Search

This paper proposes a novel microphone array speech denoising scheme based on tensor filtering methods including truncated HOSVD (High-Order Singular Value Decomposition), low rank tensor approximation and multi-mode Wiener filtering. Microphone array speech signal is represented in three-order tensor space with channel, time, and spectrum modes and then tensor filtering model can be designed to process the multiway array data. As to the first method, noise can be reduced through the truncated HOSVD which is a simple scheme in tensor processing. It is more accurate to find the lower-rank approximation of the three-order tensor with Tucker model. Then MDL (Minimum Description Length) criterion is used to estimate the optimal tensor rank in the second method. Further, multimode Wiener filtering approach upon tensor analysis can be considered as the spanning of one-mode wiener filtering. How to take advantages of tensor model to obtain a set of filters is the heart of the novel scheme. The performances of the proposed three approaches are evaluated with objective indexes and listening quality test. The experimental results indicate that the proposed tenor filtering methods have potential ability of retrieving the target signal from noisy microphone array signal and the multi-mode Wiener filtering method provides the best denoising results among the three ones.

This paper deals with low-complexity joint channel estimation and decoding for faster-than-Nyquist (FTN) signaling over frequency selective fading channels. The inter-symbol interference (ISI) imposed by FTN signaling and the frequency selective channel are intentionally separated to fully exploit the known structure of the FTN-induced ISI. Colored noise due to the faster sampling rate than that of the Nyquist signaling system is approximated by autoregressive process. A Forney style factor graph representation of the FTN system is developed and Gaussian message passing is performed on the graph. Expectation propagation (EP) is employed to approximate the message from channel decoder to Gaussian distribution. Since the inner product between FTN symbols and channel coefficients is infeasible by belief propagation (BP), we propose to perform variational message passing (VMP) on an equivalent soft node in factor graph to tackle this problem. Simulation results demonstrate that the proposed low-complexity hybrid BP-EP-VMP algorithm outperforms the existing methods in FTN system. Compared with the Nyquist counterpart, FTN signaling with the proposed algorithm is able to increase the transmission rate by over 40%, with only negligible BER performance loss.

In this letter, we study the receiver design for a coded faster-than-Nyquist signaling system in time-variant intersymbol interference channels, also known as doubly selective channels. The conventional minimum mean squared error (MMSE) equalizer results in high complexity. We develop a low-complexity iterative message passing algorithm toward this end. By reformulating the problem into a linear state-space model, Gaussian message passing is performed on a vector-form factor graph. Furthermore, we consider the imperfect channel information scenario to improve the robustness of the proposed algorithm. Simulation results show that the proposed algorithm outperforms the variational inference method, and it performs close to the frequency-domain MMSE equalizer with much lower computational complexity.

In this paper, a linear precoding scheme is proposed to maximize the total throughput of a broadband geostationary earth orbit (GEO) satellite system based on traffic demands. The weighted max-min fair (WMMF) problem under the per-antenna power constraints and the traffic demands constraints is investigated to achieve this purpose. Semidefinite relaxation (SDR) approach and bisection search method are jointly adopted to solve the corresponding related problem, and then Gaussian randomization and a power control problem are proposed to get a suboptimal solution of the WMMF problem. Different traffic demands scenarios and weighted values selection schemes are considered to analyze the impact of them on the precoding design. Simulation results demonstrate that the total performance of the proposed scheme has been improved significantly over the traditional schemes and is better than the rate balance scheme.

