Low SNR Region Research Articles

Joint variable frame rate and length analysis for speech recognition under adverse conditions

This paper presents a method that combines variable frame length and rate analysis for speech recognition in noisy environments, together with an investigation of the effect of different frame lengths on speech recognition performance. The method adopts frame selection using an a posteriori signal-to-noise (SNR) ratio weighted energy distance and increases the length of the selected frames, according to the number of non-selected preceding frames. It assigns a higher frame rate and a normal frame length to a rapidly changing and high SNR region of a speech signal, and a lower frame rate and an increased frame length to a steady or low SNR region. The speech recognition results show that the proposed variable frame rate and length method outperforms fixed frame rate and length analysis, as well as standalone variable frame rate analysis in terms of noise-robustness.

Iterative Synchronization-Assisted Detection of OFDM Signals in Cognitive Radio Systems

Despite many attractive features of an orthogonal frequency-division multiplexing (OFDM) system, the signal detection in an OFDM system over multipath fading channels remains a challenging issue, particularly in a relatively low signal-to-noise ratio (SNR) scenario. This paper presents an iterative synchronization-assisted OFDM signal detection scheme for cognitive radio (CR) applications over multipath channels in low-SNR regions. To detect an OFDM signal, a log-likelihood ratio (LLR) test is employed without additional pilot symbols using a cyclic prefix (CP). Analytical results indicate that the LLR of received samples at a low SNR can be approximated by their log-likelihood (LL) functions, thus allowing us to estimate synchronization parameters for signal detection. The LL function is complex and depends on various parameters, including correlation coefficient, carrier frequency offset (CFO), symbol timing offset, and channel length. Decomposing a synchronization problem into several relatively simple parameter estimation subproblems eliminates a multidimensional grid search. An iterative scheme is also devised to implement a synchronization process. Simulation results confirm the effectiveness of the proposed detector.

Achievable Rate and Power Allocation for Single-Relay AF Systems Over Rayleigh Fading Channels at High and Low SNRs

In this paper, we propose a general method to analyze the achievable rate and to characterize the optimal power-allocation (PA) scheme for a wide range of half-duplex single-relay amplify-and-forward (AF) protocols over Rayleigh fading channels in high and low signal-to-noise-ratio (SNR) regimes. These include one-way (OW) dual-hop AF (DHAF) systems, OW nonorthogonal AF (NAF) systems, and two-phase two-way AF (2P TWAF) systems using either fixed-gain (FG) or variable-gain (VG) amplification coefficients. At high-SNR regimes, our main idea is to exploit the capacity of a two-branch maximal-ratio combining (MRC) system to obtain tight yet simple approximations to the achievable rates. In low-SNR regions, we use a simple approximation to the logarithm to compute the asymptotic achievable rates. For all considered AF protocols, the closed-form approximations are tight and easy to analyze since they involve only the exponential integral. Then, using the derived approximations, we analytically quantify the asymptotic PA schemes among the nodes to achieve the maximum rate and the sum rate for OW and TW schemes, respectively. The OW and TW protocols of interest are finally compared with the direct transmission (DT) scheme. Although OW relaying is inferior in terms of capacity in high- and low-SNR regimes, the 2P TW system might become advantageous in terms of sum-rate performance at high SNRs under the right channel conditions.

LDPC-RS Product Codes for Digital Terrestrial Broadcasting Transmission System

Product code is a promising technique for the next generation digital terrestrial broadcasting transmission system, due to its superior error correction performance. In this paper, we propose an Low Density Parity Check-Reed Solomon (LDPC-RS) product code structure, along with a novel hybrid iterative decoding scheme. The decoding scheme combines a hybrid LDPC decoding technique and RS+LDPC hard decision decoding to form a `turbo-like' decoding structure. By ingeniously ranging the order of different decoding techniques as well as performing error estimation and soft value modification at proper stages, the proposed decoding scheme greatly improves the error performance in low SNR regions while reducing the computational complexity in moderate and high SNR regions compared with the simple concatenation of LDPC and RS decoding. Moreover, we propose a rate compatible LDPC-RS product code to further reduce complexity by adaptively choosing the decoding code rate. Simulation results verify the outstanding error performance and suitability for the energy saving applications.

