Conventional training objectives for speech enhancement, such as mean squared error (MSE) in the time frequency domain and scale-invariant signal-to-distortion ratio (SI-SDR) in the waveform domain, exhibit limited correlation with human perception. Recent perceptual objectives better align with perceptual metrics, but often rely on auxiliary assessment networks or pretrained encoders, increasing training complexity and potentially causing instability. To address these limitations, this letter proposes a non-learned, closed-form objective, termed the multi-band relative contrastive loss (MBRCL). MBRCL enforces a relative margin in perceptually informed frequency bands to constrain the enhanced speech to be strictly closer to the clean reference than to the noisy input, and incorporates a lightweight temporal envelope gradient consistency term to preserve temporal dynamics. The loss is non-learned, almost everywhere differentiable, and requires only paired noisy–clean utterances for training. Experiments on BLSTM, CNN-U-Net, and GTCRN show gains over the matched fidelity baseline in terms of PESQ, STOI, SI-SDRi, and DNSMOS-OVRL; for the BLSTM setting, MBRCL also outperforms PESQ-inspired PMSQE, Quality-Net-based learned-loss training, and a recent frozen WavLM feature-space objective.

We propose PhysMVNet, a physics-inspired end-to-end framework for multichannel speech enhancement that integrates a learnable MVDR beamformer, a Helmholtz-inspired STFT-domain regularizer, and a residual spectral mapping module. The beamformer is trained with a reconstruction loss, while the regularizer encourages local smoothness in the STFT spectrogram to improve robustness to noise and array perturbations. To mitigate spectral distortions introduced by beamforming, we incorporate a three-band residual spectral mapping network to restore fine details. Experiments on CHiME-3/4 show that PhysMVNet achieves state-of-the-art perceptual quality and intelligibility while maintaining a lightweight design suitable for realtime application. It also remains stable under extreme low-SNR conditions and array perturbations. Ablation studies confirm the contribution of each component, highlighting the benefits of physics-inspired priors in deep beamforming networks for robust, high-fidelity speech enhancement.

In this paper, the performance of opportunistic user scheduling in reconfigurable intelligent surface (RIS)-aided multiuser communications is investigated by invoking 1-bit coding scheme. In particular, we derive new expressions for the outage probability (OP) and delay outage rate (DOR) by leveraging copula theory to model a two-fold interdependency stemming from the presence of discrete phase noise and a shared channel. We also provide simplified expressions for the OP and DOR using Clayton and Gumbel copulas. The newly derived expressions reveal the effects of different channel and RIS parameters simply. Finally, numerical results are provided to validate the derived exact closed-form expressions.

In this paper, we propose a novel multichannel speech enhancement method that is based on adaptive beamforming and spectral recovery. The core idea consists of dynamic spatial filtering (DSF) and residual spectral mapping (RSM). The DSF module utilizes depthwise separable convolutions to efficiently suppress directional noise and reverberation, while the residual spectral mapping (RSM) network refines the beam-formed output by restoring lost speech components. Experimental results on the CHiME 3/4 dataset demonstrate that the proposed model outperforms state-of-the-art methods in both speech quality and intelligibility while maintaining lower computational complexity.

This study investigates the anti-jamming space-air-ground integrated network (SAGIN) scenario wherein a reconfigurable intelligent surface (RIS) is deployed on a fixed Unmanned Aerial Vehicle (UAV) to counteract malevolent jamming attacks. In contrast to existing research, in this paper, we consider that a Low Earth Orbit (LEO) satellite is sending the signal to the user on the ground in the presence of jamming from a Geostationary Equatorial Orbit (GEO) satellite side. We aim to maximize the signal-to-jamming plus noise ratio (SJNR) by optimizing the RIS beamforming and transmit power of the LEO satellite. Assuming the availability of global channel state information (CSI) at the RIS, we propose alternating optimization (AO) and semidefinite relaxation (SDR) techniques to address the complexity. Simulation results show that the optimization schemes lead to considerable performance improvements. The results also indicate that, given the high jamming power and the relatively small number of RIS elements, deploying the RIS on UAVs near the user is more effective in mitigating the impact of jamming interferers.

In this paper, we propose a robust channel estimation technique for reconfigurable intelligent surface (RIS)-aided generalized frequency division multiplexing (GFDM) systems operating in high mobility scenarios. Leveraging the channel’s inherent sparsity in the delay-Doppler domain, the method employs a sparse Bayesian learning (SBL) framework constructed with a hierarchical Laplace prior. The expectation-maximization (EM) algorithm is then used to iteratively update the prior model parameters. Furthermore, to reduce computational complexity in the RIS phase optimization, only the strongest delay-Doppler channel path that maximizes the effective channel gain is selected. Simulation results confirm that the proposed approach achieves reliable performance with reduced pilot overhead.

This paper addresses the problem of resource allocation in aerial intelligent reflecting surface (AIRS)-assisted wireless networks, where intelligent reflecting surfaces (IRS) are deployed on flying platforms. Specifically, we investigate the joint optimization of AIRS placement and phase shifts, along with base station subcarrier and power allocation to maximize the system sum rate. To address the problem's non-convexity, an alternating optimization framework is proposed. In this framework, we employ K-means clustering to determine optimal AIRS locations and decompose the problem into two interdependent subproblems: AIRS phase optimization on the one hand and base station subcarrier and power allocation on the other hand, each solved iteratively until convergence is reached. In addition, the successive convex approximation method is used to tackle the non-convexity of AIRS phase shift optimization. Numerical results demonstrate that the proposed approach outperforms benchmark schemes, highlighting the potential of AIRS to extend cellular coverage in challenging scenarios such as emergencies or regions with limited infrastructure.

Unmanned aerial vehicles (UAVs) face escalating cybersecurity threats, particularly from GPS spoofing and jamming attacks, which endanger flight safety and mission integrity. This paper introduces a novel UAV cybersecurity framework that integrates statistical extreme value meta-learning (EVML) with a dual-path classifier to detect and classify GPS attacks. The method overcomes the key limitations of some existing approaches, such as excessive training data needs, zero-day threat vulnerability, and disjointed detection/classification. In particular, the proposed framework includes two stages. The first one enables zero-day attack detection through few-shot meta-learning with prototype-based anomaly detection where support sets contain only benign flight data while query sets include both benign samples and synthetically generated attack patterns. Moreover, a prototypical OpenMax layer and extreme value theory are exploited to identify suspicious patterns in GPS telemetry data. The second stage utilizes a novel dual-path architecture that independently processes position-related and signal-related features to classify detected attacks as spoofing or jamming. Our results on real-world attack data demonstrates that the proposed method has exceptional performance with 97.33 % detection accuracy and 0% false alarm rate, which significantly outperform the state-of-the-art. Furthermore, attack classification achieves 82.33 % accuracy for spoofing and 94.31 % for jamming attacks, with an overall F1 score of 0.88.

Human activity recognition (HAR) has attracted growing interest due to its wide ranging applications in healthcare, security, and smart surveillance. In particular, radar-based HAR offers a robust, non-contact alternative to conventional sensor based methods. Despite its promise, existing approaches often rely on substantial labeled data and exhibit limited adaptability in small-shot incremental learning scenarios. To address these challenges, we propose a novel few shot class incremental learning (FSCIL) framework for radar-based HAR. Our framework exploits vision transformers (ViTs) augmented with adaptive prompt mechanisms, hybrid attention blocks (HAB), and dynamic feature refinement strategies, which can substantially improve incremental learning performance. We validate the proposed approach on a fused radar dataset containing four sets of diverse radar datasets collected under varying environmental conditions. Our validation shows a harmonic accuracy (HAcc) up to 69.54 % and a Top-1 average accuracy (Top-1 Avg.) up to 91.20 %, outperforming state-of-the-art methods, with a session performance degradation as low as 10.80%. These results underscore the robustness of our approach in recognizing novel activities while preserving previously acquired knowledge.

In this paper, we address optimum design of uncrewed aerial vehicle (UAV)–based wireless networks with a focus on computation offloading in the presence of an active aerial attacker. Our design aims to minimize the maximum computation time among the tasks of ground users while satisfying the energy consumption requirements. To this end, we propose a joint optimization problem of partial computation offloading, ground user association, multiple UAVs trajectory control, computation resource, and sub-channel assignment. To tackle the underlying non-convex mixed-integer nonlinear optimization problem, we use the alternating optimization approach to iteratively solve the five sub-problems, namely, user-UAV association, user scheduling, partial offloading control and bit allocation over time slots, computation resource and sub-channel assignment, and UAV trajectory control until convergence. Moreover, the successive convex approximation method is employed to solve the non-convex sub-problems and improve the resilience of the system against jammer attacks. Additionally, we propose low-complexity algorithms to solve the involved sub-problems. Via extensive numerical studies, we illustrate the effectiveness of our proposed design compared to baselines under the impact of an aerial attacker.

This paper proposes a robust decentralized beam-forming (BF) scheme for downlink non-orthogonal multiple access (NOMA) transmission in an integrated satellite and aerial network (ISAN) to reduce both power consumption and signaling overhead. By employing the imperfect channel state information (CSI) and the imperfect successive interference cancellation (SIC), we formulate an optimization problem to minimize the total transmit power, subject to the rate requirements of both satellite and aerial terminals, and the transmit power budget of satellite and aerial platforms. To address this complex problem, we adopt S-procedure to transform the nonconvex constraints into convex ones and then propose a decentralized BF algorithm using Lagrange duality to obtain the satisfactory solutions in an efficient way. Finally, simulation results demonstrate that our proposed scheme can achieve a similar performance but at a lower signaling overhead as compared with the centralized BF method, and confirm the superiority of the proposed scheme in terms of power consumption over other existing works.

In this paper, we study unmanned aerial vehicle (UAV) trajectory control and channel assignment for UAV-based wireless networks with wireless backhauling. Our design aims to maximize the sum rate achieved by ground users while satisfying their data demand where spectrum reuse and co-channel interference management are considered. To tackle the underlying non-convex mixed-integer nonlinear optimization problem, we use the alternating optimization approach where we iteratively optimize the channel assignment and UAV trajectory control until convergence. Particularly, we propose an efficient heuristic algorithm to solve the channel assignment sub-problem. Moreover, the successive convex approximation (SCA) is used to solve the non-convex UAV trajectory control sub-problem. Via extensive numerical studies, we illustrate the effectiveness of our proposed design considering different network settings.

In this paper, we study the computation offloading problem against an active attacker in space-air-ground integrated networks (SAGIN), where joint optimization of partial computation offloading, unmanned aerial vehicle (UAV) trajectory control, computation and resource allocation is performed. Our design aims to minimize the maximum computation time of individual tasks among ground users while satisfying energy consumption constraints. To tackle the underlying non-convex optimization problem, we use the alternating optimization approach to iteratively solve three sub-problems, namely, partial offloading control and bit allocation over time slots, computation resource and bandwidth allocation, and UAV trajectory control, until convergence. Furthermore, the successive convex approximation method is employed to solve the non-convex sub-problems and improve the resilience of the SAGIN against active attacks. Via extensive numerical studies, we illustrate the effectiveness of our proposed design compared to baselines under the effect of an active attacker.

Reconfigurable intelligent surface (RIS) has received remarkable attention for its potential to improve the capacity and coverage of wireless communication networks. In this paper, we evaluate the performance of RIS-assisted communication systems in the presence of phase noise with the help of Student-T copula in two scenarios. The first one is the cascade link or RIS link and the second scenario involves direct link in conjunction with cascade link. In particular, we first analyze the probability density function and the cumulative distribution function of the signal-to-noise-ratio with/without direct link. Then, we investigate the outage probability and ergodic capacity of the RIS-assisted network by using Student- T copula function to characterize the non-linear dependency among the signal components. Furthermore, we reveal the relation between the Student- T copula dependency parameter and Pearson correlation coefficient. Finally, numerical results are presented confirm the validity of the analytical closed-form expressions.

In this work, we propose a new speech enhancement model referred to as auditory scene-attention model (ASAM), that can adapt dynamically to changes in the auditory scene components, such as speaker gender, input SNR levels, and background noise properties. To this end, a representative set of so-called Universal Scene Models (USM), each associated to a different auditory scene component, are first created, where each model attempts to predict a corresponding ideal ratio mask (IRM). The dynamic adaptation to changes in the auditory scene is then carried by computing the outputs of the USMs and forming a weighted combination of the most relevant scene models. This adaptation process is implemented via a frame-based attention mechanism, allowing to realize a soft selection of USM models, and taking advantage from both scene-dependent and scene-independent models. The evaluation of the proposed ASAM model, under different noise conditions and input SNR levels, shows substantial improvements in terms of standard speech Quality and intelligibility measures.

Unmanned aerial vehicles (UAVs) have improved the capacity and coverage of wireless networks. Mobile edge computing (MEC) has provided substantial computation capability to user equipment (UEs). The integration of UAV and MEC can take advantages of both to provide flexible computation service. In UAV assisted MEC networks, delay and energy consumption are two main concerns, which are conflicting to a certain extent. This paper investigates delay and energy consumption jointly in a multiple UAVs assisted MEC network. A cost function is defined to balance the delay and the energy consumption. The user access, task offloading, and computational resource allocation are jointly considered to minimize the long-term cost. To tackle this difficult problem, we formulate the long-term problem into sequential decision problem and treat all UEs as intelligent agents. Each UE decides its access UAV, task offloading proportion, and required edge computation resource to minimize the its own cost. Moreover, the optimal task offloading proportion and required computation resource can be obtained in closed-form given user access so that the action space can be significantly reduced. Then, an adversarial multi-armed bandit based algorithm is employed at each UE and a distributed scheme is proposed to solve the joint optimization problem. Simulation results validate the effectiveness and robustness of the distributed scheme and show its superiority to benchmarks.

This paper investigates a semi-grant-free (SGF) based transmission strategy to provide a flexible connectivity for various kinds of users in an integrated satellite-aerial-terrestrial network (ISATN). Herein, a high-altitude platform (HAP) termed as a grant-based user (GBU), which serves multiple mobile terminals (MTs) through space division multiple access (SDMA), wants to access a satellite network with multiple earth stations (ESs) termed as grant-free users (GFUs) simultaneously via non-orthogonal multiple access (NOMA) assisted SGF. To this end, we first propose two SGF-based uplink transmission schemes for both perfect channel state information (CSI) and imperfect CSI cases. When perfect CSI is available, a zero-forcing based beamforming (BF) scheme is used in HAP network while an adaptive transmit power allocation (ATPA) approach is adopted for SGF transmission. When only imperfect CSI is available, BF scheme employing the derived channel correlation matrix of HAP-MT link is proposed to achieve SDMA, and a novel ATPA strategy with rate probability constraint is proposed to guarantee quality-of-service of the GBU. Next, we derive the closed-form throughput expressions to evaluate the performance of the considered ISATN with the proposed two SGF-based schemes. Finally, computer simulations are conducted to validate the theoretical performance analysis and show the superiority of the proposed schemes over the related works. Moreover, our numerical results not only demonstrate a satisfactory performance of the proposed SGF-based scheme using imperfect CSI, but also reveal the impact of CSI errors on the system performance.

This paper aims at the proportional-integral (PI) control of the cone vibration of the electrodynamic loudspeakers system recently described using a linear fractional order model. After introducing the fractional order model of the circuit of these loudspeakers, firstly, a new method is developed to design a fractional order PI controller to place the poles of the system in a desired area of the complex plane which is called D-stabilizing. The design parameters of the method depend directly on the speed of the system output response, the cone vibration. Moreover, the offered fractional order controller avoids any non-minimum phase zero, which causes undesired undershoots in the output, for the closed-loop control system. Secondly, considering uncertainties in the coefficients of the model, a methodology is presented to determine up to how much the uncertainties can increase such that the controller is still able to maintain both D-stability and the absence of non-minimum phase zeros for the control system. Finally, the merit of the presented results and the superiority of the designed fractional order controller over its conventional integer order counterpart are illustrated through numerical simulations.