Wireless networks are growing in the value of application in many areas, in which accurate clock synchronization is required when tasks are performed in a collaborative fashion among nodes. Especially, cooperative synchronization techniques lead to significant performance improvement compared with traditional methods. However, the correlation among agents renders the performance analysis of cooperative network synchronization difficult. In this paper, we introduce the concept of information coupling intensity to the analysis of interaction between agents. Our approach enables us to derive closed-form asymptotic expressions under specific network topologies, and relate them to various network parameters., National Natural Science Foundation of China (Grant No. 61571041), National Natural Science Foundation of China. Foundation for the Author of National Excellent Doctoral Dissertation (Grant 201445)

Location information has become attractive for a variety of applications in wireless networks. Cooperative localization was proposed to improve the performance in harsh environment where conventional localization methods failed due to the insufficient number of anchors. In this paper, distributed cooperative localization is studied based on message passing on factor graph. The joint a posteriori distribution of nodes? positions is represented by factor graph. Due to the nonlinearity between the positions and range measurement, the expressions of messages cannot be obtained in closed form by directly applying sum-product algorithm. Most existing methods resort to particle-based representation, which leads to both high computational complexity and large communication overhead. We propose to replace the factor node of the likelihood function by a linear Gaussian model. Accordingly, we are able to derive Gaussian messages on the revised factor graph, which only requires to update the mean vectors and the covariance matrices of multivariate Gaussian distributions. Simulation results show that the proposed method significantly outperforms the distributed maximum likelihood estimator and the extended Kalman filter, and performs very close to or even better than SPAWN estimator with much lower communication overhead and computational complexity in both static and mobile wireless networks. HighlightsA distributed cooperative localization based on message passing is proposed in the paper.Gaussian parametric messages are utilized to represent the messages on factor graph to reduce the communication overhead.Relying on the linearization on the nonlinear observation functions, messages are derived with Gaussian forms, which leads to efficient computations.To further reduce the communication overhead, a broadcast message scheme is considered.The proposed algorithm reaches superior performance compared to the SPAWN, EKF and ML.

The nonlinear distortion of wideband signal due to the filtering and efficiently operated high power amplifiers limits the performance of satellite communications. Volterra series can be used to describe the nonlinear satellite channels effectively. Most existing equalizers simply ignore the nonlinear terms or treat all the nonlinear combinations of symbols as interference. In this study, by properly exploiting information from nonlinear terms, we propose three turbo equalizers for nonlinear satellite channels, namely, joint Gaussian (JG), soft interference cancellation-minimum mean square error (SIC-MMSE) and linear minimum mean square error (LMMSE) equalizers. In JG and SIC-MMSE-based equalizers, both the linear and nonlinear terms that contain the symbol of interest are considered as desired signals. Accordingly, the required statistics are calculated based on the a priori probabilities of coded bits from output of channel decoder. For LMMSE-based equalizer, we propose to calculate the extrinsic information from output of equalizer by excluding the prior information in both the linear and nonlinear terms. Simulation results demonstrate that the proposed equalizers significantly outperform the method which ignores the presence of nonlinear interferences. Moreover, the nonlinear terms that contain the symbol of interest can be exploited to further improve the performance of turbo equalization.

In this paper, a code-aided maximum-likelihood and moment-based joint SNR estimator is proposed for M- ary amplitude phase shift keying (APSK) signals over AWGN channels. The proposed estimator significantly improves the performance at low SNRs by utilizing the syndrome in the LDPC codes to act as a reference measurement of estimation performance. Moreover, a methodology to measure the performance of the estimation by the syndrome of the LDPC codes is derived and the simulation results reveal that the number of 0s in the syndrome presents a positive correlation with the SNR. Compared with code-aided maximum-likelihood (ML) estimators and momentbased estimators for M-APSK signals, it is validated that the proposed joint SNR estimator has integrated the advantages of the classical approaches and simulation results also show that the proposed estimator exploiting the decision metric of the selector performs better at the SNR estimation range.

Localization techniques are crucial for wireless networks in the age of big data. Exploiting pairwise measurements between agents, the scheme of cooperative localization yield superior performance and has been discussed extensively. However, despite its great significance for network design, the efficiency of cooperation remains an open topic. By introducing a decomposition of the Fisher information matrix (FIM), this paper gives a natural definition of cooperation efficiency, and relate it to the information coupling among nodes. Geometric interpretations of this decomposition is given and illustrated by numerical examples.

