Byonghyo Shim

Generalized Orthogonal Matching Pursuit

As a greedy algorithm to recover sparse signals from compressed measurements, orthogonal matching pursuit (OMP) algorithm has received much attention in recent years. In this paper, we introduce an extension of the OMP for pursuing efficiency in reconstructing sparse signals. Our approach, henceforth referred to as generalized OMP (gOMP), is literally a generalization of the OMP in the sense that multiple <i>N</i> indices are identified per iteration. Owing to the selection of multiple “correct” indices, the gOMP algorithm is finished with much smaller number of iterations when compared to the OMP. We show that the gOMP can perfectly reconstruct any <i>K</i>-sparse signals (<i>K</i> >; 1), provided that the sensing matrix satisfies the RIP with δ_NK <; [(√<i>N</i>)/(√<i>K</i>+3√<i>N</i>)]. We also demonstrate by empirical simulations that the gOMP has excellent recovery performance comparable to <i>l</i>₁-minimization technique with fast processing speed and competitive computational complexity.

Reliable low-power digital signal processing via reduced precision redundancy

In this paper, we present a novel algorithmic noise-tolerance (ANT) technique referred to as reduced precision redundancy (RPR). RPR requires a reduced precision replica whose output can be employed as the corrected output in case the original system computes erroneously. When combined with voltage overscaling (VOS), the resulting soft digital signal processing system achieves up to 60% and 44% energy savings with no loss in the signal-to-noise ratio (SNR) for receive filtering in a QPSK system and the butterfly of fast Fourier transform (FFT) in a WLAN OFDM system, respectively. These energy savings are with respect to optimally scaled (i.e., the supply voltage equals the critical voltage V/sub dd-crit/) present day systems. Further, we show that the RPR technique is able to maintain the output SNR for error rates of up to 0.09/sample and 0.06/sample in an finite impulse response filter and a FFT block, respectively.

Recent trend of multiuser MIMO in LTE-advanced

Recently, the mobile communication industry is moving rapidly toward Long Term Evolution, or LTE, systems. The leading carriers and vendors are committed to launching LTE service in the near future; in fact, a number of major operators such as Verizon have initiated LTE service already. LTE aims to provide improved service quality over 3G systems in terms of throughput, spectral efficiency, latency, and peak data rate, and the MIMO technique is one of the key enablers of the LTE system for achieving these diverse goals. Among several operational modes of MIMO, multiuser MIMO (MU-MIMO), in which the base station transmits multiple streams to multiple users, has received much attention as a way of achieving improvement in performance. From the initial release (Rel. 8) to the recent release (Rel. 10), so called LTE-Advanced, MUMIMO techniques have evolved from their premature form to a more elaborate version. In this article, we provide an overview of design challenges and the specific solutions for MU-MIMO systems developed in the LTE-Advanced standard.

Sphere Decoding With a Probabilistic Tree Pruning

In this paper, we present a near ML-achieving sphere decoding algorithm that reduces the number of search operations in the sphere-constrained search. Specifically, by adding a probabilistic noise constraint on top of the sphere constraint, a more stringent necessary condition is provided, particularly at an early stage, and, hence, branches unlikely to be survived are removed in the early stage of sphere search. The tradeoff between the performance and complexity is easily controlled by a single parameter, so-called pruning probability. Through the analysis and simulations, we show that the complexity reduction is significant while maintaining the negligible performance degradation.

On the Recovery Limit of Sparse Signals Using Orthogonal Matching Pursuit

Orthogonal matching pursuit (OMP) is a greedy search algorithm popularly being used for the recovery of compressive sensed sparse signals. In this correspondence, we show that if the isometry constant δ<i>K</i>+1 of the sensing matrix Φ satisfies δ<i>K</i>+1 <; 1/(1/√<i>K</i>+1) then the OMP algorithm can perfectly recover <i>K</i>-sparse signals from the compressed measurements <b>y</b>=Φ<b>x</b>. Our bound offers a substantial improvement over the recent result of Davenport and Wakin and also closes gap between the recovery bound and fundamental limit over which the perfect recovery of the OMP cannot be guaranteed.

Multipath Matching Pursuit

In this paper, we propose an algorithm referred to as multipath matching pursuit (MMP) that investigates multiple promising candidates to recover sparse signals from compressed measurements. Our method is inspired by the fact that the problem to find the candidate that minimizes the residual is readily modeled as a combinatoric tree search problem and the greedy search strategy is a good fit for solving this problem. In the empirical results as well as the restricted isometry property-based performance guarantee, we show that the proposed MMP algorithm is effective in reconstructing original sparse signals for both noiseless and noisy scenarios.