User-centric base station (BS) deployment has been designed for the fifth-generation (5G) dense millimeter wave (mmWave) networks for alleviating the inter-cell interference and improving the cell-edge user experience. However, the system power consumption increases sharply with the network density. In this paper, we investigate a user-centric simultaneous wireless information and power transfer (SWIPT) mmWave system employing a time-switching protocol at users to allow both energy harvesting (EH) and data decoding. To enable user-centric BS cooperation, adaptive BS clustering model is used to determine the user’s serving cluster based on its channel condition. Considering both linear and non-linear EH models, we analyze the joint coverage, namely, the probability that the user harvests enough energy in a given time slot and receives the required data from its serving cluster. The random serving clusters and the correlation between the amount of harvested energy and received data rate make the joint coverage analysis more challenging. A tractable tight approximation of the joint coverage probability is thus derived for ultra-dense networks. A mathematical optimization model for the time switching coefficient is also developed to maximize the system joint rate and energy coverage performance. All mathematical expressions are validated by Monte-Carlo simulations. Our results show that the proposed analytical framework is accurate and efficient for the design and deployment of SWIPT-enabled user-centric mmWave networks.

An integrated satellite and aerial network (ISAN) is considered a promising candidate to provide seamless connectivity for future wireless communication systems. In this paper, we propose a multi-objective based robust beamforming (BF) scheme for an ISAN to support heterogeneous services with high flexibility, where the satellite network serves various heterogeneous satellite terminals through multicast non-orthogonal multiple access (MC-NOMA), while the aerial network offers services to many internet of things devices using layered division multiplexing (LDM). Specifically, we first formulate a multi-objective optimization problem (MOOP) to achieve a good trade-off between sum rate maximization and total transmit power minimization. To tackle this mathematically intractable problem, we exploit the weighted Tchebycheff approach to transform the MOOP into a single-objective problem. Since only the angular information based channel state information is available, we exploit the angular discretization method and sequential convex approximation to design a robust BF algorithm to obtain the Pareto optimal solutions. Finally, simulation results demonstrated that our proposed scheme can achieve a optimal trade-off between multiple performance metrics with high spectrum and energy efficiency, so as to support heterogeneous services in the ISAN and fill the gap of only single type of serivce in the existing ISAN works.

This paper proposes a novel bi-directional non-orthogonal multiple access (NOMA)-based cooperative underlay satellite-terrestrial network (CUSTN), in which two secondary satellite users can serve as a potential relay for each user, thus gaining higher cooperative diversity order (DO). We analyze the exact and asymptotic outage probabilities (OPs) of both users under a multi-antenna setup and a practical residual hardware impairment (RHI) consideration. Numerical results are provided to validate the analysis, reveal the impacts of key parameters on the system performance, and demonstrate the advantages of our proposed scheme over other benchmarks.

Although many state-of-the-art approaches for improving the accuracy of Voice Activity Detection (VAD) have been proposed, their performance under adverse noise conditions with low Signal-to-Noise Ratio (SNR) remains limited. In this paper, we introduce a novel attention model-based deep neural network (DNN) architecture for VAD which takes advantage of complex Ideal Ratio Mask (cIRM). The proposed model, named AM-cIRM, consists of three sequential modules: extraction of cIRM features from the noisy speech using a DNN-based architecture; combination of cIRM with log-Mel spectrogram features along with temporal contextual extension; and VAD using an attention model that exploits the spectro-temporal information in the transformed features. Experimental results show that the proposed AM-cIRM achieves improved VAD performance when compared to state-of-the-art methods under different noise conditions.

Satellite and unmanned aerial vehicle (UAV) integrated networks (SUINs) are considered as a promising method to offer various Internet of Remote Things (IoRT) applications. In this article, we investigate the downlink transmission of SUINs where the satellite-to-UAV link uses the free-space optical (FSO) technology with an equal gain combining (EGC) scheme while the links from UAV to IoRT devices exploit radio frequency (RF) with the space-division multiple access (SDMA) technique. Specifically, considering that only statistical channel state information (CSI) is available, we first formulate an optimization problem to maximize the ergodic sum rate (ESR) of the system, which is constrained by the total transmit power budget and IoRT devices’ rate requirements. Then, a beamforming (BF) scheme based on the alternating direction method of multipliers (ADMM) is proposed to solve the nonconvex problem. Furthermore, a zero-forcing (ZF)-based suboptimal approach is also presented to reduce the implementation complexity. Finally, by assuming that the FSO link and RF links are subject to Gamma–Gamma fading and Nakagami-$m$fading, respectively, we derive closed-form ESR expressions for the considered network with the proposed BF schemes. Simulation results are provided to confirm the accuracy of the theoretical analysis. Moreover, it is revealed that our proposed EGC scheme for FSO communication and BF schemes for RF transmission can both achieve better performance than the existing works.

In this paper, we investigate the outage performance of an uplink (UL) non-orthogonal multiple access (NOMA)-based hybrid satellite-terrestrial relay network (HSTRN), in which two users communicate with the satellite through a decode-and-forward (DF) relay due to the lack of direct link. To provide a comprehensive yet hitherto unexplored outage analysis framework, we consider a more generalized channel model, i.e., the terrestrial and satellite links, respectively, undergo$\alpha -\mu $and$\kappa -\mu $shadowed fadings. Under fixed power allocation (FPA) for both multiple access phase and relaying phase, we firstly study three successive interference cancellation (SIC) decoding schemes, of which the first two are refined from existing schemes, while the third one, named asextended SIC (ESIC), is proposed in this paper to satisfy the quality of service (QoS) decoding criterion, which is shown to offer better performances for both users as compared to the former two SIC schemes. We also propose a novel dynamic power allocation (DPA) scheme for the multiple access phase, termed asenhanced DPA (EDPA), to overcome both users’ error floor (EF) issue yet provide better user fairness than the conventional DPA in the literature. We then analyze the exact and asymptotic outage performance for three SIC and theEDPAschemes under the generalized channel setting. It is shown that both the proposedESICandEDPAcan circumvent the EF issue and moreover, each has its own advantage in terms of the diversity order (DO). Our results also reveal that there exists a trade-off betweenESICandEDPA, since the former requires a premise on users’ targets rates to overcome the EF and once this premise is satisfied, no DO degradation will occur, while the latter does not entail such a premise but may face potential DO degradation. Finally, we present numerical results to verify the theoretical analysis, manifest the impacts of key parameters on the system performance, and demonstrate the advantages of our proposed network over other benchmarks.

In this paper, we investigate the impact of side information (SI) on the performance of physical layer security (PLS) under correlated fading channels. By considering non-causally known SI at the transmitter and exploiting the copula technique to describe the fading correlation, we derive closed-from expressions for the average secrecy capacity (ASC) and secrecy outage probability (SOP) under positive/negative dependence conditions. We indicate that considering such knowledge at the transmitter is beneficial for system performance and ensures reliable communication with higher rates, as it improves the SOP and brings higher values of the ASC.

This paper investigates a user-centric clustering model for a hybrid network comprising both millimeter-wave (mmWave) and terahertz (THz) base stations (BSs). Based on the proposed model, a user can choose to be cooperatively served by multiple mmWave or multiple THz BSs depending on their link quality. Besides, to maximize the cooperation gains, the serving clusters are dynamically adjusted to user's channel conditions pertaining to the different properties of the mmWave and THz networks. Finally, we evaluate the coverage probability of the hybrid network by using stochastic geometric tools and validate the analysis through numerical simulations.

In this paper, we propose a joint beamforming (BF) and power allocation scheme for non-orthogonal multiple access based multibeam satellite systems. Unlike the existing works where perfect channel state information (CSI) is required, we use imperfect CSI in formulating a constrained optimization problem aiming to minimize the maximum individual antenna powers subject to the outage constraints of quality-of-service requirements. Since the original problem is mathematically intractable, we first adopt Bernstein-Type II inequality to convert the outage constraints into deterministic forms. Then, an alternating optimization algorithm is proposed to jointly design BF vectors and power allocation coefficients. Finally, simulation results are provided to demonstrate the robustness and superiority of the proposed scheme compared with benchmark schemes.

Integrating non-orthogonal multiple access (NOMA) in intelligent reflecting surface (IRS) is expectedly an effective solution to enhance system's spectrum efficiency. In this paper, we investigate joint beamforming and power allocation for uplink NOMA transmission in an IRS-assisted cognitive satellite and terrestrial network operating at millimeter wave frequency band. Specifically, based only on imperfect channel state information in terms of the angular information of both primary users (PUs) and secondary users, we formulate an optimization problem to maximize the sum rate of the PUs in terrestrial network. To handle the resulting intractable optimization problem, we first transform the uncertainty channel vectors into a deterministic form with the aid of angular discretization. Then, by combining successive convex approximation with Taylor expansion and S-procedure methods, we propose an optimization scheme to jointly optimize the beamforming weight vector and power coefficients. Finally, simulation results show that the proposed scheme can achieve outstanding sum rate performance compared to state-of-the-art schemes.

In this paper, we propose a transformer-based architecture, called two-stage transformer neural network (TSTNN) for end-to-end speech denoising in the time domain. The proposed model is composed of an encoder, a two-stage transformer module (TSTM), a masking module and a decoder. The encoder maps input noisy speech into feature representation. The TSTM exploits four stacked two-stage transformer blocks to efficiently extract local and global information from the encoder output stage by stage. The masking module creates a mask which will be multiplied with the encoder output. Finally, the decoder uses the masked encoder feature to reconstruct the enhanced speech. Experimental results on the benchmark dataset show that the TSTNN outperforms most state-of-the-art models in time or frequency domain while having significantly lower model complexity.

This paper investigates the outage performance of a novel two-user non-orthogonal multiple access (NOMA)-based hybrid satellite-terrestrial relay network (HSTRN), where one user has direct link to the satellite (termed as direct-link user), while the other user needs to seek the help of the direct-link user or a dedicated relay to acquire its desired signal. An adaptive switch-and-stay combining relaying (ASSCR) scheme is proposed for boosting the system performance without excessive use of relaying resources of the network. Exact and asymptotic expressions of the outage probabilities (OPs) of both users are derived under the fixed-gain (FG) amplify-and-forward (AF) protocol. Finally, numerical results are presented to verify the theoretical analysis, manifest the impacts of key parameters on the system performance, and demonstrate the advantages of our proposed network over other benchmarks.

The user-centric base station cooperation is a new approach that allows a mobile user to be connected to a set (cluster) of base stations instead of being associated with a single one. This approach is highly valuable in millimeter wave networks where the base stations are expected to be densely deployed. In this paper, we evaluate the performance, in terms of coverage probability, of user-centric millimeter wave networks with dynamic clustering. First, we propose a dynamic clustering model for base stations that will cooperate to serve a given user. Then, based on the proposed model, we investigate analytically the coverage probability performance of the considered user-centric network using stochastic geometry tools. Finally, numerical and simulation results are provided, showing that the proposed dynamic clustering model always outperforms static clustering and single base station selection schemes for given network parameters.

Speech exhibits strong dependencies among its samples in both time and frequency domains. In this paper, we propose a low-complexity composite model for speech enhancement (SE) that integrates a convolutional neural network (CNN) and a long short-term memory (LSTM) network. These two modules take full advantage of the spectral and temporal information of input speech and extract in parallel a complementary set of features. The CNN is enabled to capture non-local spectral information via dilated frequency convolutions. It also incorporates an attention mechanism to recalibrate its weights without imposing considerable additional complexity. A grouping strategy is adopted for LSTM implementation to reduce its complexity while keeping performance almost unchanged. Our composite model is carefully designed to address concerns in real-time applications including limited computational resources, low-latency processing, and causal architecture. Through extensive and comparative simulation studies, it is shown that the proposed model significantly outperforms some other DNN-based SE methods in the recent literature.

In this paper, a novel multimodal convolutional neural network (CNN) fusion architecture is proposed for heart sound signal classification. Instead of using features from just one domain, general frequency features as well as Mel domain features are extracted from the raw heart sound. The multimodal CNN fusion architecture is individually trained based on the feature maps resulting from various feature extraction methods. These feature maps are then merged for optimizing the diversified extracted features. The proposed method provides an opportunity to explore the optimal selection of features for heart sound classification. Extensive experimentations are carried out, showing that an outstanding accuracy of 98.5% is achieved by the multimodal CNN architecture, which outperforms the other state-of-the-art approaches.

In this paper, we propose a transformer neural network based U-Net architecture, called context-aware U-Net (CAUNet), for end-to-end speech denoising in time domain. The proposed model adopts the dilated-dense block in both encoder and decoder layers of the U-Net to strengthen feature propagation and enlarge the receptive field of features. It also uses stacked two-stage transformer blocks to efficiently extract local and global contextual information from the encoder output, based on which the enhanced speech is reconstructed at the decoder. Experimental results show that our model outperforms most state-of-the-art methods in time and frequency domains, while it maintains a relatively low model complexity.

This paper presents a novel two-step method for channel estimation in millimeter wave hybrid MIMO systems over doubly (time and frequency) selective channels. In the first step, the dominant delay taps of the channel are selected by using an energy detector. In the second step, utilizing the basis expansion model (BEM) for effectively representing doubly selective channels, a BEM-based block orthogonal matching pursuit algorithm is proposed for estimating the gains of the selected delay taps. The proposed method benefits from exploiting the entire available training sequence to estimate all channel parameters, is capable of capturing channel variations across the whole training frames and does not require any feedback. Computer simulations and complexity evaluations show that the proposed approach can significantly improve the mean squared error performance compared with the existing techniques without increasing the computational complexity.

This paper proposes a novel joint hybrid precoder and combiner design for maximizing the average achievable sum-rate of single-user OFDM millimeter wave massive MIMO systems. The analog precoder and combiner design is formulated as a constrained Tucker2 decomposition and solved using the projected alternate least square method. This formulation allows maximizing the sum of the effective baseband gains over every subcarrier while suppressing the interference among the same subcarrier’s data streams. The digital precoder and combiner are obtained from the effective baseband channel’s singular value decomposition on a per-subcarrier basis. Numerical simulation results show that the proposed method outperforms some of the existing designs while having similar computational complexity.

To satisfy the explosive access demands of Internet-of-Things (IoT) devices, various kinds of multiple access techniques have received much attention. In this article, we investigate the multicast communication of a satellite and aerial-integrated network (SAIN) with rate-splitting multiple access (RSMA), where both satellite and unmanned aerial vehicle (UAV) components are controlled by network management center and operate in the same frequency band. Considering a content delivery scenario, the UAV subnetwork adopts the RSMA to support massive access of IoT devices (IoTDs) and achieve desired performances of interference suppression, spectral efficiency, and hardware complexity. We first formulate an optimization problem to maximize the sum rate of the considered system subject to the signal-interference-plus-noise-ratio requirements of IoTDs and per-antenna power constraints at the UAV and satellite. To solve this nonconvex optimization problem, we exploit the sequential convex approximation and the first-order Taylor expansion to convert the original optimization problem into a solvable one with the rank-one constraint, and then propose an iterative penalty function-based algorithm to solve it. Finally, simulation results verify that the proposed method can effectively suppress the mutual interference and improve the system sum rate compared to the benchmark schemes.

Satellite-based Internet of Things (IoT) is recognized as a cost-effective approach for global access. In this article, we aim at improving the spectrum efficiency of satellite systems to serve a huge number of IoT devices. To this end, we present a cognitive satellite-terrestrial framework, where a multibeam satellite system with full frequency reuse shares the spectrum with terrestrial networks based on the underlay paradigm. Considering DVB-S2X recommendations, geometric configurations, and channel characteristics, we investigate a robust multigroup multicast beamforming design for the satellite-based IoT coexisting with terrestrial networks in the presence of a phase error on channel state information, and characterize the achievable rate region under the outage probability constraint for the terminal and the power consumption constraint for the satellite. Based on the concept of rate profile, an associated optimization problem is formulated to design robust beamformers and determine the Pareto boundary of the region. To solve the intractable problem, we propose a two-level iterative algorithm on the basis of joint bisection search and penalty function enabled nonsmooth optimization. In particular, we develop a Bernstein-type inequality aided method and a large deviation inequality aided method to obtain a tractable and conservative approximation for the probabilistic constraint, respectively. Numerical results are provided to confirm the validity and superiority of our proposed scheme over the existing approaches and reveal the impact of key parameters on the achievable system performance.