In conventional localization, agents may not be able to reach sufficient number of anchors to obtain unambiguous locations, especially in sparse networks. Cooperative localization is a promising solution in that harsh environment, which enables the agents to cooperate with each other by exchanging location information and performing range measurements. In this paper, a distributed cooperative localization method based on factor graph is proposed in wireless networks. Gaussian parametric messages are used to represent the messages passed on factor graph. However, because of the nonlinear observation model, no closed-form solutions can be obtained. To solve this problem, the Taylor expansion is used to approximate the nonlinear terms in message updating, which leads to the Gaussian message passing on factor graph. Accordingly, only two parameters of the Gaussian distribution have to be transmitted and the communication overhead for localization can be significantly reduced. The two proposed algorithms for the application with accurate and inaccurate anchors, respectively, are evaluated by Monte Carlo simulations and compared with the SPAWN and maximum likelihood (ML) estimator. The results show that the proposed algorithms can perform very close to SPAWN with much lower computational complexity, and it outperforms the ML estimator significantly.

Localisation in wireless sensor networks (WSNs) has received much attention, where most studies focus on mitigating the effects of measurement noise under the assumption of accurate anchors’ positions. However, anchors’ positions could be inaccurate for the inevitable errors in practical observations. This paper studies the sensor localisation with both inaccurate anchors’ positions and noisy range measurements in WSNs. To solve the intractable integrals in likelihood function, the authors propose to use expectation-maximisation (EM) algorithm to obtain the maximum likelihood (ML) estimation iteratively. The ‘a posteriori’ distribution of the anchor's position uncertainty is approximated to a circularly symmetric Gaussian distribution by minimising the Kullback-Leibler divergence between them. Building on this, the authors derive the expectation step in a closed-form expression. In the maximisation step, based on the Taylor expansion of the confluent hypergeometric function of the first kind presented in the expectation step, analytical solutions are obtained. Simulation results show that the proposed EM estimator significantly outperforms the approximated ML estimator. The performance gain by using the EM estimator becomes larger as the increase of anchors’ position uncertainties. Moreover, the performance of the EM estimator is close to that of the Monte Carlo-based estimator with much less computational complexities.

In this paper, we investigate a robust filter-and-forward (FF) beamforming design for two-waymulti-antenna relaying networks, where multiple relays assist two terminals to exchange information. With Gaussian distributed channel errors, the proposed robust beamforming design aims at maximizing the signal-to-interference-plus-noise-ratio (SINR) under individual transmit power constraints at each relay. Exploiting the elegant convex optimization mathematical tools, the optimization problem can be efficiently solved. Finally, simulation results demonstrate that the proposed robust beamformer reduces the sensitivity of the two-way multi-antenna relay networks to channel estimation errors, and outperforms the algorithm with estimated channels only. Moreover, the length of robust beamformer also influences the system performance.

Location-aware wireless networks can provide precise location information in harsh environments only when all anchors are well-functioning. In this paper, we propose a distributed, robust TOA-based localization approach based on the Expectation-Maximization (EM) algorithm, which enables the agents to detect the malfunctioning anchors cooperatively while localizing themselves. Closed-form solution is obtained by using Taylor approximation. Moreover, performance limit is analyzed using Cramer-Rao lower bound (CRLB), which reveals that cooperative outlier detection brings theoretical localization performance gain. Simulation results show that the proposed method attains the performance limit with a significantly lower computational cost compared with the existing algorithms.

The location awareness of a “passive” target has drawn much attention in recent years. Several papers have investigated passive localization in different scenarios. In this paper, we build a factor graph framework and show that the factor graph can be modified easily for different passive localization problems. Then the beliefs of unknown variables are obtained via message passing algorithms. To reduce the huge complexity of nonparametric message passing, we propose a damped expectation propagation method which restricts the messages on factor graph to Gaussian distributions. Simulations results verify the feasibility of proposed algorithm.