An Adaptively Weighted Least Square Estimation Method of Channel Mismatches in Phase for Multichannel SAR Systems in Azimuth

Multichannel synthetic aperture radar (SAR) systems in azimuth can achieve high-resolution and wide-swath imaging. However, the quality of final SAR image can be degraded by the channel mismatch in phase which increases the energy outside the processed Doppler bandwidth (PDB). To address this problem, a calibration algorithm is proposed in this letter by minimizing the energy outside the PDB. Theoretical analysis shows that the presented method can be interpreted as an adaptively weighted least square estimation problem, where the weights are related to the signal-to-noise ratio (SNR) of the echoes from different directions. Simulation results reveal that our method outperforms the conventional methods in the case of quasi-uniform sampling, particularly at the low-SNR region.

A Simplified Posteriori Estimator to Cascade Channel based on Amplify-and-Forward Multi-relaying Systems

This paper analyzes the cascade channel estimation of Amplify-and-Forward (AF) multi-relaying systems. To improve the accuracy, a maximum a posteriori (MAP) estimator is obtained with the prior probability distribution function (PDF) of the cascade channel. For lessening the computational amount of the MAP method, the complicated generalized-K distribution is approximated with a Gamma PDF based on the moment-matching method. Moreover, a closed-form and simplified MAP (sMAP) estimator is derived. Numerical simulation shows that the proposed sMAP estimator decreases the computational duration sharply than the MAP method with marginal performance loss, and outperforms its least square (LS) counterpart in the low SNR region.

High‐resolution, large dynamic range field map estimation

PurposeWe present a theory and a corresponding method to compute high‐resolution field maps over a large dynamic range.Theory and MethodsWe derive a closed‐form expression for the error in the field map value when computed from two echoes. We formulate an optimization problem to choose three echo times which result in a pair of maximally distinct error distributions. We use standard field mapping sequences at the prescribed echo times. We then design a corresponding estimation algorithm which takes advantage of the optimized echo times to disambiguate the field offset value.ResultsWe validate our method using high‐resolution images of a phantom at 7T. The resulting field maps demonstrate robust mapping over both a large dynamic range, and in low SNR regions. We also present high‐resolution offset maps in vivo using both, GRE and multiecho gradient echo sequences. Even though the proposed echo time spacings are larger than the well known phase aliasing cutoff, the resulting field maps exhibit a large dynamic range without the use of phase unwrapping or spatial regularization techniques.ConclusionWe demonstrate a novel three‐echo field map estimation method which overcomes the traditional noise‐dynamic range trade‐off. Magn Reson Med 71:105–117, 2014. © 2013 Wiley Periodicals, Inc.

Low Complexity User Scheduling, Ordering and Transmit Covariance Matrix Optimization Algorithms for Successive Zero-Forcing Precoding

In this paper we consider user scheduling, ordering and transmit covariance matrix optimization problems under successive zero-forcing (SZF) precoding for multiuser multiple-input multiple-output downlink. We propose a heuristic user scheduling metric and an intermediate user grouping technique to develop a low complexity greedy scheduling algorithm. A suboptimal user ordering technique is also proposed for transmit covariance matrix optimization under SZF. Proposed algorithm is of low complexity, but performs closely to the highly complex exhaustive search algorithm. For transmit covariance optimization under SZF, a dirty paper coding based algorithm has been previously proposed, which is computationally very complex. In this paper, we propose a suboptimal but much simplified algorithm, which employs an iterative procedure similar to a known multiple access channel (MAC) covariance optimization algorithm, but does not involve multiple levels of covariance matrix transformations. With the proposed suboptimal user ordering the exhaustive search through all possible user orders is avoided during transmit covariance matrix optimization resulting in a significant complexity reduction, and without a significant performance penalty. Simulation results show that the proposed algorithm performs very close to the known algorithm in the low SNR region.