High-throughput satellites (HTSs) play an important role in future millimeter-wave (mmWave) aeronautical communication to meet high speed and broad bandwidth requirements. This paper investigates the outage performance of an aeronautical broadband satellite communication system’s forward link, where the feeder link from the gateway to the HTS uses free-space optical (FSO) transmission and the user link from the HTS to aircraft operates at the mmWave band. In the user link, spot beam technology is exploited at the HTS and a massive antenna array is deployed at the aircraft. We first present a location-based beamforming (BF) scheme to maximize the expected output signal-to-noise ratio (SNR) of the forward link with the amplify-and-forward (AF) protocol, which turns out to be a phased array. Then, by supposing that the FSO feeder link follows Gamma-Gamma fading whereas the mmWave user link experiences shadowed Rician fading, we take the influence of the phase error into account, and derive the closed-form expression of the outage probability (OP) for the considered system. To gain further insight, a simple asymptotic OP expression at a high SNR is provided to show the diversity order and coding gain. Finally, numerical simulations are conducted to confirm the validity of the theoretical analysis and reveal the effects of phase errors on the system outage performance.

This paper investigates the massive access for a satellite-aerial-terrestrial network (SATN), where a high-altitude platform (HAP) is deployed as a relay to assist the uplink transmission from terrestrial user equipment (UE) to satellite. Unlike previous works, we adopt radio frequency (RF) and free space optical for the aerial-terrestrial and satellite-aerial links, respectively. Specifically, by assuming that imperfect angular information (IAI) of each UE is known at the HAP, we develop a space division multiple access (SDMA) scheme to maximize the ergodic sum rate (ESR). To this end, we first exploit the IAI to calculate the analytical expression of channel correlation matrix. Then, by considering the limitation of array freedom, we propose a subspace-based UE grouping and scheduling scheme to cluster all UEs into groups. Next, we present a computationally effective beamforming (BF) scheme for each UE at HAP to efficiently implement SDMA in the RF link. Furthermore, a closed-form expression for the ESR of the SATN is derived to validate the proposed BF and SDMA schemes. Finally, simulation results corroborate the derived theoretical formulas and reveal the impacts of array size, angular estimation error, the number of UEs and scheduling threshold on the system performance.

In millimeter wave (mmWave) networks, base stations (BSs) are expected to be densely deployed in order to meet the demands of mobile users. A major challenge in dense mmWave networks is the interference experienced by the user from the neighboring BSs which limits the density of deployed BSs. A promising solution to this challenge is to adopt user-centric BS cooperation that allows a user to be connected to a set (cluster) of BSs instead of being associated with a single one. In this paper, we investigate the performance of user-centric mmWave networks with dynamic clustering. First, we propose a dynamic clustering model to enable the user-centric BS cooperation. Then, based on the proposed model, we analyze the coverage probability and average spectral efficiency (ASE) performances using stochastic geometry tools. We also propose a BS clustering optimization framework to achieve maximum ASE performance for given network configuration. Finally, numerical and simulation results are provided, showing that the proposed dynamic clustering schemes always outperform static clustering and single-BS selection schemes and yield an optimum system performance for given network parameters.

This paper investigates the throughput improvement of cell-free massive multiple-input multiple-output (MIMO) systems by non-orthogonal multiple access (NOMA) for future cellular networks under stochastic access point and user locations. In this context, the node locations are modeled with Poisson point processes. The time division duplexing mode is employed, and uplink channels are estimated locally using uplink pilots. Furthermore, unique pilot sequences are used between NOMA clusters, while pilot reuse occurs within each cluster to strike a balance between the training overhead and the number of clusters. Matched-filter-based precoding is utilized for downlink transmission. The aggregate received signal is analytically characterized by deriving the moment generating function and approximations via moment matching. Then, the asymptotic achievable rates of the NOMA users are derived, thereby quantifying the adverse impact of error propagation owing to imperfect successive interference cancellation. Special scenarios with prior downlink channel state information and log-distance power control are also considered. We show that NOMA greatly increases the achievable average rate, especially under low path loss exponents and dense networks, while user fairness may be boosted by the adoption of a log-distance transmit power control scheme with proper parameter selection (i.e. lower values for the power control parameter).

In this paper, we investigate secrecy-energy efficient hybrid beamforming (BF) schemes for a satellite-terrestrial integrated network, wherein a multibeam satellite system shares the millimeter wave spectrum with a cellular system. Under the assumption of imperfect angles of departure for the wiretap channels, the hybrid beamformer at the base station and digital beamformers at the satellite are jointly designed to maximize the achievable secrecy-energy efficiency, while satisfying signal-to-interference-plus-noise ratio constraints of both the earth stations (ESs) and cellular users. Since the formulated optimization problem is nonconvex and mathematically intractable, we propose two robust BF schemes to obtain approximate solutions with low complexity. Specifically, for the case of a single ES, we integrate the Charnes-Cooper approach with an iterative search algorithm to convert the original nonconvex problem into a solvable one and obtain the BF weight vectors. In the case of multiple ESs, by exploiting the sequential convex approximation method, we convert the original problem into a linear one with multiple matrix inequalities and second-order cone constraints, for which we obtain a solution with satisfactory performance. The effectiveness and superiority of the proposed robust BF design schemes are validated via simulations using realistic satellite and terrestrial downlink channel models.

The next generations of cellular wireless systems promise an improved user experience with respect to the throughput, reliability, latency, and connectivity. To this end, multiple-input multiple-output (MIMO) and massive MIMO systems hold enormous potential. Moreover, due to heterogeneity and densification, user and base station locations are increasingly random. Thus, in this paper, we characterize the performance of a typical downlink user within a massive MIMO Voronoi cell when the base stations employ matched filter based precoding in the downlink. We consider a typical Voronoi cell where the base stations follow a Poisson point process (PPP), and also consider the users to be randomly distributed according to a PPP independent from the base stations. Furthermore, a wireless channel with log-distance path loss and Rayleigh fading is assumed. Using a novel framework to model the signal to noise ratio, the outage probability of a user is derived in closed-form while taking into account full and partial loading for the cell's base station due to the randomness of user numbers. Numerical results show that the outage probability is heavily dependent on the base station density, and that the performance decreases when the maximum number of users served by a base station increases.

Simultaneous wireless information and power transfer (SWIPT) is increasingly seen as a viable technique to power energy constrained user devices while transmitting data. SWIPT can be readily implemented within the recently proposed cell-free massive MIMO networks, where a large number of interconnected access points (APs) simultaneously serve users. This paper characterizes the coexistence and interplay between these two technologies by using tools from stochastic geometry. To this end, we consider a spatially random network, where the APs are modeled stochastically using a Poisson point process, and a time-switching protocol is used for the SWIPT operation at the users. A time-division-duplexing protocol is considered in which uplink pilots are used to obtain channel state information at the APs, while conjugate beamforming is performed in the downlink. Moreover, we consider blockages due to obstacles in the channel and the resulting line-of-sight and non-line-of-sight conditions affecting the fading and path loss. We derive the mean and variance of the harvested energy along with the average achievable rate in the downlink for an energy user. The tradeoff between the downlink data throughput and harvested energy is quantified, and thereby, we show that spatially-distributed APs in a cell-free arrangement can boost the energy-rate trade-off of SWIPT.

This paper investigates the uplink space division multiple access in a satellite-aerial-terrestrial network, where multiple users access to satellite with the aid of a unmanned aerial vehicle, and the two hops adopt radio frequency (RF) and free space optical (FSO), respectively. We first formulate an optimization problem to maximize the ergodic sum rate (ESR) of the considered system. Then a statistical channel state information (CSI)-based beamforming (BF) scheme is proposed to solve the non-convex ESR maximization problem. The main advantage of the proposed method is that only statistical CSI is used to obtain BF vectors so that a low implementation complexity can be achieved. Then, by assuming that the RF and FSO links undergo Nakagami-m and Gamma-Gamma fading, respectively, closed-form expression for ESR of the considered system is derived. Simulation results validate the derived expression and confirm the effectiveness of the proposed BF scheme.

In this paper, we propose a novel robust downlink beamforming (BF) design for satellite-terrestrial integrated networks. Under a realistic assumption that the angular information of eavesdroppers is not perfectly known, we establish an optimization framework for hybrid BF at the terrestrial base station and digital BF at the satellite to maximize the secrecy-energy efficiency of the system, while satisfying the quality-of-service constraints of both earth station and cellular user. Since the formulated optimization problem is mathematically intractable, we present an iterative algorithm based on the Charnes-Cooper approach to optimize the BF weight vectors. The effectiveness and superiority of the proposed robust hybrid BF scheme are validated via computer simulations.

Most localization methods for mixed far-field (FF) and near-field (NF) sources are based on uniform linear array (ULA) rather than sparse linear array (SLA). In this paper, we propose a localization method for mixed FF and NF sources based on the generalized symmetric linear arrays, which include ULAs, Cantor array, Fractal array and many other SLAs. Our method consists of two steps. In the first step, the high-order statistics of the array output is exploited to increase the degree of freedom. Then the direction-of-arrivals (DOAs) of the FF and NF sources are jointly estimated by using the recently proposed atomic norm minimization (ANM), which belongs to the gridless super-resolution method since the discretization of the parameter space is not required. In the second step, the ranges are given by MUSIC-like one-dimensional searching. Simulations results are provided to demonstrate the advantages of our method.

With the help of a power-domain non-orthogonal multiple access (NOMA) scheme, satellite networks can simultaneously serve multiple users within limited time/spectrum resource block. However, the existence of channel estimation errors inevitably degrade the judgment on users' channel state information (CSI) accuracy, thus affecting the user pairing processing and suppressing the superiority of the NOMA scheme. Inspired by the advantages of machine learning (ML) algorithms, we propose an improved support vector machine (SVM) scheme to reduce the inappropriate user pairing risks and enhance the performance of NOMA based satellite networks with imperfect CSI. Particularly, a genetic algorithm (GA) is employed to optimize the regularization and kernel parameters of the SVM, which effectively improves the classification accuracy of the proposed scheme. Simulations are provided to demonstrate that the performance of the proposed method is better than that with random user paring strategy, especially in the scenario with a large number of users.

In this paper, we present a deep neural network (DNN) based algorithm to restore the high-frequency (HF) component of the enhanced speech processed by Kalman filtering, where the DNN is applied for estimating the magnitude of HF component from the low-frequency (LF) counterpart. The complete HF component is then computed with the estimated magnitude given by the DNN and the phase of the Kalman filtered speech. By incorporating our restoration algorithm into Kalman filter based speech enhancement method, our new speech enhancement system is able to recover the HF component with better perceptual quality and less distortion. Experimental results demonstrate that the proposed method outperforms the state-of-the-art Kalman filter based method in terms of both speech quality and intelligibility.

In this paper, the outage performance of nonorthogonal multiple access (NOMA) based coordinated direct and relay transmission (CDRT) system is analyzed. Exact and asymptotic outage probabilities of both cell-center user (CCU) and celledge user (CEU) are derived. Our analysis is conducted based on the independent but not identically distributed (i.n.i.d) Nakagami-m fading model and the fixed-gain amplify-and-forward relaying protocol, and hence the result is applicable to general relay-aided communications, where only statistical channel state information is available. Monte Carlo simulation results are provided to verify the accuracy of the derived analytical expressions with comparison to conventional orthogonal multiple access (OMA) counterpart as well as the decode-and-forward NOMA-based CDRT, demonstrating the superiority of the proposed scheme.

In cellular networks that exploit the millimeter wave bands, access points are expected to be densely deployed and to share limited radio resources. In such systems, the geometric distribution of the downlink signal-to-interference-plus-noise ratio (SINR) is affected by several random variables such as path loss, blockage, small scale channel fading, interferer density, antenna array orientation, and thermal noise. Therefore, using fixed modulation in such varying environments is inefficient as it degrades the spectral efficiency and/or increases the outage probability. To tackle this issue, the transmitter needs to adapt the modulation order to the link condition. In this paper, we introduce an adaptive modulation technique, which depends on the geometric distribution of the SINR, for millimeter wave cellular systems in order to enhance their performances in terms of average spectral efficiency. In this case, the transmitter adjusts the modulation type of the transmitted signal based on the receiver geometry and the link condition. First, we present an analytical model using geometry tools to compute the statistical distributions of the downlink SINR. Then, the performance of the adaptive modulation scheme is studied and evaluated in terms of the average spectral efficiency. Numerical results show that, while keeping the outage probability as minimum as possible, this geometry-based adaptive modulation can efficiently enhance the millimeter wave system average spectral efficiency.

We propose a joint hybrid precoder and combiner design for narrowband single-user massive MIMO systems. The analog precoder and combiner are computed by an iterative algorithm based on the constrained singular value decomposition (CSVD), which guarantees the constant modulus constraints of the phase-shifters in the analog beamforming are attained. The digital precoder is obtained from the SVD of the reduced-dimension effective channel, after taking into account the effect of the analog beamforming. Numerical simulation results show that the proposed HBF design outperforms several designs existing in the literature in terms of the achievable sum-rate, even when only imperfect channel state information (CSI) is available.

Simultaneous wireless information and power transfer (SWIPT) is a promising solution for future wireless networks as it provides convenient and perpetual energy supplies to wireless users. This study proposes three new transceiver design schemes in two‐way relay networks with SWIPT. Firstly, a transmit power minimisation solution is proposed with an aim to minimise the transmit power while meeting an energy harvesting requirement. Then, a harvested energy maximisation solution that has the same complexity as the first solution but provides a much better sum harvested energy is proposed. Finally, a signal‐to‐interference‐plus‐noise ratio maximisation solution is developed, which gives the best sum‐rate performance for the two‐way relay networks while ensuring a desired level of energy harvesting. Furthermore, the complexity of the three new transceiver solutions is discussed. Simulation results have demonstrated the effectiveness of the proposed transceiver design schemes in two‐way relay networks with SWIPT.

In this work, a novel learning‐based approach for the design of a compressed sensing measurement matrix is proposed. In contrast with the state‐of‐the‐art methods, the suggested approach takes into account the correlation within entries of each column of the measurement matrix, namely, the inter‐column correlation (ICC). The new method makes use of a rather small number of training sparse signal vectors in a recursive scheme to obtain their corresponding measurement vectors. The latter is exploited to estimate the copula function of measurements which, in turn, is used to generate an arbitrary number of measurement vector ensembles. By using the latter, the autocorrelation of the measurement vectors is estimated precisely and then, the ICC of measurement matrix under design is obtained from the autocorrelation. Given the resulting ICC, the measurement matrix columns are to be generated independently, e.g. by employing a proper random Gaussian vector generator. Performance evaluations using both synthetic and real‐world data confirm the superiority of the proposed approach to the less recent methods.

Grayscale-thermal tracking has attracted a great deal of attention due to its capability of fusing two different yet complementary target observations. Existing methods often consider extracting the discriminative target information and exploring the target correlation among different images as two separate issues, ignoring their interdependence. This may cause tracking drifts in challenging video pairs. This paper presents a collaborative encoding model called joint correlation and discriminant analysis based inver-sparse representation (JCDA-InvSR) to jointly encode the target candidates in the grayscale and thermal video sequences. In particular, we develop a multi-objective programming to integrate the feature selection and the multi-view correlation analysis into a unified optimization problem in JCDA-InvSR, which can simultaneously highlight the special characters of the grayscale and thermal targets through alternately optimizing two aspects: the target discrimination within a given image and the target correlation across different images. For robust grayscale-thermal tracking, we also incorporate the prior knowledge of target candidate codes into the SVM based target classifier to overcome the overfitting caused by limited training labels. Extensive experiments on GTOT and RGBT234 datasets illustrate the promising performance of our tracking framework.