Code-aided (CA) iterative carrier phase synchroniser can greatly improve the accuracy of phase estimation in turbo receivers. However, because of the iteration involved between phase recovery and decoding, most existing studies depend on extensive simulations rather than on theoretical analysis to evaluate the performance improvement of phase recovery by exploiting the coding constraints. In this study, the authors propose analytical methods to fill this void. The first step is to approximate the cross-talks caused by phase offset as Gaussian noise at low signal-to-noise ratios. Then, a semi-analytical method is proposed to express the distribution of extrinsic information from channel decoder as a function of phase offset. Building on this model, both the open-loop and closed-loop performances of CA iterative phase synchronisers are derived in closed-form. The analytical results explicitly reveal how extrinsic information contributes to the performance of carrier phase estimation. Monte Carlo simulation results corroborate that the proposed methods are able to accurately characterise the performance of CA iterative carrier phase recovery for systems with different channel codes.

We analyze the performance of a code-aided (CA) decision-directed (DD) timing synchronizer, which can exploit the dependence structure across coded symbols to improve the timing recovery accuracy. Due to the inherent coupling between timing recovery and decoding, most existing studies rely on extensive simulation rather than on analytical methods to evaluate performance of timing recovery for coded systems. We propose analytical methods in this paper towards this end. A first key step is to approximate timing-offset-induced inter-symbol interference (ISI) as an additive Gaussian noise, since in the low signal-to-noise ratio (SNR) regime the background noise is large enough to mask the ISI. Then, we derive semi-analytical expressions for the mean and variance of extrinsic information as functions of timing offset, building on which we characterize both open-loop and closed-loop performance of decision-directed timing synchronizers. Monte Carlo simulation results corroborate that the proposed method accurately characterizes the performance of CA DD timing recovery, for systems with a wide range of channel bandwidth and different channel codes.

This paper studies TOA-based localization of a passive target with one transmitter and multiple distributed receivers. Different from the existing studies which assume synchronous or quasi-synchronous network, we consider a more practical case in which all the receivers have different time offsets to the transmitter. To tackle the multi-dimensional optimization problem in the maximum likelihood estimation, an expectation maximization (EM) algorithm is proposed. Taylor expansion to the result obtained in the expectation step is applied, which enables us to give a closed-form solution of localization in the maximization step. The Cramer-Rao bound of the target's position estimation with asynchronous receivers is derived. Simulation results show that the proposed EM-based algorithm outperforms the traditional methods and can perfectly attach the derived Cramer-Rao bound.

In this paper, we propose a distributed simultaneous localization and tracking (SLAT) algorithm in wireless sensor networks. Belief propagation (BP) algorithm is applied on the factor graph which represents the factorization of the joint posterior distribution function. Due to the nonlinearity between the observations and location variables, closed-form expression of messages cannot be obtained by directly applying BP on factor graph. We resort to the Taylor expansion to approximate the nonlinear terms. Accordingly, Gaussian messages and the beliefs of the location variables can be derived. Due to the noncooperation of the target, the posterior position distribution has to be calculated by sensors distributively. We propose to use an average consensus algorithm to estimate the parameters of the target's posterior position distribution. Monte Carlo simulations showed that the proposed SLAT algorithm performs close to the particle-based BP algorithm, with significantly lower computational complexity and communication overhead, which makes it very attractive in practical applications.

In wireless sensor networks, acquiring accurate timing information is a crucial requirement for time-based sensor localization. Utilizing a joint localization and synchronization method in sensor networks can improve positioning speed and accuracy. In this paper, we present a unified factor graph framework based on time of arrival (TOA) measurements to solve the problem of joint localization and time synchronization. A novel distributed cooperative joint estimation method based on belief propagation (BP) is proposed. We linearize the nonlinear terms in messages on factor graph in order to obtain a closed Gaussian form solution of message update. Accordingly, only the means and variances have to be updated and transmitted, which significantly reduce the communication overhead and computational complexity. To further reduce the communication overhead, we propose a message passing schedule. Simulation results show that the proposed BP method reach close performance to particle-based approaches with lower complexity.

