Earticle

본 논문에서는 sparse 채널에서 대표적인 채널 추정 기법들의 오차 성능을 비교 및 분석한다. 오차 성능을 비교하기 위해 크라머-라 오 경계를 이용하여 최소평균자승오차 추정기법의 하한 경계를 구하고 이를 정합 추적 기법의 상한 경계와 분석한다. 분석 결과로 부터 추정 탭 개수와 신호 대 잡음비에 따라 기존에 sparse 채널에서 효율적인 추정기법으로 알려진 정합 추적 기법보다 최소평균 자승오차 추정기법의 오차가 적을 수 있음을 보인다. 레일리이 페이딩 분포를 갖는 두 개의 sparse 채널에 대한 전산모의실험 결과 신호 대 잡음비에 따라 두 추정 기법의 오차 성능이 반전되는 경우를 보였다.

In this paper, we evaluate and analyze representative estimation methods for the sparse channel. In order to evaluate error performance of matching pursuit(MP) and minimum mean square error(MMSE) algorithm, lower bound of MMSE is determined by Cramer-Rao bound and compared with upper bound of MP. Based on analysis of those bounds, mean square error of MP which is effective in the estimation of sparse channel can be larger than that of MMSE according to the number of estimated tap and signal-to-noise ratio. Simulation results show that the performances of both algorithm are reversed on the sparse channel with Rayleigh fading according to signal-to-noise ratio.

Many wireless channels encountered in practice tend to exhibit the structure of sparse multipath due to large bandwidth or large number of antennas. High-rate data wireless communications over multipath wireless channels usually require that the channel response be known at the receiver. In this paper, a novel scheme for the estimation of sparse wireless channels is developed. The initial estimation of the channel taps is obtained by the unstructured least-squares (LS) method. Then, the presence of a channel tap is detected via an improved threshold obtained by applying the nature of sparse channels and the statistics of the noise vector. At last, the channel estimate is refined by deploying the knowledge previously acquired on the position of the nonzero taps and the structured LS method. The proposed method is compared and contrasted with the existing sparse estimation methods. And the results show the validity of the proposed method.

It is widely accepted that SC-FDE (Single-Carrier Frequency Domain Equalization) is an excellent candidate for broadband wireless systems. Channel estimation is one of the key challenges in SC-FDE, since accurate channel estimation can significantly improve the equalization at the receiver and consequently enhance the communication performances. In this paper, we proposed a simple sparse channel estimation method for SC-FDE system based on noise space, through sorting the results of the Least-squares (LS) channel estimation, channel tap locations and value are evaluated. The proposed system can realize channel estimation at a very low complexity, and simulation result shows that it can achieve significantly improved performance in frequency selective fading sparse channel.

It is widely accepted that single carrier frequency division multiple access (SC-FDMA) is an excellent candidate for broadband wireless systems. Channel estimation is one of the key challenges in SC-FDMA, since accurate channel estimation can significantly improve equalization at the receiver and, consequently, enhance the communication performances. In this paper, we study the application of compressive sensing for sparse channel estimation in a SC-FDMA system. By skillfully designing pilots, their patterns, and taking advantages of the sparsity of the channel impulse response, the proposed system realizes channel estimation at a low cost. Simulation results show that it can achieve significantly improved performance in a frequency selective fading sparse channel with fewer pilots.

천해에서 음파 전달은 매우 복잡하며, sparse한 전달 특성을 갖는다. 이러한 환경에서 수중음향통신 시스템의 성능을 향상시키기 위하여 채널을 추정하기 위한 여러 방법들이 연구되었다. 본 논문에서는 기존의 sparse-aware LMS(Least Mean Square) 알고리즘들보다 빠른 수렴속도를 갖는 LMS 기반 채널 추정 알고리즘을 제안하였다. 제안한 방법은 $L_p$-norm LMS 알고리즘과 soft decision 과정을 결합한 것이다. 모의실험은 실제 해상 실험을 통하여 얻은 수중 음속 데이터를 바탕으로 수행되었다. 그 결과 제안한 방법이 기존의 방법들보다 빠른 수렴속도와 향상된 성능을 보이는 것을 확인하였다.

Transmission characteristics of the sound propagation is very complicate and sparse in shallow water. To increase the performance of underwater acoustic communication system, lots of channel estimation technique has been proposed. In this paper, we proposed the channel estimation based on LMS(Least Mean Square) algorithm which has faster convergence speed than conventional sparse-aware LMS algorithms. The proposed method combines $L_p$-norm LMS with soft decision process. Simulation was performed by using the sound velocity profile which acquired in real sea trial. As a result, we confirmed that the proposed method shows the improved performance and faster convergence speed than conventional methods.

본 논문은 볼록 규준화 RLS(Recursive Least Squares)에 쓰이는 규준화를 위한 상수를 정하는 법을 제안한다. Eksioglu와 Tanc는 희소성 음향 채널 추정을 위해서 볼록 규준화 RLS 알고리즘을 구현하였다. 그러나 이 알고리즘은 더 좋은 추정 성능을 위해서 채널의 참 임펄스 응답 정보가 사용된다. 본 논문에서는 이 같은 참 임펄스 응답 정보가 필요 없는 규준화 상수를 위한 두 가지 선정법을 제안한다. 그리고 제안한 방법을 사용했을 때 Eksioglu와 Tanc의 방법에 필적한 추정 성능을 유지함을 보인다.

We develop two methods to select a constant in the RLS (Recursive Least Squares) with the convex regularization. The RLS with the convex regularization was proposed by Eksioglu and Tanc in order to estimate the sparse acoustic channel. However the algorithm uses the regularization constant which needs the information about the true channel response for the best performance. In this paper, we propose two methods to select the regularization constant which don't need the information about the true channel response. We show that the estimation performance using the proposed methods is comparable with the Eksioglu and Tanc's algorithm.

Since a reverberation filter is generally much shorter than the corresponding dereverberation filter, a single-channel speech dereverberation method based on reverberation filter estimation has been developed to improve its performance. Unfortunately, a typical reverberation filter still requires too many coefficients to be accurately estimated using limited speech observations. In order to exploit sparseness of reverberation filter coefficients, in this paper, we present an algorithm to impose a sparse prior to the process of reverberation filter estimation. Simulation results demonstrate that the sparse prior imposition further improves performance of the speech dereverberation method based on reverberation filter estimation.