Non-orthogonal multiple access (NOMA) and millimeter wave communications are key technologies for the fifth-generation of cellular networks and beyond, and the coexistence of these two techniques is critical. This paper characterizes the system performance through the outage probability and achievable downlink rate for large-scale millimeter wave NOMA networks by using stochastic geometry. In order to reflect spatial randomness, we consider homogeneous Poisson point processes to model the base stations and user equipments. Moreover, blockages which affect the channel characteristics, power allocation based on the combined channel gain, and imperfections in the successive interference cancellation are considered. The aggregate co-channel interference at a user is characterized based on the moment generating function. Finally, the outage probability and downlink rate are derived for a two-user NOMA scenario under two user-base station association schemes: 1) closest base station association and 2) closest line-of-sight base station association. It is seen that using NOMA under millimeter wave channels increases the achievable downlink rate while keeping the performance impact on individual users low, and that the closest line-of-sight base station association scheme is comparatively advantageous. Moreover, a dense base station deployment generally improves the performance further.

In this paper we propose a fully convolutional neural network (CNN) for complex spectrogram processing in speech enhancement. The proposed CNN consists of one-dimensional (1-d) convolution and frequency-dilated 2-d convolution, and incorporates a residual learning and skip-connection structure. Compared with the state-of-the-art, the proposed CNN achieves a better performance with fewer parameters. Experiments have shown that the complex spectrogram processing is effective in terms of phase estimation, which benefits the reconstruction of clean speech especially in the female speech case. It is also demonstrated that the model yields a convincing performance with small memory footprint when the number of parameters is limited.

In this paper, we investigate the outage performance of a multiuser threshold-based decode-and-forward integrated satellite-terrestrial relay network (ISTRN) with opportunistic user scheduling, where the satellite link and terrestrial links undergo Shadowed-Rician (SR) fading and correlated Rayleigh fading, respectively. First, we propose a new probability density function (PDF) to statistically characterize the square sum of independent and identically distributed SR random variables, which is more accurate and concise than existing expressions. Next, based on the new PDF, we derive an analytical expression of outage probability (OP) of the considered network. Furthermore, the asymptotic OP expression is developed at high signal-to-noise ratio to reveal the achievable diversity order of the considered ISTRN. Finally, computer simulation is conducted to validate the analytical results, and show the impact of various parameters on the outage performance.

Cell-free massive multiple-input multiple-output (MIMO) is a form of distributed massive MIMO aiming to provide massive access and improve spectral efficiency by inheriting favorable properties of traditional massive MIMO, while mitigating detrimental effects such as shadowing and spatially correlated fading. This paper investigates how the throughput of cell-free massive MIMO is affected by non-orthogonal multiple access (NOMA) for the next-generation cellular networks under stochastic access point (AP) and user locations. Thus, we consider homogeneous Poisson point processes (PPPs) to model node locations, while considering a Rayleigh channel with log-distance path loss. The time division duplexing (TDD) mode is employed and uplink channels are estimated autonomously/locally at each AP via uplink pilots sent by users. Moreover, while unique pilots are used between NOMA clusters, pilots are reused within each cluster in order to strike a balance between the training overhead and number of clusters. Matched filter based precoding is performed within the downlink based on the estimated channels. The aggregate signal received from all access points is characterized based on the moment generating function and approximated via moment matching. Thereby, the achievable rates for the users are derived, under the consideration of error propagation due to imperfect successive interference cancellation (SIC). It is shown that NOMA increases the overall rate under environments with low path loss exponents and networks with high access point densities, while careful power allocation can significantly improve user fairness.

In this paper, we present a novel deep neural network (DNN) based Kalman filter (KF) algorithm for speech enhancement, where DNN is applied for estimating key parameters in the KF, namely, the linear prediction coefficients (LPCs). By training the DNN with a large database and making use of the powerful learning ability of DNN, our proposed DNN-KF algorithm is able to estimate LPCs from noisy speech more accurately and robustly, leading to an improved performance as compared to traditional KF based approaches in speech enhancement. Experimental results demonstrate that our DNN-KF method outperforms two existing KF based speech enhancement methods in terms of both speech quality and intelligibility.

In this study, the authors present and investigate a novel cooperative relaying scheme for cognitive radio networks (CRNs), which is based on non‐orthogonal multiple access (NOMA). In the proposed scheme, following the detection of an idle channel, the secondary base station transmits a power domain NOMA signal to a first nearby secondary user (SU). In addition to decoding its own signal, this user applies a decode‐and‐forward strategy to relay the signal intended to a second SU. In contrast to previous works, where the spectrum sensing and transmission phases are treated separately, the authors here consider both phases jointly in the design and analysis of the proposed scheme. To characterise performance of the latter, analytical expressions are derived for the outage probability and the ergodic rate of the two SUs by assuming a flat Rayleigh fading channel model. The performance of two traditional orthogonal multiple access schemes is also analysed for comparison. Simulation and numerical results are presented to demonstrate the effectiveness of the proposed cooperative relaying scheme for CRN, as well as the accuracy of the analytical results.

This paper investigates the performance of a dual-hop decode-and-forward (DF) relaying-aided land mobile satellite communication over Shadowed-Rician (SR) fading channels. A practical model for the satellite relaying system is first developed, where the impacts of satellite multi-beam antenna, radio propagation loss, and random shadowing are taken into account. Next, by assuming that the multi-beam satellite suffers from hardware impairments (HIs) and is perturbed by interference signals, we derive an equivalent end-to-end signal-to-interference-plus-noise-and-distortion-ratio of the system, and justify that the maximum ratio transmission at the source and the maximum ratio combining at the destination are the optimal transmit-receive beamforming schemes on the proposed HIs model. Then, closed-form expressions for the probability density function (PDF) of the sum of independent and identically distributed (i.i.d) squared SR random variables in the case of integer and rational Nakagami-m fading parameters are derived. Based on the derived PDF, new analytical expressions for the outage probability (OP) and average throughput are obtained in the presence of HIs and interference. Moreover, the asymptotic OP and average throughput at high signal-to-noise ratio are investigated to reveal the achievable diversity order of the system. Finally, Monte Carlo simulation results are provided to corroborate the analytical results.

Massive multiple-input multiple-output (MIMO) systems operating within the millimeter wave frequency range offer exciting opportunities for the future fifth-generation (5G) wireless networks. While the increased bandwidth and spectral efficiency are attractive, transitioning to millimeter massive MIMO presents significant challenges with respect to blockages, high attenuation, and channel estimation errors. To address these challenges, this paper evaluates the outage performance of a millimeter wave cellular network using massive MIMO under a stochastic set-up subject to pilot contamination and matched-filter precoding. We model the cellular users and base stations with Poisson point processes. Furthermore, we consider blockages from random objects, and employ different path loss and fading models for the line-of-sight (LOS) and non-line-of-sight scenarios. Moreover, both fixed power transmissions and path loss inversion-based power control are considered along with the sectored antenna patterns. Using stochastic geometry, we derive the moment generating function of the interference experienced by a typical cellular user and its outage probability. It is observed that the environments with different path loss exponents have varying behaviour for similar blockage sizes and densities. In addition, the ratio of the number of cellular users to that of base stations, and the antenna beamwidth are critical parameters affecting the outage performance.

Sleep apnea, a serious sleep disorder affecting a large population, causes disruptions in breathing during sleep. In this paper, an automatic apnea detection scheme is proposed using single lead electroencephalography (EEG) signal to discriminate apnea patients and healthy subjects as well as to deal with the difficult task of classifying apnea and nonapnea events of an apnea patient. A unique multiband subframe based feature extraction scheme is developed to capture the feature variation pattern within a frame of EEG data, which is shown to exhibit significantly different characteristics in apnea and nonapnea frames. Such within-frame feature variation can be better represented by some statistical measures and characteristic probability density functions. It is found that use of Rician model parameters along with some statistical measures can offer very robust feature qualities in terms of standard performance criteria, such as Bhattacharyya distance and geometric separability index. For the purpose of classification, proposed features are used in K Nearest Neighbor classifier. From extensive experimentations and analysis on three different publicly available databases it is found that the proposed method offers superior classification performance in terms of sensitivity, specificity, and accuracy.

Massive multiple-input multiple-output (MIMO) enabled base stations employing non-orthogonal multiple access (NOMA) hold immense potential in increasing the spectral efficiency of future cellular networks. In this paper, we evaluate the achievable rate of N-user NOMA under a Poisson process of massive MIMO enabled base stations. We adopt a timedivision duplexing (TDD) mode, where the uplink pilots are reused among the different base stations, and within each NOMA cluster of each base station, while matched-filter based precoding is employed in the downlink. The achievable rate by a typical NOMA user is characterized by taking into account imperfect successive interference cancellation (SIC) and error propagation. To this end, the moment generating function of the interference from other base stations due to pilot contamination is derived along with the signal detection probability. It is shown that NOMA can significantly improve the rate performance under most system parameters, and that the rate performance can be increased further through denser networks. Moreover, we show that the individual user rates and fairness amongst users within a NOMA cluster are significantly impacted by the specific power allocation algorithm.

Non-orthogonal multiple access (NOMA) and millimeter wave communications are key technologies for the fifth generation (5G) of cellular networks in order to increase the spectral efficiency, ensure massive connectivity, and to provide large bandwidths to mobile internet services. However, the coexistence of these two techniques is critical, especially under increasingly random dense base station assignments. To this end, this paper characterizes the system coverage for users utilizing NOMA in a stochastic multi cell set-up under millimeter wave frequencies. We employ the Poisson point process (PPP) to model the base stations and user devices. Moreover, our system model considers the power allocation, imperfections in the successive interference cancellation (SIC), and blockages that affect the channel path loss and fading characteristics. The aggregate co-channel interference experienced at a user is characterized based on the moment generating function, and the outage probability is derived for a two user NOMA scenario. Analytical and numerical results show that NOMA is a feasible candidate technique to increase the spectral efficiency under millimeter wave channels if careful power allocation is implemented, with limited performance penalties for each user. It is also shown that the performance is further improved by utilizing a dense base station deployment.

In this paper, we investigate the secure communication of a satellite-terrestrial integrated network (STIN). By supposing that the satellite employs multi-beam antenna while the base station (BS) is equipped with a uniform planar array (UPA), we first formulate a joint constrained optimization problem to maximize the sum rate of STIN while satisfying both the quality-of- service (QoS) requirement of the information receivers and earth stations (ESs), the energy harvest (EH) requirement of the energy receivers (ERs), the secrecy constraint at ERs. Since the formulated optimization problem is non-convex and mathematically intractable, we then propose a joint beamforming (BF) scheme to obtain the optimal solutions through an iterative algorithm, which exploits the sequential convex approximation (SCA) and Taylor expansion to convert the original non-convex problem into a solvable one. Finally, simulation results are given to demonstrate the effectiveness of the proposed joint BF schemes.

Channel estimation is essential for precoding/combining in millimeter wave (mmWave) communications. However, accurate estimation is usually difficult because the receiver can only observe the low-dimensional projection of the received signals due to the hybrid architecture. We take the high speed scenario into consideration where the Doppler effect caused by fast-moving users can seriously deteriorate the channel estimation accuracy. In this paper, we propose to incorporate the nested array into analog array architecture by using RF switch networks with an objective of reducing the complexity and power consumption of the system. Based on the covariance fitting criterion, a joint Doppler and channel estimation method is proposed without need of discretizing the angle space, and thus the model mismatch effect can be totally eliminated. We also present an algorithmic implementation by solving the dual problem of the original one in order to reduce the computational complexity. Numerical simulations are provided to demonstrate the effectiveness and superiority of our proposed method.

In this paper, a novel noncircularity-based localization method for mixed near-field (NF) and far-field (FF) sources is proposed with a symmetric uniform linear array (ULA) in the presence of unknown mutual coupling (UMC). Based on the principle of rank reduction (RARE), the multiple parameters of the sources including direction of arrival (DOA), range and mutual coupling coefficient (MCC) are decoupled, so that only several one-dimensional (1-D) spectral searches are required for their estimation. Meanwhile, the proposed method can also distinguish the types of sources without any extra processing. Simulation results are provided to demonstrate the effectiveness of the proposed method for the classification and localization of mixed sources under UMC.

In this paper, a new approach for the design of measurement matrix, Φ, for compressive sensing (CS) in a generic context is proposed. In accordance with well-known classical CS theory, we take the elements of Φ to be random, yet, we include correlations within the elements of the individual columns of Φ. To this end, a new structure for Φ is proposed where the correlations of interest are controlled by a selectable parameter. We aim at optimizing the proposed Φ with respect to the latter correlation parameter by leveraging an appropriate criterion in a learning-based framework. We evaluate the performance of the proposed Φ and compare it with the state-of-the-art literature including random Φ with independent and identically distributed (i.i.d.) elements. Performance advantage of the proposed approach is validated in different CS scenarios.

The atomic norm minimization (ANM) has been successfully incorporated into the two-dimensional (2-D) direction-of-arrival (DOA) estimation problem for super-resolution. However, its computational workload might be unaffordable when the number of snapshots is large. In this paper, we propose two gridless methods for 2-D DOA estimation with L-shaped array based on the atomic norm to improve the computational efficiency. Firstly, by exploiting the cross-covariance matrix an ANM-based model has been proposed. We then prove that this model can be efficiently solved as a semi-definite programming (SDP). Secondly, a modified model has been presented to improve the estimation accuracy. It is shown that our proposed methods can be applied to both uniform and sparse L-shaped arrays and do not require any knowledge of the number of sources. Furthermore, since our methods greatly reduce the model size as compared to the conventional ANM method, and thus are much more efficient. Simulations results are provided to demonstrate the advantage of our methods.

Conventional reconstruction algorithms of compressed sensing suffer from serious performance degradation in impulsive noise environment. In this paper, we propose a modified Lorentzian greedy iterative algorithm to further improve the quality of the reconstructed sparse signals in such an adverse condition. By using bi-directional support estimation in conjunction with an iterative unconstrained Lorentzian norm optimization, we detect the impulse noise support first and then estimate the signal support to compute the values of sparse signals. Experimental results demonstrate that our proposed algorithm gives a higher average SNR and a lower computational cost than the conventional Lorentzian iterative thresholding algorithm does.

This paper addresses the energy efficient power allocation issue in massive multi-user multiple-input multiple-output (MU-MIMO) communication systems for both uplink and downlink transmission with circuit power consumption taken into account. The maximum-ratio combining (MRC) and zero-forcing (ZF) detectors in the uplink together with maximum-ratio transmission (MRT) and ZF precoder in the downlink are considered. The key contribution of this paper lies in formulating the original energy efficient power allocation and converting such a complicated optimization problem to a geometric programming problem which can then be solved by a standard optimization software package.

This study investigates the secrecy performance of a hybrid satellite‐terrestrial relay network (HSTRN) in the presence of multiple eavesdroppers, where the satellite link undergoes Shadowed‐Rician fading, while the terrestrial link follows Rayleigh fading. The authors suppose the direct link between the satellite and the intended user is unavailable due to heavy shadowing and adopt a multi‐antenna relay using either a decode‐and‐forward (DF) or an amplify‐and‐forward (AF) protocol to assist the transmission. By employing perfect channel state information of each link at the relay, the authors first apply receive maximal ratio combining beamforming (BF) and transmit zero‐forcing BF schemes to obtain the output signal‐to‐noise ratios (SNRs) of the intended user and eavesdroppers. Then, based on the Meijer‐G function and the moment generating function, the authors derive the analytical expressions of the ergodic secrecy rate for the considered HSTRN for both DF and AF protocols. Finally, Monte‐Carlo simulations are conducted to validate the theoretical performance analysis and reveal the effects of certain representative parameters on the system secrecy performance.

This paper studies the robust secure beamforming (BF) issue of fifth generation (5G) cellular system operating at millimeter wave frequency and coexisting with a satellite network. By employing an uniform planar array at the base station (BS) and assuming known imperfect angle-of-arrival-based channel state informations of multiple eavesdroppers (Eves), a constrained optimization problem is first formulated to maximize the worst-case achievable secrecy rate of the cellular user under the constraints of the transmit power of BS and the interference threshold of satellite earth station. Then, we propose two robust BF methods to solve the complex optimization problem for both coordinated and uncoordinated Eves. For the case of coordinated Eves, we propose a heuristic BF scheme, which transfers the worst-case problem into a min-max one such that the BF weight vectors can be obtained analytically. For uncoordinated Eves, we convert the non-convex problem into a convex one, and further propose an iterative penalty function-based algorithm to obtain the optimal BF weight vectors. Finally, simulation results are provided to confirm the effectiveness and superiority of the proposed robust BF schemes.

