Juha Karjalainen

Iterative Frequency Domain Joint-over-Antenna Detection in Multiuser MIMO

Multiuser multiple-input-multiple-output (MIMO) wireless systems have great potential in improving information rate, diversity and resistance to against interference. The primary objective of this paper is to derive for broadband signaling a new iterative frequency domain (FD) multiuser MIMO signal detection technique for joint-over-antenna (JA) detection. The proposed detector is based on soft-cancellation and minimum mean square error (MMSE) filtering, followed by maximum a posteriori probability (MAP) detector to detect several of each users transmit antennas. The purpose of jointly detecting several transmit antennas is to preserve the degrees of freedom (DoF) for MMSE. Computational complexities with FD and its time domain (TD) counterpart are evaluated in this paper, and it is shown that FD requires significantly lower complexity than TD. Numerical results show that JA significantly outperforms the receiver that detects transmit antenna signals antenna-by- antenna (AA). The proposed iterative FD JA technique achieves larger performance gains compared to AA when the total number of transmit antennas is larger than the number of receiver antennas, as well as in the presence of spatial correlation.

EXIT Chart-Based Power Allocation for Iterative Frequency Domain MIMO Detector

Transmission power allocation in single-carrier multiple-input multiple-output (MIMO) systems with iterative frequency-domain (FD) soft cancellation (SC) minimum mean-squared error (MMSE) equalization is considered. A novel framework for transmission power minimization subject to equalizer convergence constraints, referred as convergence constrained power allocation (CCPA) method, is proposed based on extrinsic information transfer (EXIT) chart analysis. The proposed method decouples the spatial interference between the streams using singular value decomposition (SVD), and minimizes the transmission power while achieving the target mutual information for each stream after iterations at the receiver. We show that the transmission power optimization can be formulated as a convex optimization problem. Three CCPA methods, one approximately optimal, and other two heuristic methods inspired by the Lagrange duality are derived. The numerical results demonstrate that the proposed scheme outperforms the existing linear precoding schemes. Moreover, the proposed heuristic schemes can achieve performance close with that of the approximately optimal method in terms of the equalizer convergence properties as well as transmission power.

On the Convergence Property of an MMSE Based Multiuser MIMO Turbo Detector with uplink Precoding

In this the paper, impact of minimum mean squared error (MMSE) based linear precoding on the convergence of iterative multiuser detector in multiuser multiple input multiple output (MIMO) uplink single carrier communications is studied. The influence of linear precoding on iterative MMSE based multiuser detector is investigated via the multi-dimensional EXIT analysis. It is shown that the use of the linear precoding enhances the separability of the EXIT planes of the simultaneous streams over without precoding; This invokes the idea that different code rate be allocated to the each transmitted streams at the transmitter. Especially, in the case of multiuser communication precoding has significant roles in the convergence property of iterative detector.

On the Activation Ordering of Detector, Decoder, and Channel Estimator in Iterative Receiver for MIMO-OFDM

An iterative receiver for a multiple-input multiple-output (MIMO) orthogonal frequency-division multiplexing (OFDM) system is considered to jointly decode the transmitted bits and estimate the channel state. The receiver consists of the a posteriori probability (APP) algorithm, the repeat-accumulate (RA) decoder, and the least-squares (LS) channel estimator. An obvious problem, with more than two blocks in an iterative receiver, is to find the optimal activation schedule of the different blocks. This paper proposes to use extrinsic information transfer (EXIT) charts to characterize the behavior of the receiver blocks and find out the optimal activation schedule for them. An analytical expression of the EXIT function is derived for the LS channel estimator. An algorithm is proposed to generate the EXIT function of the APP algorithm as a function of channel estimate mutual information (MI). Surface fitting is used to get closed form expressions for the EXIT functions of the APP algorithm and the RA decoder. A trellis search based algorithm is shown to find the convergence with the lowest possible complexity using the EXIT charts. With the proposed concept, the activation scheduling can be adapted to prevailing channel circumstances and unnecessary iterations will be avoided.

Combating Timing Asynchronism in Relay Transmission for 3GPP LTE Uplink

Relays are used between the source and the destination to improve network coverage and reliability in cooperative relay system. Imperfect time delay synchronization at the receiver arises because of time difference in propagation of the signals from the source and the relay to the receiver. The performance degrades due to interference from the previous block and the defective channel matrix. In this paper, we study the effect of time delay asynchronism and propose to use interblock interference cancelation and cyclic prefix reconstruction techniques on amplify and forward relay based distributed space-time block coding for single carrier-frequency domain equalization system, which removes the unwanted interference part and reconstructs the defective channel matrix to circulant one. The proposed method significantly reduces the effect of time delay asychronism at the receiver without disturbing the 3rd generation partnership project-long term evolution (3GPP LTE) uplink frame structure, data rate and increasing the complexity over frequency selective channels.

