Chungyong Lee

Non-orthogonal Multiple Access in a Downlink Multiuser Beamforming System

In this paper, we propose a non-orthogonal multiple access-based multiuser beamforming (NOMA-BF) system designed to enhance the sum capacity. In the proposed NOMA-BF system, a single BF vector is shared by two users, so that the number of supportable users can be increased. However, sharing a BF vector leads to interference from other beams as well as from the other user sharing the BF vector. Therefore, to reduce interference and improve the sum capacity, we additionally propose a clustering and power allocation algorithm. This clustering algorithm, which selects two users with high correlation and a large gain-difference between their channels, can reduce the interference from other beams and from the other user as well. Furthermore, power allocation ensures that each users transmit power is allocated so as to maximize the sum capacity. Numerical results verify that the proposed NOMA-BF system improves the sum capacity, compared to the conventional multiuser BF system.

Enhanced channel-estimation technique for MIMO-OFDM systems

In multi-input-multi-output orthogonal frequency-division multiplexing systems, conventional channel-estimation techniques using comb-type training symbols give relatively large mean squared errors (MSEs) at the edge subcarriers. To reduce the MSEs at these subcarriers, a cyclic comb-type training structure is proposed. In the proposed cyclic training structure, all types of training symbols are transmitted cyclically at each antenna. At the receiver, the channel frequency responses that are estimated using each training symbol are averaged with weights obtained from the corresponding MSEs. Computer simulations showed that the proposed cyclic training structure gives more signal-to-noise ratio gain than the conventional training structure.

Optimizing feature extraction for speech recognition

We propose a method to minimize the loss of information during the feature extraction stage in speech recognition by optimizing the parameters of the mel-cepstrum transformation, a transform which is widely used in speech recognition. Typically, the mel-cepstrum is obtained by critical band filters whose characteristics play an important role in converting a speech signal into a sequence of vectors. First, we analyze the performance of the mel-cepstrum by changing the parameters of the filters such as shape, center frequency, and bandwidth. Then we propose an algorithm to optimize the parameters of the filters using the simplex method. Experiments with Korean digit words show that the recognition rate improved by about 4-7%.

An information-theoretic perspective on feature selection in speaker recognition

This letter studies feature selection in speaker recognition from an information-theoretic view. We closely tie the performance, in terms of the expected classification error probability, to the mutual information between speaker identity and features. Information theory can then help us to make qualitative statements about feature selection and performance. We study various common features used for speaker recognition, such as mel-warped cepstrum coefficients and various parameterizations of linear prediction coefficients. The theory and experiments give valuable insights in feature selection and performance of speaker-recognition applications.

Integrated echo and noise canceler for hands-free applications

An integrated acoustic echo and noise canceler is proposed for hands-free telephony applications. The proposed system includes a computationally efficient double-talk detection (DTD) method and a residual echo cancellation algorithm. The DTD method uses two cross-correlation coefficients and is robust in adverse conditions. In the residual echo cancellation approach, a linear prediction error filter removes the short-term correlation of the residual echo and the resulting whitened residual echo is canceled by a noise-reduction filter. The proposed system was implemented using a general purpose digital signal processor (DSP) with 16-b arithmetic.

An efficient antenna shuffling scheme for a DSTTD system

A new antenna shuffling scheme for DSTTD (double space time transmit diversity) is proposed. The proposed method obtains the shuffling pattern directly from the estimated channel by maximizing minimum post-processing SNR (signal to noise ratio), while the conventional method minimizes channel correlation. Since the minimum post-processing SNR is directly related with error performance, the proposed method shows better bit error rate performance than the conventional method.

Calibration of NLOS error for positioning systems

In positioning system NLOS (non-line of sight) errors always cause remarkable positive bias to the estimation of TOA (time of arrival) or TDOA (time difference of arrival) and directly affect range measurement errors. A new method is proposed to calibrate NLOS errors by using the relationship between the mean excess delay and the delay spread measured at a mobile station. The computer simulations showed that the proposed calibration technique effectively reduces the positioning error caused by the urban NLOS environment.

An eigen-based MIMO multiuser scheduler with partial feedback information

An eigen-based MIMO multiuser scheduler which is robust to spatial channel correlation is proposed. The proposed scheduler uses the principal right singular vector for reduced feedback information and exploits the eigen-mode transmission using a precoder matrix, whose columns are the principal right singular vectors of selected users. Since the proposed scheduler uses only each principal eigen channel of the user, it is hot affected by spatial channel correlation even if all spatial channels are equal. Moreover, since the distribution of a singular value is concentrated on the principal singular value under the spatially correlated environments, more channel gain can be obtained. By computer simulations, it is shown that the proposed scheduler has better performance for both capacity and fairness under the moderate spatial channel correlation than the existing schedulers.

A wireless position location system using forward pilot signal

We propose a handset based TDOA position location system for a CDMA network. To avoid a critical modification of the handset, the proposed system exploits several delay lock loop (DLL) modules of fingers or off-line DLL modules to estimate the TOAs from the base stations. Since the synchronization performance of the DLL is degraded by the near-far effect in general, we adopt an interference cancellation technique, subtracting the reconstructed pilot signal from the received signal at the handset. It can increase the audibility and enhance the performance in estimating the TOAs of weak pilot signals from other base stations. Furthermore, the error in estimating the TOA is minimized by using an N-th order polynomial approximation. Through computer simulations, we showed that the proposed system has improved the performance in estimating parameters and locating positions.

A practical MMSE-ML detector for a MIMO SC-FDMA system

We propose a new minimum mean square error (MMSE) based maximum likelihood (ML) detector for a single carrier frequency division multiple access system. By utilizing the distance between decision line and MMSE soft-output, the proposed detector reduces the resources required to check superfluously accurate reliability of received symbol. The proposed detector provides almost same bit error rate with the conventional MMSE-ML detector while requiring significantly low computational complexity (below 10%) as SNR varies.