This paper investigates the secure communication of a cognitive satellite terrestrial network with software-defined architecture, where a gateway is acting as a control center to offer the resource allocation for the wireless systems. Specifically, we propose beamforming (BF) schemes to utilize the interference from the terrestrial network as a green source to enhance the physical-layer security for the satellite network, provided that the two networks share the portion of millimeter-wave frequencies. Supposing that the satellite employs multibeam antenna while the base station is equipped with a uniform planar array, we first formulate a constrained joint optimization problem to minimize the total transmit power while satisfying both the quality-of-service requirement of the terrestrial user and the secrecy rate (SR) requirements of the satellite users. Since the formulated optimization problem is nonconvex and mathematically intractable, we then propose two BF schemes to obtain the optimal solutions with high computational efficiency. For the case of one eavesdropper (Eve), we present a method to convert the nonconvex SR constraint to a second-order cone one and then adopt a penalty function approach to obtain the BF weight vectors. In the case of multiple Eves, by introducing a list of auxiliary variables, we propose a two-layer iterative BF scheme using penalty function approach together with gradient-based method to calculate the BF weight vectors. Finally, simulation results are given to demonstrate the effectiveness and superiority of the proposed BF schemes.

This letter proposes a beamforming (BF) scheme to enhance wireless information and power transfer in terrestrial cellular networks coexisting with satellite networks. By assuming that the energy receivers are the potential eavesdroppers overhearing signals intended for information receivers (IRs), we first formulate a constrained optimization problem to maximize the minimal achievable secrecy rate of the IRs subject to the constraints of energy harvest requirement, interference threshold, and transmit power budget. Through exploiting the sequential convex approximation method, we convert the original problem into a linear one with a series of linear matrix inequality and second-order cone constraints. An iterative algorithm is then proposed to obtain the BF weight vectors. Finally, simulation results demonstrate the effectiveness and superiority of the proposed scheme.

In this paper, we investigate the performance of a multi-pair massive MIMO AF relay network, where multiple source users communicate with multiple destination users through a relay equipped with a very large number of antennas, when zero-forcing (ZF) reception/transmission is utilized at the relay. In contrast to prior works, we derive new exact analytical closed-form expressions of the outage probability and the ergodic achievable rate, and these results are valid for any finite number of relay antennas. In particular, the asymptotic performance and power-scaling laws are further analyzed and compared under two different asymptotic cases for the number of relay antennas going to infinity with a fixed and large number of user pairs, respectively. Both analytical results and Mont-Carlo simulations have consistently shown the effects of the number of relay antennas and the number of user pairs on the outage probability, the achievable rate, and power-scaling laws.

In this paper, a new method for two-dimensional (2-D) direction-of-arrival (DOA) estimation is proposed. We first reconstruct the covariance matrix of the coarray with block-Toeplitz structure and then retrieve the DOAs. Our method is computationally efficient as supported by the derived closed-form expression for the estimated covariance matrix. Unlike other methods, which require fully loaded arrays, the proposed method can be applied in the case of common rectangular arrays with arbitrary geometries. The estimated azimuth and elevation angles are automatically paired. Moreover, our method is of high estimation accuracy and immune to the angle ambiguity effect. Numerical simulations are carried out to verify the effectiveness of the proposed method.

This paper investigates the outage performance of cognitive broadband satellite systems and terrestrial cellular network in millimeter wave (nunWave) scenario. Considering the state-of-art standard recommendations and nun Wave propagation model, we first define a general framework for the coexistence of broadband satellite system and terrestrial cellular networks with distinct geometry, configuration, and channel characteristics. Then, by employing a interference temperature constraint at the fixed satellite service (FSS) receiver to limit the interference below an acceptable level, closed-form expression for the outage probability (OP) of the cognitive cellular user is derived, which is general and applicable to various channel parameters and geometric scenarios. Eventually, simulation results are carried out to verify the theoretical derivations, and shows the impact of key system parameters on the performance of the terrestrial cellular user with the coexistence of FSS.

In this paper, a novel pilot-aided algorithm is proposed for the detection of integer frequency offset (IFO) in orthogonal frequency division multiplexing (OFDM) systems. By transforming the IFO into two new integer parameters, the proposed method can largely reduce the number of trial values for the true IFO. The two new integer parameters are detected using two different pilot sequences, a periodic pilot sequence and an aperiodic pilot sequence. It is shown that the new scheme can significantly reduce the computational complexity while achieving almost the same performance as compared to previous methods.

This study investigates the spectral efficiency of multi‐pair massive multiple‐input multiple‐output amplify‐and‐forward relay networks by considering the composite channel aging effect. The rational of this work is that the channel aging effect caused by phase noise and node movements is an inevitable practical channel impairment in time‐varying fading channels and substantially degrades the performance. The proposed model comprises multiple sources and multiple destinations, each equipped with a single antenna, which communication via a relay equipped with a very large number of antennas by employing maximal‐ratio combining/maximum‐ratio transmission. Based on this model, the authors first derive a closed‐form lower bound expression for the achievable rate of per source–destination pair. Then, by using the derived expression, they study how the transmitted powers of each source and the relay can be reduced without compromising the spectral efficiency when the number of relay antennas approaches infinity. They also discuss the effect of channel aging coefficients, including Doppler shift and phase noise increment variance, on the asymptotic spectral efficiency under different power scaling laws. Both theoretical analysis and Monte Carlo simulations disclose that the channel aging effect does not affect the power scaling laws but degrades the spectral efficiency.

Feature selection and fusion is of crucial importance in multi‐feature visual tracking. This study proposes a multi‐task kernel‐based sparse learning method for multi‐feature visual tracking. The proposed sparse learning method can discriminate the reliable and unreliable features for optimal multi‐feature fusion through using a Fisher discrimination criterion‐based multi‐objective model to adaptively train the kernel weights of different features such as pixel intensity, edge and texture. To guarantee a robustness of the sparse representation method, a mixed norm is employed in the sparse leaning method to adaptively select correlated particle observations for multi‐task sparse reconstruction. Experimental results show that the proposed sparse learning method can achieve a better tracking performance than state‐of‐the‐art tracking methods do.

It is known that affinity propagation can perform exemplar-based unsupervised image clustering by taking as input similarities between pairs of images and producing a set of exemplars that best represent the images, and then assigning each nonexemplar image to its most appropriate exemplar. However, the clustering performance of affinity propagation is largely limited by the adopted similarity between any pair of images. As the scale invariant feature transform (SIFT) has been widely employed to extract image features, the nonmetric similarity between any pair of images was proposed by “hard” matching of SIFT features (e.g., counting the number of matching SIFT features). In this letter, we notice, however, that the decision of hard matching of SIFT features is binary, which is not necessary for deriving similarities. Hence, we propose a novel measure of similarities by replacing hard matching with the so-called soft matching. Experimental examples show that significant performance gains can be achieved by the resulting affinity propagation algorithm.

In this paper, a noise robust formant frequency estimation scheme is developed based on a spectral model matching algorithm. Considering the vocal tract as an autoregressive system, a spectral model of repeated autocorrelation function (RACF) of band-limited speech signal is proposed. It is shown that because of the repeated autocorrelation operation on band-limited signal, the proposed model can exhibit prominent formant characteristics. First from given noisy speech observations, an adaptive band selection criterion is developed. Next, on each resulting band-limited noisy speech signal, a repeated autocorrelation operation is carried out, which not only reduces the effect of noise but also strengthens the dominant poles corresponding to the formant frequencies. Finally, spectrum of the RACF is computed and instead of direct spectral peak picking, a model fitting scheme is introduced to find out model parameters which lead to formant estimation. The proposed algorithm has been tested on natural vowels as well as some naturally spoken sentences in the presence of different environmental noises. It is found that the proposed scheme provides better formant estimation accuracy in comparison to some of the existing methods at low levels of signal-to-noise ratio.

In this paper, a transmit antenna selection (TAS)-based incremental hybrid decode-amplify-forward (IHDAF) scheme is proposed to enhance physical layer security in cooperative relay networks. Specifically, TAS is adopted at the source in order to reduce the feedback overhead. In the proposed TAS-based IHDAF scheme, the network transmits signals to the destination adaptive select direction transmission (DT) mode, AF mode, or DF mode depending on the capacity of the source-relay link and source-relay link. In order to fully examine the benefits of the proposed TAS-based IHDAF scheme, we first derive its secrecy outage probability (SOP) in a closed-form expression. We then conduct asymptotic analysis on the SOP, which reveals the secrecy performance floor of the proposed TAS-based IHDAF scheme when no channel state information is available at the source. Theoretical analysis and simulation results demonstrate that the proposed TAS-based IHDAF scheme outperforms the selective decode-and-forward, the incremental decode-and-forward, and the noncooperative DT schemes in terms of the SOP and effective secrecy throughout, especially when the relay is close to the destination. Furthermore, the proposed TAS-based IHDAF scheme offer a good tradeoff between complexity and performance compared with using all antennas at the source.

In this paper, we study the energy efficiency of secure communication in an underlay cognitive radio network (CRN). We first formulate an optimization problem to maximize the secrecy energy efficiency (SEE) while meeting the quality-of-service (QoS) requirement for the primary user and the transmit power constraint at each base station. Since the problem is non-convex and very difficult to solve, we then convert the original fractional form into a subtractive one, and adopt the difference of two-convex functions (D.C.) approximation method to obtain an equivalent convex problem. Furthermore, a two-layer iterative algorithm is presented to solve the problem and obtain the optimal beamforming (BF) weight vectors. Finally, numerical results are provided to demonstrate the superiority of the proposed scheme.

In this paper, we present a new dereverberation algorithm based on the weighted prediction error (WPE) method. In contrast to the conventional WPE method which alternatively estimates the reverberation prediction weights and early speech spectral variance, the proposed algorithm estimates the latter efficiently by employing a geometric spectral enhancement approach and a proper estimate for late reverberant spectral variance (LRSV). Hence, our algorithm does not require iterations to estimate the reverberation prediction weights nor needs alternation between the prediction weights and the spectral variance of early speech. Performance assessments demonstrate considerable improvements in terms of speech quality measures and computational load compared to previous WPE-based dereverberation methods.

This paper addresses the issue of direction-of-arrival (DOA) estimation with an objective to eliminate the off-grid effect of the sparsity-based methods and enlarge the maximum number of distinguishable signals in the subspace-based methods. We first reconstruct the covariance matrix of the array output in the Toeplitz structure and then employ the reconstructed covariance matrix together with root-MUSIC to estimate the DOAs. The proposed covariance matrix reconstruction approach (CMRA) can be used for uniform and sparse linear arrays. It can also estimate the DOAs of multiple signals that are larger than the number of sensors by taking advantage of the array geometry. In contrast to the sparsity-based methods, CMRA is formulated in the continuous angle space rather than the discretized one, and hence it is immune to the off-grid effect. Simulations are carried out to verify the effectiveness of our method.

This paper addresses the pilot and data power allocation issue in time division duplexing (TDD) massive multi-user multiple-input multiple-output (MU-MIMO) systems. By using minimum mean square error (MMSE) channel estimation along with a maximum-ratio combining (MRC) detector for the uplink transmission and a maximum-ratio transmission (MRT) precoder for the downlink transmission, a novel pilot and data power allocation scheme is proposed to minimize the total uplink and downlink transmit power under per-user signal to interference-plus-noise ratio (SINR) and power consumption constraints. The main contribution of this paper lies in formulating the original energy efficient power allocation problem and converting such a complicated optimization problem to a geometric programming one. Computer simulation shows that the proposed scheme can save up to 78% of the total power as compared to the equal power allocation among all the mobile users.

The integration of cognitive radio (CR) into satellite networks is recognized as an effective strategy to enhance the efficiency of radio spectrum. This paper investigates the ergodic capacity of a multiple antenna cognitive satellite terrestrial network, where the secondary terrestrial system can coexist with the primary satellite system as long as the interference imposed from the secondary user (SU) to the primary user (PU) is below a predefined threshold. Specifically, the Meijer-G function based analytical expression for the ergodic capacity of the secondary network is derived, which not only provides an efficient means to evaluate the system performance but also characterize the impact of various channel parameters on the network. Finally, simulation results are provided to demonstrate the validity of the theoretical analysis.

This paper is concerned with the energy efficiency of secure transmission in an underlay cognitive relay network (CRN), where a secondary source communicates with a secondary destination via a multi-antenna relay in the presence of an eavesdropper. We first establish an optimization problem to maximize the secrecy energy efficiency (SEE) under the constraints of data rate and transmit power of the cognitive transmission as well as the interference limitation to the primary user. Then, we recast the original non-convex problem in fractional form into an equivalent subtractive one with an additional rank-one constraint. Moreover, we incorporate the rank-one constraint into the objective function as the penalty term and apply the difference of two-convex functions (D.C.) approach to obtain an equivalent convex problem. Finally, we present an iterative algorithm to obtain the optimal solution for the SEE maximization problem in the CRN. Numerical results are provided to demonstrate the effectiveness of the proposed scheme.

In this paper, we propose a single channel speech enhancement algorithm using a subband iterative Kalman filter. A wavelet filterbank is first used to decompose the noise corrupted speech into a number of subbands. To achieve the best tradeoff among the noise reduction, speech intelligibility and computational complexity, a partial reconstruction scheme based on consecutive mean squared error is proposed to synthesize the low-frequency (LF) and high-frequency (HF) subbands. An iterative Kalman filter is then applied to the partially reconstructed HF subband speech. Finally, the enhanced HF subband speech is combined with the partially reconstructed LF subband speech to reconstruct the fullband enhanced speech. Experimental results show that the proposed subband iterative Kalman filter based algorithm is capable of reducing adverse environmental noises for a wide range of input SNRs. The overall performance of our method in terms of segmental SNR, perceptual evaluation of speech quality (PESQ) and computational cost is superior to several existing Kalman filter based algorithms.

In this paper, we propose a joint pilot and data power allocation scheme aiming to improve the energy efficiency of time division duplexing (TDD) massive multi-user multiple-input multiple-output (MU-MIMO) communication systems for both uplink and downlink transmission. The proposed scheme uses a maximum-ratio combining (MRC) detector in the uplink together with a maximum-ratio transmission (MRT) precoder in the downlink. By using minimum mean square error (MMSE) channel estimation, the total uplink and downlink transmit power is minimized under per-user signal to interference-plus-noise ratio (SINR) requirement and per-user power consumption constraints. Lower bounds of the average SINR are derived and used in the power allocation algorithm in order to simplify the optimization problem. The tightness of the derived SINR lower bounds and the advantage of the proposed power saving scheme as compared to equal power allocation among all users are validated by computer simulation.

This study aims to improve the energy efficiency of time‐division duplexing massive multi‐user multiple‐input multiple‐output communication systems for both uplink and downlink transmissions. By using minimum mean square error channel estimation, two novel pilot–data power allocation schemes are proposed to minimise the total uplink and downlink transmit power under per‐user signal‐to‐interference‐plus‐noise ratio (SINR) requirement and per‐user power consumption constraints. The proposed schemes take into account the maximum‐ratio combining and zero‐forcing (ZF) detectors in the uplink transmission together with maximum‐ratio transmission and ZF precoder in the downlink transmission. In order to simplify the proposed optimisation problems, lower bounds of the average SINR are derived and used in the power allocation algorithms. The key contribution of this study lies in formulating the original energy‐efficient power allocation problem and converting such a complicated optimisation problem to a geometric programming problem. Computer simulation validates the tightness of the derived SINR lower bounds and shows that the proposed schemes can save up to 78% of the total power as compared with the equal power allocation among all the mobile users.