Antenna-by-antenna and joint-over-antenna MIMO signal detection techniques for turbo-coded SC/MMSE frequency domain equalization

This paper investigates iterative frequency domain techniques for the reception of spatially multiplexed single carrier signals transmitted over frequency-selective multiple input multiple output (MIMO) channels. The investigated equalizers are based on the soft-cancellation (SC) and minimum mean square error (MMSE) altering technique for turbo-coded single carrier point-to-point MIMO systems. We consider two different transmit antenna separation techniques in the frequency domain: (1) antenna-by-antenna (AA) and (2) joint over antenna techniques (JA). (1) aims to separate signals in different layers by MMSE filtering antenna-by-antenna, whereas (2) aims to detect the composite signal comprised of the signals transmitted from the multiple antennas. In (2) the composite received signal is decomposed by using the spatial maximum a posteriori probability (MAP) algorithm. We evaluate performances in terms of frame error rate (FER) and throughput in point-to-point MIMO frequency-selective fading channels. Impacts of spatial correlation on performance of the two detectors are also investigated in this paper.

Scheduling of the Activations in Iterative Detection, Decoding, and Channel Estimation for MIMO-OFDM

An iterative receiver for a multiple-input-multiple-output (MIMO) orthogonal frequency-division multiplexing (OFDM) system is considered to jointly decode the transmitted bits and estimate the channel state. The receiver consists of the a posteriori probability (APP) algorithm, the repeat-accumulate (RA) decoder, and the least-squares (LS) channel estimator. An obvious problem, with more than two blocks in an iterative receiver, is to find the optimal activation schedule of the different blocks. This paper proposes to use extrinsic information transfer (EXIT) charts to characterize the behavior of the receiver blocks and find out the optimal activation schedule for them. A semi-analytical expression of the EXIT function is derived for the decision directed LS channel estimator. An algorithm is proposed to generate the EXIT function of the APP algorithm as a function of channel estimate mutual information (MI). Surface fitting is used to get closed form expressions for the EXIT functions of the APP algorithm and the RA decoder. Trellis search based algorithms are shown to find the convergence with the lowest possible complexity using the EXIT charts. With the proposed concept, the activation scheduling can be adapted to prevailing channel circumstances and unnecessary iterations will be avoided.

On convergence constraint precoder design for iterative frequency domain multiuser SISO detector

Convergence constraint power allocation (CCPA) in single carrier multiuser (MU) single-input single-output (SISO) systems with iterative frequency-domain (FD) soft cancelation (SC) minimum mean-squared error (MMSE) equalization is considered in this paper. In order to exploit full benefit of iterative receiver, convergence properties need to be considered. The proposed scheme can guarantee that the desired mutual information point/value after sufficient amount of iterations is achieved. In this paper, successive convex approximation algorithm is proposed as a solving the non-convex convergence constraint power minimization problem. Furthermore, the results of EXIT-chart analysis demonstrate that the CCPA design can achieve the objectives described above.

Timing and Frequency Offsets Compensation in Relay Transmission for 3GPP LTE Uplink

Relays can be used between the source and the destination to improve network coverage or reliability. In the relay based distributed space-time block-coded (DSTBC) single-carrier (SC) transmission scheme, asynchronism in time and frequency causes degradation in system performance because of the interblock interference, non-coherent combining and multiuser interference. The simultaneous presence of timing offset and carrier frequency offset also destroy the orthogonal structure of DSTBC. In this paper, we study the combined effect of timing offset and carrier frequency offset and propose a two stage equalization technique for this system at the destination terminal. Technique is based on interference cancellation and cyclic prefix reconstruction. The proposed equalization technique maintains the orthogonal structure of DSTBC and allows the use of low complexity one-tap frequency-domain equalizer. The technique significantly alleviates the effect of time offsets and frequency offsets at the destination without increasing the complexity of the receiver or disturbing the 3rd generation partnership project-long term evolution (3GPP LTE) uplink frame structure or the data rate over frequency selective channels.

Convergence analysis of MMSE based multiuser MIMO turbo detector with linear precoding strategies

This paper studies mutual information transfer properties of iterative multiuser detector with linear precoding schemes for single carrier communications in multipoint-to-point multiple input multiple output (MIMO) channels. Based on multidimensional extrinsic information transfer (EXIT) analysis, we analyze the impact of linear precoding on the convergence properties of the detectors jointly designed with the precoders based on minimum sum mean squared error (MMSE) and maximum sum mutual information rate criteria. Results demonstrate that the use of the linear precoding schemes enhance the separability of the EXIT curves of the simultaneous streams over without precoding; This invokes the idea that different code rate be allocated to the each transmitted streams at the transmitters.