Kimmo Kansanen

Multilevel-coded QAM with MIMO turbo-equalization in broadband single-carrier signaling

A new scheme employing multilevel coded bit-interleaved transmission allowing for efficient turbo-equalization is proposed for multiple-input-multiple-output (MIMO) broadband single-carrier signaling. The proposed scheme is based on block-partitioned hierarchical constellations and offers robustness in data throughput efficiency against varying spatio-temporal characteristics of fading channels. The proposed scheme is compared to a standard bit-interleaved-coded-modulation (BICM) system in frequency-selective channels, and found to offer better overall throughput. It is also shown that level-wise retransmission control with the proposed multilevel single-carrier signaling scheme can offer further throughput improvement. Channel measurement-based simulations are used to evaluate performances of both schemes in real fields. Channel parameter analysis using a superresolution technique is performed to clarify the underlying reasons for the performance characteristics of the systems.

Outage-based LDPC code design for SC/MMSE turbo-equalization

A semi-analytical EXIT chart analysis approach is presented to perform code design of low-density parity-check (LDPC) codes with a frequency-domain SC/MMSE turbo equalizer. The approach utilizes the equivalent channel assumption at the equalizer output to analytically compute the equalizer output mutual information. The analysis is used to compute the EXIT functions of the receiver for a sample set of random channel realizations. LDPC code design is then performed to match the channel code to the random convergence properties of the equalizer to obtain a desired outage probability. Substantial performance improvement is achieved with the code design compared to a standard regular LDPC code. Target block error probability is not reached, which is considered to be the result of some of the approximations made in the analysis and code design.

A computationally efficient MIMO turbo-equaliser

A reduced complexity version of the soft-interference cancelling MMSE turbo equalizer is proposed for single-carrier MIMO systems. In the receiver exact symbol-by-symbol interference covariance matrice inverses are replaced by time averages acquired through matrix inversion lemma iterations. Consequently, also the requirement to explicitly estimate signal-to-noise ratio is removed. After a number of iterations the equalizer is further simplified by utilizing the matched filter approximation. Remaining multipath and multiuser interference, in addition to receiver noise, are accounted for in the approximation. The performance of the receiver is verified through simulations with two MIMO radio network configurations.

Frequency-domain MMSE turbo equalization of multilevel coded QAM-convergence in real fields

Convergence properties of frequency domain turbo equalization with multilevel bit-interleaved coded modulation (MLBICM) and bit-interleaved coded modulation (BICM) are studied in measured quasi-static fading channels by the means of semi-analytical convergence analysis. Spatial channel parameters are matched to the reached mutual information on a per-measurement basis and trends in the behavior of the equalizer are obtained, when the channel characteristics, in terms of transmit and receive side azimuth spreads, change. Robust convergence of frequency-domain turbo equalization combined with multilevel BICM is demonstrated through the convergence analysis. Three channel codes are evaluated in the measurement-based simulation showing the effects of outer code selection

Turbo equalization of multilevel coded QAM

A turbo-equalization technique is presented for transmissions, where QAM constellations are constructed via block-partitioning as linear combinations of rotated, separately channel-coded BPSK components. The multilevel encoded QAM can be decoded with multistage decoding level-by-level. Iterative equalization and multistage decoding is performed via soft interference cancellation and MMSE equalization followed by soft-in-soft-out channel decoders. Due to binary channel codes both the equalizer and decoder can operate with binary likelihood values without requiring symbol/bit conversion. The resulting scheme is suitable for robust transmission when the transmitter has little or no knowledge of channel quality.