This paper investigates the physical layer security of a satellite network, whose downlink spectral resource is shared with a terrestrial cellular network. We propose to employ a multi-antenna base station (BS) as a source of green interference to enhance secure transmission in the satellite network. By taking the mutual interference between these two networks into account, we first formulate a constrained optimization problem to maximize the instantaneous rate of the terrestrial user while satisfying the interference probability constraint of the satellite user. Then, with the assumption that imperfect channel state information (CSI) and statistical CSI of the link between the BS and satellite user are available at the BS, we present two beamforming (BF) schemes, namely, hybrid zero-forcing and partial zero-forcing to solve the optimization problem and obtain the BF weight vectors in a closed form. Moreover, we analyze the secrecy performance of primary satellite network by considering two practical scenarios, namely: Scenario I, the eavesdroppers CSI is unknown at the satellite and Scenario II, the eavesdroppers CSI is known at the satellite. Specifically, we derive the analytical expressions for the secrecy outage probability for Scenario I and the average secrecy rate for Scenario II. Finally, numerical results are provided to confirm the superiority of the proposed BF schemes and the validity of the performance analysis, as well as demonstrate the impacts of various parameters on the secrecy performance of the satellite network.

In this paper, we address the issue of joint beamforming (BF) and power control for a device-to-device (D2D) communication underlaying cellular network, where the wireless channels of the D2D link and the base station to user equipment link experience Rician and correlated Rayleigh fading, respectively. Based on the property of the integral network, we first formulate a constrained optimization problem to minimize the total transmit power of the devices in the network, while meeting the quality-of-service requirement of both the D2D and cellular users and suppressing the mutual interference to a certain level. Then, by adopting the available statistical channel state information and proposing an approximation method to relax the constraints, a support-vector-machine-based algorithm is presented to solve the optimization problem for the transmit powers and BF weight vectors of each user. Furthermore, we derive the analytical expressions for the cumulative density function and the generalized moments of the output signal-to-interference-plus-noise ratios, thereby developing some novel theoretical formulas for the ergodic capacity and the average symbol error rate of each user in the network. Finally, computer simulation results are provided to demonstrate the validity and efficiency of the proposed scheme and its performance analysis.

In this paper, we investigate new methods for preamble-aided coarse timing estimation in orthogonal frequency division multiplexing systems. Two novel timing metrics using high-order statistics-based correlation and differential normalization functions are first proposed. The performance of the new timing metrics is then evaluated using different criteria, including class separability, robustness to the carrier frequency offset, and computational complexity. It is shown that the new timing metrics can considerably increase the class separability due to their more distinct values at correct and wrong timing instants, and thus give a significantly better detection performance as compared with the existing timing metrics. Furthermore, a new method for coarse estimation of the start of the frame is proposed, which remarkably reduces the probability of intersymbol interference (ISI). The improved performances of the new schemes in multipath fading channels are shown by the probabilities of false alarm, missed detection, and ISI obtained through computer simulations.

In this paper, we propose a real-valued ESPRIT algorithm for two-dimensional direction of arrival (2D DOA) estimation of noncircular signals using arbitrarily spaced acoustic vector sensor array. By utilizing the noncircularity of the signals, the proposed algorithm provides a better estimation performance while having only slightly larger computational complexity as compared to the traditional complex ESPRIT algorithm. Furthermore, the proposed algorithm gives automatically paired azimuth and elevation angle estimates without requiring extra pair matching. Simulation results are presented to demonstrate the estimation performance of the proposed algorithm as compared to the traditional ESPRIT algorithm as well as Cramer-Rao bound (CRB).

In this paper, we present a new algorithm for the estimation of the noise power spectral density (PSD) matrix, as needed for multi-microphone speech enhancement in a general non-stationary noisy environment. First, we propose a recursive scheme for noise PSD estimation in which the current, previous and close subsequent noisy speech frames are properly weighted. The forgetting factor for the recursive updating of the smoothed PSD is obtained based on an overall measure of the SNR across all microphone signals. Since this SNR measure depends on the noise statistics, we choose to iteratively update it using the latest available estimate of the noise PSD matrix. Finally, to obtain better estimation accuracy in the proposed method, we further apply a direct extension of the minimum tracking approach to the estimated noise PSD matrix. Performance of the proposed algorithm is evaluated in terms of objective measures and its superiority is shown with respect to two recent noise PSD estimation methods in the context of speech enhancement.

In his paper a method of seizure detection has been proposed based on the Discrete Wavelet Transform (DWT) analysis of the dominant Intrinsic mode function(IMF) resulting from the Empirical Mode Decomposition(EMD) of the EEG signals. Considering the normalized energy, Fourier spectrum and cross-correlation coefficient analysis, only the 4th Level DWT coefficients of the dominant IMF is found reasonable for feature computation. In order to reduce the dimension of the feature vector, Higher order statistics of these coefficients are employed to form he feature vector. The reduced feature vector thus formed is found effective for distinguishing seizure and non-seizure EEG signals when fed to a k-nearest neighborhood (k-NN) classifier. Extensive simulations are carried out using a benchmark EEG dataset. It is shown that the proposed method is capable of producing greater sensitivity, specificity and accuracy in comparison to that obtained by a sate-of-the-art method using the same EEG dataset and classifier.

This study proposes a transmit power allocation (TPA) scheme for bi‐directional relay networks with an objective of minimising the total power consumption to meet both the service quality (i.e. the outage probability) and individual power requirements. This new scheme focuses on the amplify‐and‐forward protocol‐based multiple‐access broadcast mode with asymmetric network traffics where the bi‐directional relay channel (BDRC) statistics are assumed to be available at the transmitters. A two‐step method is devised to solve the optimisation problem pertaining to the total power minimisation. In the proposed method, the system outage probability is first minimised subject to both the individual and total power constraints, and then the total power consumption of the network is minimised subject to the given service quality constraint based on the preliminary solutions achieved in the first step. This two‐step optimisation mechanism leads to a novel TPA algorithm for the relay and two sources of the network. Simulation results are provided to validate the proposed algorithm, showing that the proposed new power allocation scheme can significantly reduce the total power consumption, especially when the BDRC or the network traffic is asymmetric.

Moving object detection plays a key role in video surveillance. A number of object detection methods have been proposed in the spatial domain. In this study, the authors propose a compressed sensing‐based algorithm for the detection of moving object. They first use a practical three‐dimensional circulant sampling method to yield sampled measurements. Then, they propose an object detection model to simultaneously reconstruct the foreground support, background and video sequence using the sampled measurements directly. Experimental results show that the proposed moving object detection algorithm outperforms the state‐of‐the‐art approaches and it is robust to the movement turbulence, camera motion and video noise.

This paper presents a speech enhancement approach, where an adaptive threshold is statistically determined based on Student$t$Modeling of Teager energy (TE) operated perceptual wavelet packet (PWP) coefficients of noisy speech. In order to obtain an enhanced speech, the threshold thus derived is applied upon the PWP coefficients by employing a Student$t$pdf dependent custom thresholding function, which is designed based on a combination of modified hard and semisoft thresholding functions. Extensive simulations are carried out using the NOIZEUS database to evaluate the effectiveness of the proposed method for car and multi-talker babble noise corrupted speech signals. Several standard objective measures and subjective evaluations including formal listening tests show that the proposed method outperforms some of the state-of-the-art speech enhancement methods at high as well as low levels of SNRs.

We consider a cognitive radio (CR) network consisting of a secondary transmitter (ST), a secondary destination (SD) and multiple secondary relays (SRs) in the presence of an eavesdropper, where the ST transmits to the SD with the assistance of SRs, while the eavesdropper attempts to intercept the secondary transmission. We rely on careful relay selection for protecting the ST-SD transmission against the eavesdropper with the aid of both single-relay and multi-relay selection. To be specific, only the “best” SR is chosen in the single-relay selection for assisting the secondary transmission, whereas the multi-relay selection invokes multiple SRs for simultaneously forwarding the ST's transmission to the SD. We analyze both the intercept probability and outage probability of the proposed single-relay and multi-relay selection schemes for the secondary transmission relying on realistic spectrum sensing. We also evaluate the performance of classic direct transmission and artificial noise based methods for the purpose of comparison with the proposed relay selection schemes. It is shown that as the intercept probability requirement is relaxed, the outage performance of the direct transmission, the artificial noise based and the relay selection schemes improves, and vice versa. This implies a trade-off between the security and reliability of the secondary transmission in the presence of eavesdropping attacks, which is referred to as the security-reliability trade-off (SRT). Furthermore, we demonstrate that the SRTs of the single-relay and multi-relay selection schemes are generally better than that of classic direct transmission, explicitly demonstrating the advantage of the proposed relay selection in terms of protecting the secondary transmissions against eavesdropping attacks. Moreover, as the number of SRs increases, the SRTs of the proposed single-relay and multi-relay selection approaches significantly improve. Finally, our numerical results show that as expected, the multi-relay selection scheme achieves a better SRT performance than the single-relay selection.

This paper addresses robust beamforming (BF) design for the uplink transmission of wireless networks, where the base station (BS) equipped with a very large number of antennas communicates with multiple users on the same frequency band simultaneously. Based on the assumption that the wireless channels undergo correlated Rayleigh fading, we first formulate an optimization problem to maximize the output signal-to-interference-plus-noise ratio (SINR) of the intended users. Then, by using the fact that channel uncertainty is norm-bounded and imperfect channel state information (CSI) is available at the BS, we transform the optimization problem to a support vector machine (SVM) regression one, and obtain the robust solution for the BF weight vectors by means of quadratic programming (QP) technique or iterative reweighted least squares (IRWLS) procedure. The computational cost of the proposed robust BF scheme depends on the number of channel vector samples rather than that of the antennas, thus it is suitable for the wireless systems with large-scale antennas. Finally, the efficiency and superiority of the proposed new scheme are confirmed through computer simulation.

In this paper, we address the problem of estimating the two-dimensional (2-D) directions of arrival (DOA) of multiple signals, by means of a sparse L-shaped array. The array consists of one uniform linear array (ULA) and one sparse linear array (SLA). The shift-invariance property of the ULA is used to estimate the elevation angles with low computational burden. The source waveforms are then obtained by the estimated elevational angles, which together with each sensor of the SLA, considered as a linear regression model, will be used to estimate the azimuth angle by the modified total least squares (MTLS) technique. The new algorithm yields correct parameter pairs without requiring the computationally expensive pairing operation, and therefore, it has at least two advantages over the previous L-shaped array based algorithms: less computational load and better performance due to using the SLA. Simulation results show that our method provides accurate and consistent 2-D DOA estimation results which could not be achieved by other methods with comparable computational complexity.

Bayesian estimators of short time spectral amplitude (STSA) have received considerable attention in the field of speech enhancement. In this paper, we propose new multi-microphone extensions for the conventional Ephraim and Malah's speech spectral amplitude estimation method. Unlike the conventional estimators where the spectral phase is assumed to be uniformly distributed, the proposed extensions treat the latter as an unknown parameter to be estimated. It is shown that the proposed methods can exploit spectral phase estimates to improve the performance of the current speech STSA estimators and have the potential to provide even further improvement given a more accurate estimate of the spectral phase. Experimental results indicate the superiority of the new approaches in terms of noise reduction and speech distortion measures, in addition to the reduced computational complexity provided by the proposed minimum mean square method as compared to state-of-the-art solutions.

This paper presents a speaker-independent speech emotion recognition method, where emotional features are derived from the Teager energy (TE) operated wavelet coefficients of speech signal. Due to TE operation, the enhanced detail as well as approximate Wavelet coefficients thus obtained is then used to compute entropy. Entropy values of TE operated detail and approximate wavelet coefficients not only reduces feature dimension but also form an effective feature vector for distinguishing different emotions when fed to a Euclidean distance based classifier. Extensive simulations are carried out using EMO-DB German speech emotion database containing four class emotions, such as angry, happy, sad and neutral. Simulation results show that the proposed method is capable of outperforming an existing speaker-independent emotion recognition method thus solving a four-class emotion recognition problem in terms of higher recognition accuracy with lower computation.

This paper studies the asymptotic capacity of large MIMO amplify-and-forward (AF) relay networks with one source-destination pair equipped with M antennas and K single-antenna relays. In contrast to prior works, we consider not only the multipath fading but also the path loss effect. By using random matrix theory, we first derive the asymptotic eigenvalue distribution (A.E.D.) of the channel covariance matrix between the source and the destination as M and K approach infinity while their ratio remains constant. Based on the resulting A.E.D., the system capacity is then obtained. More specifically, a linear relay distribution is considered to evaluate the asymptotic capacity, which can realistically model the large MIMO AF relay networks. Both theoretical analysis and Mont-Carlo simulations have consistently shown the impact of path loss parameters on the asymptotic capacity.

To improve the mean acquisition time performance of the zero-padding (ZP) method, we propose to fully exploit the computation capability of FFT by partially folding the local pseudo-noise (PN) code. The detection and mean acquisition time performance of the proposed method is analyzed and compared with that of the zero-padding method. It is shown that the proposed method can reduce the mean acquisition time significantly at the expense of less than 1.5 dB degradation in detection performance.

Most of the existing compressed channel‐sensing methods for multiple‐input–multiple‐output orthogonal frequency division multiplexing (MIMO‐OFDM) systems did not take into account the pulse‐shaping filter in the transmitter and matched filter in the receiver. However, these two filters are commonly used in digital communication systems. The compressed channel‐sensing problem of pulse‐shaping MIMO‐OFDM systems is first formulated. A new signal‐reconstruction algorithm in the compressed sensing framework is then proposed. The algorithm is based on minimising a smoothed l 0 ‐norm regularised least‐square (LS) ( l 2 − Sl 0 ) objective function, and the unconstrained optimisation involved is performed by an approximate gradient method. Further, the proposed l 2 − Sl 0 algorithm is applied to reconstruct the channel impulse response. A number of computer simulation‐based experiments are conducted, showing a better reconstruction accuracy of the l 2 − Sl 0 algorithm as compared with the smoothed l 0 ‐norm ( Sl 0 ) algorithm. The proposed channel estimation approach can save nearly 25% pilot signals to maintain the same mean square error and bit error rate performances as given by the conventional LS method.

Compressive sampling matching pursuit (CoSaMP) is an efficient compressive sensing algorithm holding rigorous estimation error bounds and low computational complexity, when it deals with an additive noise signal model in the observation domain. However, in some applications, e.g., speech enhancement (SE), noise is added to a signal in the time domain, where the conventional CoSaMP cannot be directly applied. In this paper, we establish the theory of CoSaMP to address the time-domain noise, referred to as Tdn-CoSaMP, which extends the canonical theory of CoSaMP. In particular, we prove the existence of a new upper bound of Tdn-CoSaMP, which is found to be larger than that of the conventional CoSaMP by appending two additional terms: a multiplier$1+\sqrt{{N\over s}}$, where$N$is the dimension of the signal, and an${\ell_1}$norm of the noise${1\over\sqrt{s}}\Vert {\mbi{e}}\Vert_1$scaled by the sparse level$s$of the signal. We also apply Tdn-CoSaMP to the SE task based on the sequential denoising of overlapped frames in the discrete cosine transform (DCT) domain. The proposed system, CoSaMP-based speech enhancement (CoSaMPSE), has been evaluated in terms of both objective and subjective criteria on various types of noise. Positive results have been achieved for denoising stationary and nonstationary white Gaussian noise (WGN) and are comparable to other SE methods. Moreover, due to its low computational complexity, CoSaMPSE is possible to be combined with optimally modified log-spectrum amplitude estimation (OMLSA) and able to achieve complementary denoising effects in various noisy conditions.

In this paper, we address the problem of beamforming (BF) design in a cognitive relay network (CRN), where the cognitive network not only shares the spectrum with the primary one but also acts as a relay to assist the primary signal transmission. Considering that all of the wireless channels are subject to the correlated Rayleigh fading distribution, we first formulate a constrained optimization problem to minimize the total transmit power of the cognitive base station (CBS) with multiple antennas, while guaranteeing the quality-of-service (QoS) of the primary and secondary users and keeping the mutual interference below an acceptable level. Then, by adopting the available partial channel state information (CSI), and proposing an approximation method to relax the constraints, a support vector machine (SVM) based algorithm is presented to solve the optimization problem for the BF weight vectors. The benefit of the new approach is that the slow change of the partial CSI can be taken into account. Finally, computer simulation results are provided to confirm the superiority of the proposed BF design strategy.