Location awareness is now becoming vital requirement for many practical applications. In this paper, we consider a wireless network where there is one transmitter and multiple receivers and they aim to locate a passive target. Existing studies on passive localization assume receivers' positions are perfectly known. However, receivers' positions are generally inaccurate in actual scenarios. A factor graph based message passing algorithm is proposed to locate the passive target with inaccurate receivers. The nonlinear term in likelihood function is linearized to solve the intractable integrals. Accordingly, all messages on factor graph are obtained in Gaussian closed forms which reduce the computational complexity significantly. Simulation results show the proposed algorithm can not only estimate the target's position accurately but also determine the receivers' positions simultaneously with low complexity.

Cooperative localization has become a novel technique to improve the performance in harsh environment with insufficient anchors. In this paper, distributed cooperative localization is studied based on message passing on factor graph. The factor graph is constructed according to the joint a posteriori distribution of nodes' positions. Because of the nonlinearity between the positions and observations, the updates of messages using the sum-product algorithm cannot obtain closed forms. To tackle the nonlinearity, the linearization of range measurement based on its own and the neighbors' positions is utilized. Accordingly, we are able to derive all the messages on factor graph and the nodes' position posteriors as Gaussian distributions, which only requires to update the mean vectors and covariance matrices. Numerical results are performed for both static and mobile networks. The performance of the proposed algorithm reaches better performance than the maximum likelihood estimator, SPAWN, and extended Kalman filter. Besides, the proposed algorithm holds the low communication overhead by transmitting the parametric messages between the nodes.

Localization in wireless sensor networks has become an attractive research field in recent years. Most studies focus on the mitigation of measurement noise by assuming the positions of anchors are perfectly known, which may become impractical due to some inevitable errors in the observations of anchors' positions. This paper addresses the problem by taking into account the a priori position information of inaccurate anchors. Considering that the maximum likelihood (ML) algorithm suffers from the intractable integrals involved, we resort to expectation maximization (EM) algorithm to solve this problem iteratively. The a posteriori probability of the anchor position is approximated by circularly symmetric Gaussian distribution, with parameters optimized by minimizing Kullback-Leibler divergence of the two distributions. Building on this approximation, we are able to derive the expectation step in closed form. Particle swarm optimization is then followed to perform the maximization step. Numerical results demonstrate that the proposed EM estimator is less sensitive to the anchors' uncertainties and it significantly outperforms the traditional ML estimator which ignores the prior information of anchors.

A low complexity non-data-aided (NDA) signal-to-noise ratio (SNR) estimator is proposed for quadrature phase shift keying (QPSK) signals in complex additive white Gaussian noise (AWGN) channel. The novel SNR estimator utilizes the constant amplitude characteristics of the in-phase and quadrature components of QPSK modulation to evaluate the signal power and the noise power separately. The estimation mean and normalized mean square error (NMSE) of the proposed SNR estimator are investigated by Monte Carlo simulations. Results demonstrate that, compared with M2M4 algorithm and Ijaz's method, the proposed estimator saves about 80% and 50% multiplications, respectively. Furthermore, the proposed method performs better than other NDA SNR estimators at the same computational complexity.

Code-aided (CA) iterative carrier phase synchronizer can improve the phase estimation performance significantly. However, due to the coupling involved between phase recovery and decoding, most studies depend on extensive simulations rather than on theoretical analysis to evaluate the performance of CA phase recovery. In this paper, we propose analytical methods to fill this void. The first step is to model the cross-talks caused by phase offset as an additional Gaussian noise at low signal-to-noise ratios (SNRs). Then, a semi-analytical method is proposed to express the distribution of extrinsic information from channel decoder as a function of phase offset. Building on this model, both the open-loop and closed-loop performance of CA iterative hard/soft decision-directed phase synchronizers are derived in closed-form. The analytical results explicitly reveal how extrinsic information contributes to the performance improvement of carrier phase estimation. Monte Carlo simulation results corroborate that the proposed methods are able to accurately characterize the performance of CA iterative carrier phase recovery for systems with different channel codes.