Computationally Efficient Architecture for Accurate Frequency Estimation with Fourier Interpolation

A simplified DFT-based algorithm and its VLSI implementation for accurate frequency estimation of single-tone complex sinusoid signal are investigated. The proposed algorithm estimates frequency by interpolation using Fourier coefficients. It consists of a coarse search followed by a fine search, and its performance closely achieves the Cramer–Rao low bound (CRLB) even in low SNR region. Moreover, a pipelined triple-mode CORDIC architecture is designed to efficiently support complex multiplication, complex magnitude calculation and real division. The triple-mode CORDIC-based radix-4 architecture is employed for the hardware implementation of the frequency estimator, and is suitable for not only fast Fourier transformation but also accurate frequency estimation. A frequency estimator with 1024-point samples is implemented and verified on FPGA. It works at 215 MHz on a Xilinx XC6VLX240T FPGA device, and uses up 4,161 registers and 6,986 slice LUTs. ASIC synthesis results show that it requires an area of 60K equivalent NAND2 gates with a clock rate of 500 MHz at SMIC 0.18 μm technology. The whole latency of the frequency estimator is 2336 cycles. The proposed architecture provides a good trade off between hardware overhead, estimation performance and computation latency.

Cooperative quadrature physical layer network coding in wireless relay networks

SUMMARYThis paper proposes a cooperative quadrature physical layer network coding (CQPNC) scheme for a dual‐hop cooperative relay network, which consists of two source nodes, one relay node and one destination node. All nodes in the network have one antenna, and the two source nodes transmit their signals modulated with quadrature carriers. In this paper, a cooperative quadrature physical layer network coded decode‐and‐forward (DF) relay protocol (CQPNC‐DF) is proposed to transmit the composite information from the two source nodes via the relay node to the destination node simultaneously to reduce the number of time slots required for a transmission. The proposed CQPNC‐DF relay protocol is compared with time‐division multiple‐access amplify‐and‐forward (TDMA‐AF), TDMA‐DF, cooperative network coded DF (CNC‐DF) and cooperative analog network coded AF (CANC‐AF) relay protocols to demonstrate its effectiveness in terms of bit error rate (BER) and system throughput under different propagation conditions. The simulation results reveal that the proposed CQPNC‐DF relay protocol can significantly improve the network performance. Compared with two TDMA schemes and CNC‐DF, the proposal can provide up to 100% and 50% throughput gains, respectively. Moreover, no matter what the scene, the proposed scheme always has the lowest BER in the low SNR region. Copyright © 2013 John Wiley & Sons, Ltd.

IR-UWB 시스템을 위한 선택적 가중치 결합 에너지 검출기(S-WED)와 동기 알고리즘

본 논문에서는 IR-UWB (Ultra Wideband-Impulse Radio) 시스템을 위한 선택적 가중치를 적용한 에너지 검출기와 이에 적합한 동기 방법이 제안되었다. 구현이 용이한 에너지 검출기(ED)는 IR-UWB 시스템의 수신기로 많이 사용되나, 심볼 주기로 샘플링 된 에너지를 데이터 검출에 이용하기 때문에 수신 성능이 좋지 않은 단점이 있다. 이러한 단순한 에너지 검출기의 성능을 개선하고자 가중치를 적용한 에너지 검출기가 제안 되었다. 가중치를 적용한 에너지 검출기는 데이터 검출 이전에 동기 획득과 함께 가중치의 결정이 요구된다. 한편, 에너지를 이용한 검출 방법에서 최적의 가중치는 에너지 값이 되기 때문에 동기 획득과 가중치 획득이 동시에 이루어질 수 있다. 본 논문에서는 이러한 점을 이용하여 심볼 동기와 가중치 획득을 동시에 얻는 간단한 동기 방법을 제안한다. 또한, 제안된 알고리즘에서는 잡음 레벨 이하의 구간은 에너지 누적에서 제외함으로써 기존의 WED 보다 간단하지만 낮은 SNR에서 좀 더 향상된 성능을 가진 수신기를 얻을 수 있게 된다. 제안된 알고리즘은 IEEE 802.15.4a의 프리앰블 심볼과 채널 모델을 이용해 모의실험을 통해 검증되었다.