Amyotrophic lateral sclerosis (ALS) is a common fatal motor neuron disease that assails the nerve cells in the brain. As the nervous system controls the muscle activity, the electromyography (EMG) signals can be viewed and examined in order to detect the vital features of the ALS disease in individuals. In this paper, the discrete wavelet transform (DWT) based features, which are extracted from a frame of EMG data, are introduced to classify the normal person and the ALS patients. From each frame of EMG data, instead of using a large number of DWT coefficients, the DWT coefficients with higher values as well as their mean and maxima are proposed to be used, which drastically reduces the feature dimension. It is shown that the proposed feature vector offers a high within class compactness and between class separations. For the purpose of classification, the K-nearest neighborhood classifier is employed. In order to demonstrate the classification performance, an EMG database consisted of 5 normal subjects and 5 ALS patients is considered and it is found that the proposed method is capable of distinctly separating the ALS patients from the normal persons.

There are two main ways by which antenna arrays can significantly improve the performance of the direction-of-arrival (DOA) estimation. In the first method, one can extend the array aperture by designing a sparse antennas array. The second method makes use of the temporal information of the received signals, such as the signal waveform. A few articles have dealt with DOA estimation by combining the above two approaches. In this paper, we present a DOA estimation and tracking method by employing the known waveform of the signal and the parallel recursive least square (RLS) technique. When the waveform of the signal is known, the output of each sensor in the array can be considered as a linear regression model of which the coefficients contain the whole information for estimating the DOA. Therefore, the RLS filter is used to estimate and track these coefficients and then the relationship among the coefficients of all the sensors is exploited to obtain the final DOA value without ambiguity. Finally, computer simulation of the proposed method with comparison to the previous methods is conducted.

Fitting biophysical models to real noisy data jointly with extracting fundamental biophysical parameters has recently stimulated tremendous studies in computational neuroscience. Hodgkin-Huxley (HH) neuronal model has been considered as the most detailed biophysical model for representing the dynamical behavior of the spiking neurons. In this paper, we derive, for the first time, the dual extended Kalman filtering (DEKF) approach for the HH neuronal model to track the dynamics and estimate the parameters of a single neuron from noisy recorded membrane voltage. As unscented Kalman filter (UKF) has been already applied to the HH model, a quantitative comparison between these methods is accomplished in our simulation for different signal to observation noise ratios. Our simulations demonstrate the high accuracy of DEKF in the prediction and estimation of hidden states and unknown parameters of the HH neuronal model. Faster implementation of DEKF (than UKF) makes it particularly useful in dynamic clamp technique.

In this paper, a phase spectrum based acoustic parameter is presented for the detection of nasalized vowels from the mixture of oral and nasalized vowels of normal speakers. Acoustic analysis shows that during the event of nasalization, although additional formants (resonances) at various frequency locations are introduced, the introduction of a new formant in low frequency region around 250 Hz is found to remain consistent irrespective of female or male speakers in the modified group delay derived from the phase spectrum. By exploiting and verifying this fact on the band-limited modified group delay spectrum capable of resolving two closely spaced formants, an acoustic parameter RMGD is derived. Utilizing RMGD, the problem of detecting nasalized vowels is solved based on a threshold based scheme or a Euclidean distance based classifier. Simulation Results on TIMIT database show that the proposed method even with a simple classifier is superior in performance in comparison to that of the methods using Mel-frequency cepstral coefficients as a feature and Hidden Markov Modeling or Support Vector Machine as a classifier.

This paper investigates opportunistic relaying of a half-duplex two-way relay network, where two source users exchange information through a shared amplify-and-forward (AF) relay which is opportunistically selected from a set of candidates. Unlike existing works, in our considered network scenario, the two source users have different target rates. A novel protocol for opportunistic relaying is developed, which adaptively switches between the max-min and the pure-max policies according to the information of system traffic. The performance of the new relaying strategy in terms of the system outage probability is analyzed over Rayleigh fading channels. Simulation results demonstrate that the proposed protocol can implement opportunistic relaying efficiently, and can achieve significant performance gains in terms of the outage probability across all traffics.

In this paper, we propose a joint relay and destination optimization design for two-way half-duplex amplify-and-forward (AF) relay systems with multiple relays, each with multiple antennas. By using the sum mean-squared error (MSE) criterion and Wiener filtering principle, the joint relay and destination design is formulated as an optimization problem of the relay precoding matrix under the constraint of total relay transmit power. Then, constructing a virtual point to point MIMO channel and using the singular-value-decomposition (SVD), the optimization problem is simplified to a convex minimization of an upper bound of the sum MSE through a diagonalization process. Finally, a suboptimum scheme is proposed to solve the simplified convex optimization problem. Monte-Carlo simulation shows that the proposed suboptimal scheme gives a better MSE performance than the existing gradient descent algorithm does, and moreover, our method becomes more advantageous when the number of relays or the number of relay antennas increases.

Owing to the characteristic of single frequency network (SFN) and non-line of sight (NLOS) environment, base station (BS) identification becomes a prominent problem in SFN positioning systems. In this work, a universal BS identification algorithm is proposed for any DTV standard. The basic idea is to formulate the base station identification problem as a data classification which is then solved by the least distance classifier. The state prediction model is utilized to estimate kinematics parameters and thus decrease the system hardware requirement. In order to obtain accurate state prediction for BS identification, an interacting multiple model (IMM) method is adopted to mitigate the NLOS effect for performance improvement. Because of close relation between the position estimation and BS identification, the proposed method is more effective as compared to some existing methods. Simulation results show that the proposed algorithm can perform well in both unfixed BS set and NLOS environments.

In this study, an efficient image fusion framework for multi‐focus images is proposed based on compressed sensing. The new fusion framework consists of three parts: image sampling, measurement fusion and image reconstruction. First, the dual‐channel pulse coupled neural network model is used in the image sampling part as an important weighting factor in the fusion scheme. Second, the result from the measurement fusion part is reconstructed through a new reconstruction algorithm called self‐adaptively modified Landwebber filter. Finally, computer simulation‐based experiment is conducted, showing that the novel fusion framework is capable of saving computational resource and enhancing the fusion result and is easy to implement.

In the authors previous work, a compressive sensing (CS)‐based method has been proposed to address speech enhancement (SE) in adverse environments (CS‐SPEN) based on an assumption of sparse noise. However, this assumption may not be satisfied in practical noisy environments. In this study, the authors study this issue by relaxing this assumption to consider a general non‐sparse noise case, such that the proposed method naturally extends the previous one. In particular, they solve the theoretic difficulty of CS‐SPEN on the treatment of non‐sparse noise by using a relaxed upper bound for the constraint governing data consistency and a relaxed estimation error bound. Their main result is mathematically proved. In addition, the effectiveness of the proposed method is demonstrated by computational simulations, showing certain improvements to the previous method for both stationary and non‐stationary white Gaussian noises across various segmental signal‐noise‐ratios (SNRs). In these cases, the proposed method is shown to have comparable results to the state‐of‐the‐art SE alogrithms and some advantages over them at low SNRs. CS‐SPEN without the sparse noise assumption works evenly with CS‐SPEN with the sparse noise assumption for car internal and F16 cockpit noises.

The articles in this special issue focus on the technology and applications supported by virtual multiple antennas, or VMIMOs. The impetus for this has been spurred by the strong desire to understand VMIMO, which is a rapidly growing research area. VMIMO is believed to be a key technology for beyond 4th generation mobile communications technologies (B4G). It enables one to make use of all the neighboring terminals and amortize the cost of multiple antennas; hence, a large MIMO channel can be created to increase capacity significantly as well as improve error rate performance. Nevertheless, fundamental roadblocks need to be addressed in order to take full advantage of VMIMO.

A novel iterative adaptive filtering approach is proposed to remove the Transcranial Magnetic Stimulation (TMS) induced artifact from multi-channel recordings of neural responses to sensory stimuli. For each specific channel, the average of all trials is considered as the input to the adaptive filter whose coefficients are calculated by minimizing the mean square error between the voltage trace of that trial and the filter output. The residues of all trials serve as an initial estimate of the neural response. Once this estimate is calculated, the input of the adaptive filter is modified by subtracting the mean residue. It is shown that the modified input provides a better estimate of the mean TMS artifact, which serves the input of the adaptive filer, in the next iteration. Therefore, new filter coefficients are estimated in the next iteration, for each single trial, and the procedure continues till no considerable changes in the residues occur. We report a quantitative verification of the accuracy of our method by generating a controlled simulation. Furthermore, applying the algorithm to experimental data confirms the accuracy of our approach and its usefulness for extracting neurophysiological responses occurring in temporal proximity to TMS pulses.

An accurate direction-of-arrival (DOA) estimation algorithm with sparse sensor array is proposed. By dividing the nonuniform linear sparse array (NLSA) into two uniform linear sparse arrays (ULSA), the subarray response vectors yield a property of rotational invariance in the estimation of rough DOA without ambiguity using the so-called generalized ESPRIT. According to the estimated rough DOA, the accurate DOA of each source is then obtained by the proposed alternating null-steering technique (ANST). Simulation results that demonstrate the performance of the algorithm are provided.

This paper presents an approach to detect voice disorders based on wavelet and prosody-related voice properties. First, several statistical measures of the normalized energy contents of the Discrete Wavelet Transform (DWT) coefficients over all voice frames are determined. Then, similar statistical measures of some prosody-related voice properties, such as mean pitch, jitter and shimmer are also computed over all the frames. In order to form a feature vector to be used in both training and testing phases, a set of statistical measure of the normalized energy contents of the DWT coefficients is combined with a set of statistical measure of the extracted prosody-related voice properties. Here, the voice samples under consideration are assumed to be of two categories, namely healthy and disordered thus formulating the problem in the proposed method as a two-class problem to be solved. Finally, the feature vector as obtained above is fed to an Euclidean Distance based classifier to detect the disordered voice. By performing extensive simulations, it is shown that the statistical analysis based on wavelet and prosody-related properties are able to provide effective detection of a variety of voice disorders from the mixture of healthy and disordered voices.

Signal sparsity is the fundamental requirement of compressive sensing (CS) techniques. In our previous work, a CS-based speech enhancement algorithm has been proposed. However, several issues concerning speech sparsity have not yet been thoroughly studied. In this paper, we focus on studying the following issues: (1) the sparsity of clean speech and audio signals; (2) the sparsity of various noise signals; (3) analysis of the capacity of two sparse transforms i.e., wavelet and discrete cosine transform (DCT), to explore speech sparsity. In this respect, several measures are proposed to analytically compare the wavelet transform with DCT. We found that (1) signal compressibility is an important factor for the CS-based method. (2) DCT explores the best compressibility for noisy signals and achieves the best enhancement performance; (2) The CS-based speech enhancement methods are more efficient in reducing the noise with worse compressibility.

This paper presents a joint linear processing scheme for two-hop and half-duplex distributed amplify-and-forward (AF) relaying networks with one source, one destination and multiple relays, each having multiple antennas. Under the perfect CSI (channel state information) assumption, a joint relay and destination design is proposed to minimize the mean-square error (MSE) of the output signal at the destination. By the Wiener filter principle, it is formulated as an optimization problem with respect to the relay precoding matrix under the constraint of a total relay transmit power. And then, the constrained optimization with an objective to design the relaying block-diagonal matrix is simplified to an equivalent problem with scalar optimization variables. Moreover, it is revealed that the scalar-version optimization is convex when the total relay power or the second-hop SNR (signal to noise ratio) is above a certain threshold. The underlying optimization problem, which is non-convex in general, is solved by the complementary geometric programming (CGP). Simulation results show that the proposed scheme outperforms the existing relaying method in the case of relatively large second-hop SNR.

A joint source-and-relay beamforming scheme is proposed for multiple-input multiple-output (MIMO) systems with distributed single-antenna relays. First, a lower-bound of the signal-to-noise ratio at the destination is derived as an objective function to formulate a constrained beamforming optimisation problem. The joint beamforming problem is then divided into two sub-optimisation problems corresponding to the source and the relay beamforming, respectively. The first sub-problem is shown to be a quadratic concave minimisation, and is tackled by developing an iterative algorithm with each iteration solving a linear problem. The second one corresponds to a Rayleigh–Ritz ratio problem which is then solved by the generalised singular-value decomposition in a closed form. Based on the solutions to the subproblems, a global iterative algorithm is designed to implement the joint source-and-relay beamforming. Simulation results show that the proposed method outperforms some existing relaying schemes in terms of the capacity and outage probability of the whole MIMO relay system.

Denoising is a very important preprocessing step in signal/image processing. In this study, a new signal denoising algorithm is proposed by using neighbouring wavelet coefficients. The dual-tree complex wavelet transform is employed because of its property of approximate shift invariance, which is very important in signal denoising. Both translation-invariant (TI) and non-TI versions of the denoising algorithm are considered. Experimental results show that the proposed method outperforms other existing methods in the literature for denoising both artificial and real-life noisy signals.

In this paper, a dual-hop multiple-input multiple-output (MIMO) amplify-and-forward (AF) relay network, where the source, relay and destination are each equipped with multiple antennas, is studied. By deriving and maximizing the receive signal-to-noise ratio (SNR) at the destination, we first obtain the optimal beamforming (BF) weights for the relay network. In order to evaluate the performance of the relay network, we then investigate the outage probability (OP), probability density function (PDF) and moments of the receive SNR as well as the ergodic capacity of the system in a closed-form. Furthermore, the average symbol error rate (ASER) expression of the relay network with the optimal transmit-receive BF is derived for three commonly used modulation formats, namely, M-ary pulse amplitude modulation (M-PAM), M-ary phase shift keying (M-PSK), and M-ary quadrature amplitude modulation (M-QAM). Finally, computer simulations are conducted to demonstrate the validity and efficacy of the designed MIMO relay network and its performance analysis.

The 15 papers in this special issue are divided into four categories: challenges in broadband wireless communications; physical layer techniques; radio resource management techniques; and field measurement and channel modeling.

In this paper, a method for the estimation of pitch from noise-corrupted speech observations based on extracting a pitch harmonic and the corresponding harmonic number is proposed. Starting from the harmonic representation of clean speech, a simple yet accurate harmonic sinusoidal autocorrelation (HSAC) model is first derived. By employing this HSAC model expressed in terms of the pitch harmonics of the clean speech, a new autocorrelation-domain least-squares fitting optimization technique is developed to extract a pitch harmonic from the noisy speech. Then, the harmonic number associated with the pitch harmonic is determined by maximizing an objective function formulated as an impulse-train weighted symmetric average magnitude sum function (SAMSF) of the noisy speech. The period of the impulse-train is governed by the estimated pitch harmonic and the maximization of the objective function is carried out through a time-domain matching of periodicity of the impulse-train with that of the SAMSF. An SAMSF-based pitch tracking scheme using dynamic programming is devised to obtain a smoothed pitch contour. In order to demonstrate the efficacy of the proposed method, simulations are conducted by considering naturally spoken speech signals in the presence of white or multi-talker babble noise at different signal-to-noise ratio (SNR) levels. A comprehensive evaluation of the pitch estimation results shows the superiority of the proposed method over some of the state-of-the-art methods under low levels of SNR.

Free probability theory as a main branch of random matrix theory is a valuable tool for describing the asymptotic behavior of multiple systems, especially for those with large random matrices. But classical freeness results mainly focus on random matrices with independent and identically distributed (i.i.d.) entries. In this paper, by using the latest research results of random Vandermonde matrix (a class of matrix that does not satisfy the i.i.d. condition) in free probability theory, a new spectrum sensing scheme is proposed, which shows how the asymptotic behavior of random Vandermonde matrices can be used to assist spectrum sensing for cognitive radio. Simulation results show that the proposed scheme has a better detection performance than the energy detection technique and the maximum-minimum eigenvalue scheme even for the case of a small sample of observations.