In this paper, we focus on resource allocation for distributed broadband heterogeneous networks. Considering the fairness among different users, using the game-theoretic diagram, a fair resource allocation scheme is proposed first under each mobile user's minimal rate constraint. Then, an optimal power allocation algorithm is derived based on the Lagrange optimization method. To further improve the convergence speed and reduce the computational complexity, an adaptive scalarization method is proposed. At the end, the persuasive simulation results show that the proposed approach not only enhances the overall system capacity compared with the scheme without power allocation, but also guarantees fairness among all the users.

This paper proposes a tone labeling technique for tonal language speech synthesis. Non-uniform segmentation using Viterbi alignment is introduced to determine the boundaries to get F0 symbols, which are used as tonal label to eliminate the mismatch between tone patterns and F0 contours of training data. During context clustering, the tendency of adjacent F0 state distributions are captured by the state-based phonetic trees. Means of tone model states are directly quantized to get full tonal label in the synthesis stage. Both objective and subjective experiment results show that the proposed technique can improve the perceptual prosody of synthetic speech of non-professional speakers.

In this paper, we investigate robust beamforming design for two-way amplify-and-forward (AF) MIMO relaying system with Gaussian distributed channel errors. Take mean-square-error (MSE) of the detected data as the design criteria, beamforming matrix at relay is optimized under imperfect channel state information (CSI). Both frequency-division-duplex (FDD) system and time-division-duplex (TDD) system are considered. Then two algorithms are proposed based on a framework of quadratic programming. The effectiveness of the proposed robust design for different channel mode are finally demonstrated by the simulation results.

In this paper, we propose a novel signal-to-noise ratio (SNR) estimation method for linear modulations on additive white Gaussian noise (AWGN) channel. It estimates noise power directly without estimating the received total power. The estimation mean and mean square error (MSE) are analyzed theoretically and verified by simulations. Results show that the proposed SNR estimator performs better at low SNR than the conventional estimators while it suffers a little degradation at high SNR. Moreover, the proposed estimator can be implemented with less real multiplications than the conventional ones.

In this paper, we investigate maximum mutual information design for multi-hop amplify-and-forward (AF) multiple-input multiple-out (MIMO) relaying systems with imperfect channel state information, i.e., Gaussian distributed channel estimation errors. The robust design is formulated as a matrix-variate optimization problem. Exploiting the elegant properties of Majorization theory and matrix-variate functions, the optimal structures of the forwarding matrices at the relays and precoding matrix at the source are derived. Based on the derived structures, a water-filling solution is proposed to solve the remaining unknown variables.

The determination of the optimal fractional Fourier transform (FrFT) order is a crucial issue for FrFT. This paper introduces a novel algorithm is proposed for the estimation of FrFT order. We use the information of pitches, harmonies and formants in the correlogram of Gammatone filterbanks to get a few candidates of the transform order. The proposed method reduces the computation complexity in the searching of optimal transform order. We apply this method for speech processing such as Mel-frequency cepstral coefficients (MFCC) extraction and speech enhancement. The experiment of MFCC extraction shows that the proposed method is superior to the traditional method based on Fourier Transform in the sense of Fisher distance improvement. The experimental results of speech enhancement also show the improvement in sense of SNR and the Itakura-Saito distance of LPC coefficients.

In this paper, optimal linear transceiver designs for multi-hop amplify-and-forward (AF) Multiple-input Multiple-out (MIMO) relaying systems with Gaussian distributed channel estimation errors are investigated. Some commonly used transceiver design criteria are unified into a single matrix-variate optimization problem. With novel applications of majorization theory and properties of matrix-variate function, the optimal structure of robust transceiver is first derived. Based on the optimal structure, the original transceiver design problems are reduced to much simpler problems with only scalar variables whose solutions are readily obtained by iterative water-filling algorithms. The performance advantages of the proposed robust designs are demonstrated by the simulation results., Comment: Accepted by WCSP2011

A quality model was built to assess the influence of speech-video asynchrony on the audio-visual quality perception. The audio-visual contents were separated into two categories: “speaker inside” and “speaker outside”, depending on whether the speaker is inside the video. For the first category, speech was shifted in a small scale. DCT and MFCC coefficients were calculated from video and speech separately. A Co-inertia Analysis (CoIA) was used to decide the speech-video correlation, and as the speech progressively shifts, a correlation curve emerged. The curve was modeled by an Gaussian function, and then the function was used to predict the perceptual quality. On the other hand, a Gaussian curve was used to predict the perceptual quality of the “speaker outside” category. A subjective test proved the effectiveness of the proposed method.