Structured Compressive Sensing-Based Spatio-Temporal Joint Channel Estimation for FDD Massive MIMO

Massive MIMO is a promising technique for future 5G communications due to its high spectrum and energy efficiency. To realize its potential performance gain, accurate channel estimation is essential. However, due to massive number of antennas at the base station (BS), the pilot overhead required by conventional channel estimation schemes will be unaffordable, especially for frequency division duplex (FDD) massive MIMO. To overcome this problem, we propose a structured compressive sensing (SCS)-based spatio-temporal joint channel estimation scheme to reduce the required pilot overhead, whereby the spatio-temporal common sparsity of delay-domain MIMO channels is leveraged. Particularly, we first propose the nonorthogonal pilots at the BS under the framework of CS theory to reduce the required pilot overhead. Then, an adaptive structured subspace pursuit (ASSP) algorithm at the user is proposed to jointly estimate channels associated with multiple OFDM symbols from the limited number of pilots, whereby the spatio-temporal common sparsity of MIMO channels is exploited to improve the channel estimation accuracy. Moreover, by exploiting the temporal channel correlation, we propose a space-time adaptive pilot scheme to further reduce the pilot overhead. Additionally, we discuss the proposed channel estimation scheme in multicell scenario. Simulation results demonstrate that the proposed scheme can accurately estimate channels with the reduced pilot overhead, and it is capable of approaching the optimal oracle least squares estimator.

On further reduction of complexity in tree pruning based sphere search

In this letter, we propose an extension of the probabilistic tree pruning sphere decoding (PTP-SD) algorithm that provides further improvement of the computational complexity with minimal extra cost and negligible performance penalty. In contrast to the PTP-SD that considers the tightening of necessary conditions in the sphere search using per-layer radius adjustment, the proposed method focuses on the sphere radius control strategy when a candidate lattice point is found. For this purpose, the dynamic radius update strategy depending on the lattice point found as well as the lattice independent radius selection scheme are jointly exploited. As a result, while maintaining the effectiveness of the PTP-SD, further reduction of the computational complexity, in particular for high SNR regime, can be achieved. From simulations in multiple-input and multiple-output (MIMO) channels, it is shown that the proposed method provides a considerable improvement in complexity with near-ML performance.

A MMSE Vector Precoding with Block Diagonalization for Multiuser MIMO Downlink

Block diagonalization (BD) algorithm is a generalization of the channel inversion that converts multiuser multi-input multi-output (MIMO) broadcast channel into single-user MIMO channel without inter-user interference. In this paper, we combine the BD technique with a minimum mean square error vector precoding (MMSE-VP) for achieving further gain in performance with minimal computational overhead. Two key ingredients to make our approach effective are the QR decomposition based block diagonalization and joint optimization of transmitter and receiver parameters in the MMSE sense. In fact, by optimizing precoded signal vector and perturbation vector in the transmitter and receiver jointly, we pursue an optimal balance between the residual interference mitigation and the noise enhancement suppression. From the sum rate analysis as well as the bit error rate simulations (both uncoded and coded cases) in realistic multiuser MIMO downlink, we show that the proposed BD-MVP brings substantial performance gain over existing multiuser MIMO algorithms.

Low-Complexity Decoding via Reduced Dimension Maximum-Likelihood Search

In this paper, we consider a low-complexity detection technique referred to as a reduced dimension maximum-likelihood search (RD-MLS). RD-MLS is based on a partitioned search which approximates the maximum-likelihood (ML) estimate of symbols by searching a partitioned symbol vector space rather than that spanned by the whole symbol vector. The inevitable performance loss due to a reduction in the search space is compensated by 1) the use of a list tree search, which is an extension of a single best searching algorithm called sphere decoding, and 2) the recomputation of a set of weak symbols, i.e., those ignored in the reduced dimension search, for each strong symbol candidate found during the list tree search. Through simulations on M-quadrature amplitude modulation (QAM) transmission in frequency nonselective multi-input-multioutput (MIMO) channels, we demonstrate that the RD-MLS algorithm shows near constant complexity over a wide range of bit error rate (BER) (10-1 ~ 10-4), while limiting performance loss to within 1 dB from ML detection.