Frequency Domain Joint-over-Antenna MIMO Turbo Equalization

This paper proposes a novel iterative frequency domain multiple input multiple output (MIMO) signal detection technique for the reception of overloaded spatially multiplexed MIMO transmission for single carrier signalling in frequency-selective channels. In the presence of spatial correlation as well as the case where there are more transmit antennae than receive antennae (referred to as overloaded transmission in this paper) the performance of antenna-by-antenna (AA)-based MIMO minimum mean square error (MMSE) receivers are, in general, degraded. Therefore, we propose joint-over-antenna (JA) detection-based iterative frequency domain technique to improve the performance of receiver in the presence of overloaded antenna transmission and spatial correlation. The proposed receiver is based on the combination of maximum a posteriori probability (MAP) algorithm, and soft-cancellation (SC) and MMSE filtering for turbo-coded single carrier point-to-point MIMO systems. In the proposed receiver MIMO signal detection is comprised of two stages. At the first stage, inter-symbol interference (ISI) is suppressed with MMSE filtering. At the second stage, co-antenna interference (CAI) is suppressed by the MAP algorithm and transmitter antennae are decomposed each other. Simulation results show that performance gain of JA compared to AA technique is significant in the presence of spatial correlation as well as in overloaded cases. The performance comparison is made in terms of frame error rate (FER) with different antenna configurations and different spatial correlations in point-to-point MIMO frequency-selective fading channels

Core matrix inversion techniques for SC/MMSE MIMO turbo equalization

Conditional minimum mean square error (CMMSE) estimation problems commonly appear in iterative (turbo) signal detectors that utilize soft feedback from soft-input soft-output (SISO) channel decoders. To solve CMMSE problems, covariance matrix inverse has to be calculated for the output vector of soft-cancellation of interfering components, which requires a cubic order of computational complexity of the row size of the equivalent space-time channel matrix. This paper first proposes a computationally efficient method to calculate strictly the matrix inversion for CMMSE. It is shown that the proposed technique can be implemented in a highly pipelined manner. This paper also presents an approximated version of the algorithm that further reduces the computational complexity to a square order of the channel matrixs row size without causing any significant loss in performance. Results of model-based and field measurement data-based simulations for the approximated version of the algorithm are presented.

Groupwise interference cancellation receivers in cellular WCDMA networks

The inclusion of groupwise interference cancellation techniques into WCDMA uplink receivers is studied in this paper. Emerging third generation WCDMA signal structures for multirate transmission are utilized along with necessary uplink techniques, namely antenna diversity reception, transmitter power control and signal-to-interference ratio (SIR) estimation. A groupwise serial interference cancellation (GSIC) technique suitable for WCDMA systems using long spreading sequences is presented along with the necessary estimation techniques. A simple intercell-interference (ICI) model is incorporated and the behavior of the receiver in a power controlled environment is studied.

Performance of parallel interference cancellation for CDMA with delay estimation

While multiuser receivers for code-division multiple-access (CDMA) systems have been extensively developed in the past, there performance in fading channels with practical delay and channel estimation has gained very little attention. The performance of the hard-decision parallel interference cancellation (HD-PIC) multiuser receiver in frequency selective fading channels is studied. A practical complex channel gain and delay estimation are applied. The results show that the performance of the PIC receiver remains superior even in the presence of a practical delay estimator.

Residual interference suppression in parallel interference cancellation receivers

Parallel interference cancellation (PIC) receivers are among the most promising receiver techniques for future code-division multiple-access (CDMA) systems. Interference cancellation efficiency relies on the knowledge of the number of users and propagation paths needed in multiple-access interference (MAI) estimation. In practice, the exact number of users and paths is not known exactly, e.g., due to inter-cell interference, unknown propagation paths or new users trying to connect to the base station. As a result, there will be some residual interference even with perfect cancellation of the known signal components. In many cases, the inter-cell interference can be large enough to significantly degrade the performance of the PIC receivers. For that reason, residual interference suppression is crucial to guarantee that the PIC receivers can operate reliably. Residual interference can be reduced by applying adaptive antennas to PIC receivers. In this paper, another possibility is considered. The approach taken is to combine PIC receivers with blind adaptive interference suppression techniques in single sensor receivers.