as well as receiver (RX) IQ mismatch on the performance of multiple-antenna OFDM systems based on direct-conversion. Analytical expressions are derived for the probability of error for MIMO OFDM systems in both nonfaded and Rayleigh faded channels. The results can be used to derive matching specifications for the TX- and RX-branches. It is concluded that in fading channels RX IQ imbalance is on average more destructive than TX IQ imbalance. Additionally, it is concluded that the addition of extra RX antennas is beneficial to reduce RX IQ imbalance dependence, but increases the TX IQ imbalance impact

Impact of nonlinearities in multiple-antenna OFDM transceivers

This paper investigates the performance impact of nonlinearities in the transmitter (TX) as well as the receiver (RX) of multiple-antenna OFDM systems. First different, previously proposed, nonlinearity models are reviewed and their equivalence is shown. Subsequently, analytical expressions are derived for the probability of error of multiple-input multiple-output (MIMO) OFDM systems in Rayleigh-faded channels, which show good agreement with results from a simulation study. It can be concluded from the results that for high SNR values the TX nonlinearities cause performance floors, which are independent of the MIMO configuration. For RX nonlinearities, however, the performance in this high SNR region does depend on the MIMO configuration. The results can be used to derive limits for the permissible nonlinearities in MIMO OFDM systems

RF impairments in high-rate wireless systems - understanding the impact of TX/RX-asymmetry

Dealing with radio frequency (RF) front-end impairments will be one of the major design challenges for next- generation wireless communication systems due to conflicting requirements, such as high data rate, low cost and low power consumption. The use of digital compensation of the imperfections appears a very promising method to meet specifications. Pursuing that path, however, requires thorough understanding of the influence of the RF front-end non-idealities on the received signal and the resulting system performance. To this end, this paper reviews the impact of three important impairments, namely, phase noise, IQ imbalance and nonlinearities, on the performance of next-generation high-rate wireless systems. A specific focus is on the difference between transmitter (TX) and receiver (RX) incurred imperfections. Moreover, a generalized error model to capture to aggregate influence of different impairments is presented.