Hyounkuk Kim

An Efficient QRD-M Algorithm Using Partial Decision Feedback Detection

This paper presents a reduced-complexity QRD- M algorithm using partial decision feedback (DF) detection. Adaptive QRD-M algorithm obtains the threshold value through the simulation to prune off paths at each stage. However, the proposed algorithm efficiently acquires the threshold value using partial DF detection at each stage. In 4 times 4 V-BLAST system with 16-QAM, the complexity of the proposed algorithm is approximately 50 % lower than that of adaptive QRD-M algorithm at Eb/No = 16 dB. Moreover, the effectiveness of complexity reduction is especially predominant at low signal- to-noise ratio (SNR).

An Improved LLR Computation for QRM-MLD in Coded MIMO Systems

Near maximum-likelihood(ML) detections with reduced complexity are promising techniques for coded multiple- input multiple-output (MIMO) systems. Particularly, the QRM- MLD algorithm is often considered in practical applications since it can easily control receivers complexity. However, it provides inaccurate log likelihood ratio (LLR) values due to the limited number of candidate symbol vectors. To overcome the problem, this paper presents an efficient LLR computation algorithm. The proposed algorithm calculates approximate LLR values calculated at every stage instead of computing LLR values only at the last stage. At each stage, the proposed algorithm updates approximate LLR values to reflect their reliability on the LLR value. Simulation results show that the proposed algorithm can obtain better performance than conventional LLR calculation scheme especially when low order modulation is employed, or the number of considering candidate vectors is small.

Performance Analysis Of A Dsttd System With Decision-Feedback Detection

We investigate closed-form bit error rate (BER) expressions for the double space-time transmit diversity (DSTTD) system with a zero-forcing decision-feedback (ZF-DF) detector. We show that the lower Alamoutis STTD unit can obtain the second order diversity gain. However, the upper one cannot guarantee the fourth order diversity due to the effect of error propagation. For example, it simply gives 3.5 dB signal-to-noise ratio (SNR) advantage over the lower one at the bit error rate (BER) of 10-3. Under the same environment, overall BER performance also suffers from 5.6 dB performance degradation over a maximum likelihood (ML) detector

Efficient Successive Interference Cancellation Algorithms for the DSTTD System

We present two ordered successive interference cancellation (OSIC) schemes using the QR decomposition for the double space-time transmit diversity (DSTTD) system. These are designed by zero-forcing (ZF) and minimum-mean-squared error (MMSE) criteria. The computational complexity is significantly reduced by utilizing special structure of the channel matrix. These schemes require fewer computations than previously known OSIC schemes. Especially the ZF-OSIC-QR achieves approximately 8% lower computational efforts than the two-stage linear ZF detection with 2.4 dB performance advantage at the bit error rate (BER) of 10-4

High-Throughput Low-Complexity Link Adaptation for MIMO BIC-OFDM Systems

This paper introduces a new link adaptation (LA) approach called the adaptive modulation, coding, and spatial mode (AMCS) scheme for multiple-input multiple-output bit-interleaved coded orthogonal frequency division multiplexing (MIMO BIC-OFDM) systems. The AMCS technique can provide both minimal performance degradation and significant throughput gain by controlling the spatial streams. With this technique, we can derive a simple and accurate closed-form expression of an instantaneous bit error rate (I-BER). Based on the I-BER, the proposed AMCS scheme chooses an appropriate modulation, coding, and spatial mode (MCS) type that improves the system performance while satisfying the quality of service (QoS) requirement. A simplified MCS type search method is also proposed, which allows the receiver to effectively select a MCS type and then send it back to the transmitter by using only a small amount of information bits. The simulation results confirm the superiority of the proposed AMCS approach over conventional LA schemes in trading high-throughput for reduced complexity.

Efficient Transmit Antenna Selection for Correlated MIMO Channels

While multiple-input multiple-output (MIMO) system gives many advantages to wireless communication, the complexity of multiple RF chains gives a burden to the system. Antenna selection is introduced as a technique to reduce the burden. On the transmit antenna selection scheme, the exhaustive search for optimal antenna set quickly becomes impractical as adopting more antennas. To reduce complexity for searching, we propose an efficient transmit antenna selection method. We first make a threshold for optimal antenna set by using Poincare separation theorem. Instead of exhaustive search, we select a transmit antenna subset that exceeds the threshold. On the exponentially correlated channel of array antennas, we give an order to search for reducing complexity further. We search from the most remotely located antenna subset which in generally less correlated to the closely located antennas which is more correlated. Finally, we show the BER performance and tradeoff relation between performance and complexity by Monte-Carlo simulations.

Iterative interference cancellation algorithms for the V-BLAST system

We propose efficient iterative interference cancellation algorithms for the vertical Bell Laboratories layered space-time (V-BLAST) system. Proposed schemes consist of conventional successive interference cancellation based on the QR decomposition (SIC-QR) and diversity combining techniques such as maximal-ratio combining or selection combining. We develop upper bounds of the SIC-QR and proposed schemes. Significant performance improvement is obtained by increasing diversity order of each stream and alleviating the effect of error propagation. In particular the proposed schemes can accomplish significant performance advantage over the ordered SIC (OSIC) in small numbers of antennas system. In addition, we compare the complexity of proposed schemes with the SIC-QR and previously reported OSIC algorithms. Proposed schemes require slightly increased complexity, when compared to the SIC-QR. It is also shown that additional computations could be reduced by controlling iteration numbers

Approaching maximum-likelihood performance with reduced complexity for a double space-time transmit diversity system

A reduced-complexity detector approaching maximum-likelihood (ML) detection performance is presented for the double space-time transmit diversity system. The proposed scheme exploits both the special structure of equivalent channel matrix and decision-feedback detection. This accounts for accomplishing near-ML or ML performance with significantly relieved computational loads. Moreover, to moderate the average complexity, several distance metric selection criteria are proposed. We can control performance and computational savings according to different distance metric selection rules. Numerical results show that the proposed detector requires significantly fewer computations than that of the Schnorr-Euchner sphere-decoding algorithm in terms of both the worst-case and the average complexity.

New DSTTD Transceiver Architecture for Low-Complexity Maximum-Likelihood Detection

This paper presents a new double space-time transmit diversity (DSTTD) transceiver exploiting the special structure of the channel matrix. This accounts for accomplishing very low complexity maximum-likelihood (ML) detector. First, the proposed ML detector requires significantly fewer computations than that of the Schnorr-Euchner sphere decoding (SE-SD) algorithm in respective of both the worst-case and the average complexity. Moreover, the complexity of the proposed ML detector is further reduced with sacrificing slight performance loss, provided that the proposed DSTTD transmitter is used.

A variable rate LDPC coded V-BLAST system using the MMSE QR-decomposition

In this paper, we propose vertical Bell laboratories layered space-time (V-BLAST) system based on variable rate low density parity check (LDPC) codes to improve performance of receiver. Because of noise enhancement through zero-forcing, MMSE detection scheme is usually used to improve the performance. We consider MMSE-QR decomposition (QRD) which has the approximately same complexity as zero forcing QRD. The performance can be improved by assigning the lower rate LDPC codes in early detected layers since the poor SNR of early detected data streams makes error propagation in upper layer. The different rate LDPC codes can be made by puncturing some rows of a given parity check matrix. Under the same overall code rate, the V-BLAST system with different rate LDPC codes has the better performance than with constant rate LDPC code in fast fading channel.