Delay‐limited throughput analysis of cooperative hybrid automatic repeat request in asymmetric fading channels

This study analyses the delay-limited throughput performance of an uplink cellular relay network utilising hybrid automatic repeat request (HARQ) protocol in which one mobile station (MS) communicates with one base station (BS) with the aid of one amplify-and-forward relay station (RS). The authors’ analysis allows for a practical fading scenario where MS → RS and RS → BS channels undergo independent Rayleigh and Rician fading, respectively. Delay-limited throughput expressions for two HARQ protocols are derived, respectively, namely Repetition Time Diversity and Incremental Redundancy. Furthermore, the impact of different parameters on the achievable throughput, such as the maximum HARQ round number L, line-of-sight signal K, information rate R etc., is analysed. The authors’ analysis unveils that, for the presumed dual-hop cooperation model with the specified tolerable delay requirement and given channel condition, there exists an optimal information rate R* that achieves the maximum delay-limited throughput for the two HARQ protocols. Moreover, it is disclosed that, both HARQ protocols with the adaptive optimised information rate outperform the counterpart with the fixed information rate in low and moderate SNR regions, suggesting that an adaptive transmission scheme should be employed to achieve a better throughput performance.

A New Family of Low-Complexity STBCs for Four Transmit Antennas

Space-Time Block Codes (STBCs) suffer from a prohibitively high decoding complexity unless the low-complexity decodability property is taken into consideration in the STBC design. For this purpose, several families of STBCs that involve a reduced decoding complexity have been proposed, notably the multi-group decodable and the fast decodable (FD) codes. Recently, a new family of codes that combines both of these families namely the fast group decodable (FGD) codes was proposed. In this paper, we propose a new construction scheme for rate-1 FGD codes for 2 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">a</sup> transmit antennas. The proposed scheme is then applied to the case of four transmit antennas and we show that the new rate-1 FGD code has the lowest worst-case decoding complexity among existing comparable STBCs. The coding gain of the new rate-1 code is optimized through constellation stretching and proved to be constant irrespective of the underlying QAM constellation prior to normalization. Next, we propose a new rate-2 FD STBC by multiplexing two of our rate-1 codes by the means of a unitary matrix. Also a compromise between rate and complexity is obtained through puncturing our rate-2 FD code giving rise to a new rate-3/2 FD code. The proposed codes are compared to existing codes in the literature and simulation results show that our rate-3/2 code has a lower average decoding complexity while our rate-2 code maintains its lower average decoding complexity in the low SNR region. If a it time-out sphere decoder is employed, our proposed codes outperform existing codes at high SNR region thanks to their lower worst-case decoding complexity.

High‐resolution, large dynamic range field map estimation

We present a theory and a corresponding method to compute high-resolution field maps over a large dynamic range. We derive a closed-form expression for the error in the field map value when computed from two echoes. We formulate an optimization problem to choose three echo times which result in a pair of maximally distinct error distributions. We use standard field mapping sequences at the prescribed echo times. We then design a corresponding estimation algorithm which takes advantage of the optimized echo times to disambiguate the field offset value. We validate our method using high-resolution images of a phantom at 7T. The resulting field maps demonstrate robust mapping over both a large dynamic range, and in low SNR regions. We also present high-resolution offset maps in vivo using both, GRE and multiecho gradient echo sequences. Even though the proposed echo time spacings are larger than the well known phase aliasing cutoff, the resulting field maps exhibit a large dynamic range without the use of phase unwrapping or spatial regularization techniques. We demonstrate a novel three-echo field map estimation method which overcomes the traditional noise-dynamic range trade-off.

Energy-efficient power allocation for non-regenerative OFDM relay links

We consider a two-hop non-regenerative (amplify-and-forward) relay link with one source, one destination and one relay, in which orthogonal frequency division multiplexing (OFDM) is used. The power allocation (PA) to maximize the energy efficiency for the whole link is investigated under quality of service (Qos) requirement and separate power constraints. An approximation method with low complexity is first proposed to convert the PA problem into a quasi-concave problem, which can obtain a tight lower bound in high SNR region. Furthermore, a more accurate approximation method with higher complexity is raised, which can approach the actual performance both in the low SNR and high SNR regions. We also introduce subcarrier pairing scheme to further enhance the energy efficiency for the link. Simulation results are shown to compare the performance of the two proposed methods.