This article introduces fractional Fourier transform (FrFT) to speech enhancement, A novel algorithm is proposed for the estimation of FrFT order. The determination of the optimal FrFT order is a crucial issue for FrFT. We use the information of pitches, harmonies and formants in correlogram of Gammatone filterbank to get a few candidates of the transform order. The proposed method reduces the computation complexity in the searching of optimal transform order. The experimental results of speech enhancement show that the proposed method is superior to the conventional spectral subtraction in the sense of SNR improvement of the enhanced speeches and the Itakura-Saito distance of LPC coefficients.

In this paper, the acquisition performances of different detectors are investigated and evaluated for frame acquisition in DVB-S2 system. Mean acquisition time in both single-dwell and multi-dwell procedures are derived based on a Markov chain model. Since it is quite complicated to analyze some parameters in close formula, a semi-analytical method is adopted to simplify the analysis. Performance results show that, in contrast with post detection integration (PDI) based detector, differential detector is robust to large frequency offset with a little but acceptable complexity increase. Moreover, differential detector can provide almost the same acquisition performances in both scenarios with and without large frequency offset.

In this paper, an efficient user selection method for traditional iterative Water-Filling algorithm in Gaussian multiuser broadcast channel is developed to reduce the computational complexity. The new algorithm selects some effective users from all the users, and then allocates the entire power among the selected users applying Water-Filling algorithm. The vector Gaussian broadcast channel and the matrix Gaussian broadcast channel are considered respectively and the corresponding user selection methods are proposed. Theoretical analysis and simulation results show that the new algorithm can significantly improve the computational efficiency of the traditional iterative Water-Filling power allocation and can obtain almost the optimal sum capacity.

An improved data-aided (DA) frequency error detector (FED) and a frequency lock detector are proposed under large frequency offset for Digital Video Broadcasting Satellite Second Generation (DVB-S2) system. Computer simulations results show that the proposed error detector can increase the frequency acquisition range and decrease the acquisition time without complexity increase. Its closed-loop normalized frequency root mean square error (RMSE) improves at least 1.5 dB compared with that of conventional error detector. The proposed lock detector shows good lock indication. Its modified version can save more symbols to indicate the locking status.

We propose an acoustic feature for speech recognition based on the combination of MFCC and fractional Fourier transform (FrFT). The transform orders for FrFT are adaptively set according to the intraframe pitch change rate. This method is motivated by the fact that the speech is not stationary even in a short period of time, and the idea is shown using an AM-FM speech model and some spectrograms of an artificial periodic signal. Experiments were conducted on the intervocalic English consonants provided by Interspeech 2008 Consonant Challenge and a Mandarin connected digits corpus. The performance of the proposed method is compared with the MFCC baseline system. Experimental results show that the proposed features get a slightly better recognition rate than MFCCs presumably because they can better track the dynamic characteristics of the speech harmonics.

A non-intrusive objective measurement for estimating the quality of output speech without input clean speech is proposed for both narrowband and wideband speech based on Gaussian mixture model (GMM) and support vector regression (SVR). Perceptual linear predictive (PLP) features are extracted and clustered by GMM as an artificial reference model from clean speech. Input speech is separated into three classes, for which the consistency measures between features of the test speech signal and the pre-trained GMM reference model are calculated and mapped to an objective speech quality score using SVR method. Based on the three narrowband and two wideband MOS (mean opinion score) labeled test databases, the correlation degree between subjective MOS and objective MOS is analyzed. Experiment results show that the proposed method is an effective technique and performs better than ITU-T P.563 and MNLR (multivariate non-linear regression) method for most of the test conditions.