Optimal Power Allocation Scheme for Downlink CoMP Systems

Coordinated multi-point transmission and reception (CoMP) scheme enable LTE-Advanced systems to achieve their higher spectral efficiency. Allowing base stations to cooperate one another is one of the solutions to mitigate the inter-cell interference (ICI). In this paper, we propose an iterative power allocation scheme with MMSE procoding based on a modified water-filling for downlink CoMP systems, which achieves the optimal performance. The simulation results show that our proposed system can achieve its optimal rate according to its antenna configuration. Comparing them with a block diagonalization (BD) shows the advantages of MMSE precoding, in particular at a low SNR region.

On Optimal Linear Filtering of Speech for Near-End Listening Enhancement

In this letter the focus is on linear filtering of speech before degradation due to additive background noise. The goal is to design the filter such that the speech intelligibility index (SII) is maximized when the speech is played back in a known noisy environment. Moreover, a power constraint is taken into account to prevent uncomfortable playback levels and deal with loudspeaker constraints. Previous methods use linear approximations of the SII in order to find a closed-form solution. However, as we show, these linear approximations introduce errors in low SNR regions and are therefore suboptimal. In this work we propose a nonlinear approximation of the SII which is accurate for all SNRs. Experiments show large intelligibility improvements with the proposed method over the unprocessed noisy speech and better performance than one state-of-the art method.

SC‐GRAPPA: Self‐constraint noniterative GRAPPA reconstruction with closed‐form solution

Parallel MRI (pMRI) reconstruction techniques are commonly used to reduce scan time by undersampling the k-space data. GRAPPA, a k-space based pMRI technique, is widely used clinically because of its robustness. In GRAPPA, the missing k-space data are estimated by solving a set of linear equations; however, this set of equations does not take advantage of the correlations within the missing k-space data. All k-space data in a neighborhood acquired from a phased-array coil are correlated. The correlation can be estimated easily as a self-constraint condition, and formulated as an extra set of linear equations to improve the performance of GRAPPA. The authors propose a modified k-space based pMRI technique called self-constraint GRAPPA (SC-GRAPPA) which combines the linear equations of GRAPPA with these extra equations to solve for the missing k-space data. Since SC-GRAPPA utilizes a least-squares solution of the linear equations, it has a closed-form solution that does not require an iterative solver. The SC-GRAPPA equation was derived by incorporating GRAPPA as a prior estimate. SC-GRAPPA was tested in a uniform phantom and two normal volunteers. MR real-time cardiac cine images with acceleration rate 5 and 6 were reconstructed using GRAPPA and SC-GRAPPA. SC-GRAPPA showed a significantly lower artifact level, and a greater than 10% overall signal-to-noise ratio (SNR) gain over GRAPPA, with more significant SNR gain observed in low-SNR regions of the images. SC-GRAPPA offers improved pMRI reconstruction, and is expected to benefit clinical imaging applications in the future.

Performance Analysis of Blind Multiuser Detectors for Cdma

This paper deals with the performance of blind multiuser detectors for CDMA is analyzed. The blind multi- user detectors are Direct Matrix Inversion (DMI) blind detector and subspace blind detector. The performance analysis is performed by means of the Signal to Interference Noise Ration (SINR) and Bit Error Rate (BER). The numerical results are plotted as variation of SINR Vs SNR and ρ , SINR with respect to correlation coefficient ( ρ ) and BER Vs signal samples (M) using MAT LAB software. It is observed that the SINR Vs SNR and ρ for both blind detectors. The performance of subspace blind detector deteriorates in the high cross correlation and low SNR region whereas the performance of the DMI blind detector is less sensitive to cross correlation and SNR in this region. The performance gain offered by the subspace detector is significant for smaller value of K, and the gain diminishes as K increases. For large number of signal samples (M), both detectors converge to the true linear MMSE detector, with the subspace blind detector converging much faster than the DMI blind detector and the performance gain offered by the subspace detector is quite significant for small values of M. The BER performance of subspace approach is better than the DMI approach with increasing SNR.

Algorithm for generating unique pilot sequences that enhance joint channel and control signal estimation

Joint channel and control information estimation in orthogonal frequency division multiplexing depends strongly on the properties of the pilot sequences used, particularly the maximum difference, autocorrelation and cross-correlation coefficients. Presented is a practical algorithm for generating near-optimal sequences which can achieve excellent performance even in the low SNR region.