An precise but simple link quality model is crucial to improve the accuracy and flexibility of link adaptation design of LDPC coded OFDM system, because it acts as a link-to-system interface which provides system simulator immediately the BLER performance of a certain modulation and coding scheme corresponding to the current SINR states, which is an instantaneous vector. This contribution extends the classical exponential effective-SNR mapping model to rate-compatible LDPC codes. Theoretical analysis and simulation results indicate that the EESM model eliminates the influence of different modulation modes and channel states, by selecting proper modulation and coding adjusting factors. Therefore it can be treated as an accurate link to system interface for LDPC coded OFDM system. Furthermore adaptive modulation and coding scheme is designed to verify the accuracy of the model. The predicted results match the simulation ones very well, both over AWGN and TU50 channels, which confirms that the EESM model offers accurate and convenient look-up tables for LA design of LDPC coded OFDM system.

An accurate and simple link quality model is crucial in system evaluation of different link adaptation (LA) solutions, for it acts as a link-to-system (L2S) interface which provides the block-error rate (BLER) performance of a certain modulation coding scheme (MCS) corresponding to the current SINR states. Our previous work in the work of Chen et al. (2007) has verified the good accuracy of the MI-based model for rate compatible LDPC coded OFDM system. In this paper, adaptive modulation and coding (AMC) strategy and two hybrid ARQ schemes of variable retransmission length, partial chase combining and full incremental redundancy, have been designed to check the accuracy of the look-up tables (LUTs) obtained by the MI-based model. The predicted results are quite in accord with the simulation ones, which confirms that the MI- based model provides accurate and simple LUTs for LA design of LDPC codes.

To improve the flexibility and accuracy of link adaptation (LA) of LDPC coded OFDM system, the classical exponential effective-SNR mapping (EESM) link quality model is extended to LDPC codes. Theoretical analysis and simulation results indicate that by selecting proper modulation and coding adjusting factors, the EESM model is precise for LDPC codes, both in terms of complexity and accuracy for various modulation and coding schemes (MCS), sub- channel states and coding block lengths. Therefore it can be treated as an accurate link to system (L2S) interface for LDPC coded multi-carrier system. To verify it, a new LA scheme based on LDPC codes specified in IEEE802.16e standard is presented. Based on the EESM model, the link-level performance of BPSK over additional white Gaussian noise (AWGN) channel is used to predict the throughput performances of the LA scheme, which match the simulation results very well.

We firstly give the definition of the sliced instantaneous higher-order moments (S-IHOM) and address the use of the S-IHOM to represent non-stationary signals, substituted for the use of the instantaneous correlation function (ICK). Then we investigate the application of the S-IHOM in the selection and estimation of FH signals. The approach is based on the phase information of the S-IHOM. Results show the S-IHOM of FH signals is less sensitive to noise than the IFC of KH signals. The proposed scheme is robust to additive white noise degradations for SNR levels of 0 dB and above.

The application of regression algorithm in channel estimation (CE) with turbo codes over slow fading channel for DS-CDMA is presented in this paper. None of the classical methods like wiener filter, simple average (SA) and linear interpolation (LI) algorithms, utilizes channel coherent property sufficiently. Therefore, a low complexity improved method using low order robust regression algorithm for QPSK and 16QAM modulation is proposed. After initial channel estimation with classical methods, channel gains estimated are filtered by linear or conic regression algorithm within a given regression length. In this improved method, the length and order of regression algorithm are two key parameters, which affect the system performance significantly and the optimal values of which depend on the speed of mobile station. It is demonstrated by computer simulations that this improved method can track fading channel accurately and outperforms over classical methods substantially by selecting appropriate parameters of regression algorithm under a certain channel environment.

This paper introduced a system employing hybrid DS/FH spread spectrum (SS) coupled with RS forward error-control coding. By the flexible redefinition of frequency-occupation and frequency-collision event, the frequency-occupation probability of the DS/FH SS network was analyzed. This probability was based on the simultaneous transmission number threshold and discussed in both synchronous and asynchronous circumstances, respectively. The network throughput based on the packet correct reception probability was analyzed. Two models which had finite and infinite population respectively was discussed. Finally, simulation results were given.