This paper presents a new method of estimating the direction-of-arrival (DOA) for multiple signals using minimum redundancy linear sparse subarrays (MRLSS). The proposed method makes use of the array structure to obtain the extended correlation matrix that is constructed by Kronecker Steering Vectors (KSVs) of which each contains the ambiguous and unambiguous angle with a one-to-one relationship. Our method enjoys two advantages in comparison to the existing methods. First, the cyclic ambiguity can be resolved by the one-to-one mapping of unambiguous angle without requiring additional algorithms such as MUSIC and MODE. Second, the proposed method can deal with different unambiguous angles with the same ambiguous angles, which could not have been possible by using the traditional schemes due to the fact that our method obtains the ambiguous and unambiguous angles simultaneously.

This paper addresses the problem of the direction-of-arrival (DOA) estimation using fewer receivers than sensors. Inspired by the Compressed Sensing (CS) theory developed in recent years, we present a new preprocessing scheme for a large array using a small size receiver. Unlike the traditional ℓ2-norm-based algorithms by judicious selection of the preprocessing matrix, the proposed scheme uses a random weight generator as a measurement of the compressed sensing to form the output data for each time interval. The formulated CS problem for DOA estimation is then solved based on the convex programming via ℓ1-norm approximation such as Dantzig Selector. We consider two different scenarios in the CS domain, i.e., the angle domain and the angle-frequency domain. It is shown that the number of receivers can be reduced significantly for a given number of sensors by using the proposed CS-based DOA estimation approach.

This paper presents a relay precoding scheme under the imperfect channel state information (CSI)assumption for two-hop amplify-and-forward based MIMO relay networks with one source, one destination and multiple relays. The forward channel mean and covariance available at the relays is assumed to be for the robust design of the linear relay precoder. By using the average mutual information criterion along with a total relay transmitting power constraint, an optimization problem is formulated by assuming the imperfect CSIT(CSI at the transmitter) at the relays. We first derive the approximation of the average mutual information rate, and then simplify the constrained optimization problem into the equivalent problem involving scalar optimization variables. Through an in-depth study of the problem formulation, it is revealed that the objective function is convex only when each optimization variable is larger than a certain threshold, which can be determined mainly by the system configurations and the channel condition. The effectiveness of the proposed relay precoder is validated by Monte-Carlo simulations with comparison to that under perfect CSI assumption.

By employing the amplify-and-forward relaying strategy, a relay selection method with simplified precoding is proposed for two-hop multi-relay channels. The proposed method consists of two parts. In the first part, the authors present an optimal beamforming scheme with full channel state information (CSI) and derive a simplified and analytic expression for the output SNR through each relay, thereby selecting the best relay. The second part considers the transmiit precoding problem with the CSI known to the receiver only and provide a selection strategy with Grassmannian beamforming. The new algorithm chooses only one relay based on the maximal received SNR to assist in transmission. It can significantly reduce the system consumption, achieve excellent bit-error-rate and average throughput and maintain full diversity order. Simulation results demonstrate that the performance of the proposed schemes with low complexity approaches towards the optimal iteration method. For practical situations with a distance attenuation factor, the performance gain of the authors schemes is also evident.

A maximum mean squared error optimal relay beamformer is proposed here for two-way single-antenna distributed relaying systems. A constrained optimisation problem with respect to the relay beamforming vector is first formulated. It is then shown that the design problem of such a relay beamformer supporting both downlink and uplink transmissions simultaneously can be converted to convex optimisation when full channel state information (CSI) is available at the relays. By employing the Lagrangian multiplier method, a closed-form solution for the relaying vector is obtained. The proposed relay beamforming method is also extended to the situation where only the statistics of the CSI are available. A simulation study is conducted to confirm the merit of the proposed two-way relaying scheme.

In this paper, we first derive an optimal precoder for amplify-and-forward (AF) multiple-antenna relay systems with single-antenna source and destination terminals by maximizing the output signal-to-noise ratio (SNR). Then, the resulting maximum SNR expression and its approximation at high transmission SNR are analyzed to obtain their statistical properties for independently identically distributed (i.i.d.) Rayleigh fading channels. Based on the properties of the approximate maximum SNR, the average symbol error probability (ASEP) is investigated, leading to an explicit diversity order and array gain. The ergodic achievable rate is also studied, giving a tight closed-form upper bound. The theoretical analysis is finally validated by Monte Carlo simulations.

We consider a wireless relay network where two pairs of nodes exchange information with their partners through a single amplify-and-forward two-way relay with each node equipped with multiple antennas. We propose a new relaying scheme employing OFDMA for the multiple access transmission in the first time slot and OFDM/SDMA for the broadcast transmission in the second time slot to improve the sum rate of the network. To fully utilize spatial diversity, we design the relay beamforming matrices according to two methods respectively, i.e., signal to leakage and noise ratio (SLNR) and block diagonalization based zero-forcing (BDZF), on per subcarrier basis. We also derive the upper bound on the capacity region of this two-way relay network by using cut-set theory. Simulation results show that the proposed scheme outperforms three other relaying schemes in terms of sum rate and can approach the upper bound of capacity region.

This paper presents an optimal relay precoding scheme for two-hop amplify-and-forward based MIMO relay networks with one source, one destination and multiple relays. A constrained optimization problem for relay precoder is formulated by using the mutual information criterion along with a total relay transmitting power constraint. It is shown that due to the block-diagonal structure of the precoding matrix, the optimization problem can not be solved directly by the existing methods. We then simplify it into an equivalent problem involving scalar optimization variables only. Through an in-depth study of the problem formulation, it is revealed that the objective function is convex only when the relay power is larger than a certain threshold, which can be determined mainly by the system configurations and the channel condition. Finally, under the convex condition, the simplified problem is solved by a convex optimization method. The effectiveness of the proposed precoder is validated by Monte-Carlo simulations with comparison to some of the existing methods.

In this paper, an efficient signal-perturbation-free semi-blind approach is proposed for flat-fading channel estimation of multiple-input multiple-output (MIMO) systems with maximum ratio transmission (MRT). A novel transmit scheme is developed based on the eigenvalue decomposition of the correlation matrix of the transmitted signal. The new scheme is to send a small volume of data bearing the information of the correlation matrix to the receiver for the cancellation of the signal perturbation error. Computer simulations show that the proposed approach significantly outperforms the closed-form semiblind estimation method (CFSB) in terms of the MSE of the beamforming vector estimate.

The cascade connection of a 2ndorder Volterra filter (VF) with a linear transversal filter is often used in nonlinear acoustic echo cancellation (AEC) to model a nonlinear echo path comprising a nonlinear loudspeaker and a linear room impulse response. This cascade-form AEC, however, often suffers from divergence, since its nonlinear and linear constituents are updated separately and a global convergence toward the minimum residual echo error is hardly guaranteed. In this paper, we propose a simplified 2ndorder VF that is equivalent to the cascade structure for nonlinear AEC. The new structure has much fewer coefficients than the conventional direct-form 2ndorder VF does, while enjoys a better convergence performance compared to both the cascade canceller and the conventional 2ndorder VF. Computer simulation based experiments have confirmed the advantage of the proposed structure.

In this paper, a robust optimal distributed relay beamforming scheme is presented for two-way wireless relay systems with two sources (one base station and one mobile terminal) and multiple relays, each having a single antenna. Considering that the perfect channel state information (CSI) between the mobile terminal and the relays is usually not available, the new beamforming problem, based on the minimization of the sum MSE (mean squared error) subject to a total relay power, is formulated such that only the CSI between the base station and the relays is required. The constrained beamforming optimization problem is then solved by the Lagrangian multiplier method, leading to a closed-form solution for the distributed relay beamforming. Monte Carlo simulations show that the proposed scheme performs better than the conventional relaying method in terms of both sum rate and the bit-error-rate (BER).

Most of the existing multiple-input multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) channel estimation methods do not take into account the effect of the pulse-shaping filter in the transmitter nor of the matched filter in the receiver, thus leading to an estimation solution for the composite channel including the pulse-shaping and matched filters, instead of the pure wireless channel. This solution is not sufficiently accurate because of the extra length of the composite channel induced by the two filters especially in the scenario with a small pure channel length. In this study, the authors present a novel methodology for the estimation of the pure multipath channels of pulse-shaped MIMO-OFDM systems. By utilising the knowledge of pulse-shaping and matched filters, the authors develop two channel estimation approaches, namely, a semi-blind approach for the sampling duration-based channels, in which the multipath occurs at the sampling instant and a training-based least-square technique for the upsampling duration-based channels where the multipath may occur in a fraction of sampling duration. A number of computer simulation-based experiments are conducted, and these simulation results confirm the efficacy of the proposed approaches.

In this paper, a two-way relay precoder is designed for multiple-input multiple-output (MIMO) distributed cooperative relay systems. A constrained optimization problem with respect to (w.r.t.) the relay precoder is formulated for the most general relay and antenna scenario, namely, multiple relays each with multiple antennas, It is shown that due to the difficulty of two-way distributed relaying mechanism as well as the block-diagonal nature of the relay precoding matrix, this optimization problem cannot be solved by existing one-way relaying methods. It is then proved that with full channel state information available at relays, the underlying problem can be converted to a convex optimization w.r.t. the non-zero entries of the relay precoding matrix only, such that the Lagrangian multiplier method is applicable to the relay precoder design, leading to a closed-form relay precoding solution. A simulation study is conducted to justify the superior performance of the proposed two-way relaying scheme.

A new optimization criterion is proposed to minimize error probability for the proposed joint optimal power allocation (PA) of the MIMO systems enhanced by relay in this paper. It is proved that the cost function obtained is only convex with respect to (w.r.t.) the power parameters of the source or those of the relay separately, but not convex w.r.t. the whole parameters. In order to use convex optimization methods with high efficiency to solve this complicated problem, a tight upper bound of the sum MSE (mean squared error) is derived, and employed to modify the cost function in order to obtain a convex problem. It is verified through simulation results that the proposed PA scheme outperforms the existing one.

Formant frequency is a one of the most important speech feature, which has widespread applications in speech recognition, synthesis, and compression. In this paper, a new time-frequency domain scheme for the estimation of formant frequencies from noise-corrupted speech signals is presented. In order to overcome the adverse effect of noise, instead of conventional autocorrelation function (ACF), a repeated ACF (RACF) of the noisy speech is employed. Exploiting the characteristics of the zero lag, a set of equations containing the lower lags of the RACF of the noisy speech is used to estimate the formant frequencies. In order to avoid estimation errors that may occur in the case of weak formants, a frequency-domain algorithm is introduced utilizing the RACF of the observed speech. Formant frequency estimation accuracy is measured for different natural and synthetic vowels in noisy environments and even at low levels of signal-to-noise ratio, a better performance is obtained by the proposed scheme in comparison to some of the existing methods.

An effective method for pitch estimation from severely noise-corrupted speech is presented in this paper. We argue that an accurate estimate of a pitch-harmonic (PH) can be obtained from the Discrete Cosine Transform (DCT) power spectrum of an enhanced frame. By exploiting the PH thus extracted, a spectral matching method is developed in the DCT domain to determine the true harmonic number associated to the PH that leads to a pitch estimate. In a heavy noisy scenario, the superior efficacy of the proposed pitch estimation method relative to some of the existing ones is confirmed through simulation results using theKeeledatabase.

In this paper, a semi-blind algorithm is proposed for the detection of most significant tap (MST) in the sparse channel estimation of OFDM systems. Based on an analysis of the second-order statistics of the signal received through a noise-free sparse channel, a direct relationship between the positions of the most significant taps (MST) of the sparse channel and the lags of the nonzero correlation functions is revealled, leading to an efficient semi-blind MST detection algorithm. By using the acquired MST position, a sparse least square channel estimate is then obtained. A number of computer simulation-based experiments are carried out to confirm the effectiveness of the proposed semi-blind MST detection algorithm and the associated sparse LS channel estimation method.

The second-order statistics (SOS) of the received signal are very often used in blind and semi-blind channel estimation. In this paper, an analysis of signal perturbation in SOS of the received signal is first conducted, revealling that, even in the noise-free case, some SOS-based blind and semi-blind algorithms are subject to a signal perturbation error. Based on the analysis, a very efficient transmit scheme that can completely cancel the signal perturbation error at the receiver in the noise-free case is proposed. Computer simulations show that by employing the proposed signal perturbation cancellation approach, the mean square error (MSE) of the SOS estimate can be sufficiently suppressed in the noisy case.

The cooperative communication technology proposed in recent years enables network nodes to share their antennas to achieve diversity gain. In this paper, an efficient variation scheme on opportunistic cooperation is proposed by using an outage criterion, in which the cooperation mode will be adopted only when the channel from source to relay does not occur outage event. We derive a closed-form BER (bit error rate) expression for the proposed scheme over Rayleigh fading channels, showing that the full diversity is achieved by the new scheme. Also, the BER performance of the known coded cooperation is presented for the purpose of comparison with our scheme. Numerical results illustrate the superiority of the proposed scheme over the coded cooperation in terms of BER performance. It is pointed out that the corresponding BER advantage of the proposed scheme comes at the expense of increasing system overhead since the new scheme needs some feedback from relay to both source and destination.

In this paper, a new algorithm for pitch extraction from noisy speech signals based on both temporal and spectral representations is presented. We derive a harmonic sinusoidal correlation (HSC) model of clean speech as a temporal representation. Given only a noisy speech frame, a noise-robust least-squares minimization technique is proposed to acquire the parameters of the HSC model which are directly employed for the accurate estimation of a pitch-harmonic (PH). Exploiting the extracted PH and based on a spectral representation which is an enhanced spectrum in the discrete cosine transform domain, a two-fold criterion is developed in order to achieve the true consecutive number corresponding to PH that is finally adopted for pitch detection in the presence of noise. Simulation results using the Keele pitch extraction reference database manifest that combining the multi cues obtained from the temporal as well as spectral representations, the proposed algorithm is able to achieve a superior efficacy in comparison to some of the existing methods from high to very low signal-to-noise ratio (SNR) levels.

In this paper, a new transmit scheme combining beamforming (BF) with space-time block coding (STBC) is proposed for correlated fading channels. Based on maximizing the output mean signal-to-noise ratio (SNR) at the receiver, the transmitter BF weight vectors are first derived. Then, utilizing the minimal bit error rate (BER) upper bound as the design criterion, a simple power allocation algorithm is developed. Next, the BER performance of the system with the proposed transmit scheme is analyzed by considering an M-QAM constellation. Finally, computer simulation results are given to verify the effectiveness of the proposed scheme.

This paper presents a new algorithm for the parameter estimation of minimum-phase autoregressive moving average (ARMA) systems from noise-corrupted observations. In order to estimate the AR parameters of the ARMA system, based on a repeated autocorrelation function (ACF) of the observed data, a set of zero lag compensated equations has been developed. For the estimation of the MA parameters, first, a noise-subtraction algorithm is proposed to reduce the effect of noise from the ACF of the residual signal which is obtained by filtering the noisy ARMA signal via the estimated AR parameters. The MA parameters are then estimated by using a spectral factorization corresponding to the noise-compensated ACF of the residual signal. Computer simulations are carried out for ARMA systems of different orders under noisy environments and simulation results demonstrate a superior identification performance in terms of estimation accuracy and consistency.

A novel algorithm for pitch frequency (PF) estimation from non-stationary noise-corrupted speech observations is presented in this paper based on both spectral pre-processing and temporal representation. A modified power spectral subtraction based de-noising scheme that allows tracking the time-variation of the underlying non-stationary noise is put forward to enhance speech prior to PF estimation. The de-noised speech is then utilized to propose a squared difference function of the Linear Prediction (LP) residual which is expected to reveal more prominent dips at integral multiples of the pitch period compared to that revealed by the LP residual. The dips at different pitch-harmonic locations are added and weighted by a periodicity dependent weighting factor for every possible pitch period thus yielding a weighted and harmonically summed temporal function which is globally minimized to extract the desired PF. Simulation results using the Keele database show the superior efficacy of the proposed method in the presence of a multi-talker babble noise relative to some of the existing methods.

Most of the existing MIMO-OFDM channel estimation methods do not take into account the effect of the pulse- shaping Alter in the transmitter nor of the matched Alter in the receiver, thus leading to an estimation solution for the composite channel including the pulse-shaping and matched Alters, instead of for the pure wireless channel. This solution is neither directly applicable to practical communication systems nor sufficiently accurate due to the extra length of the composite channel induced by the two Alters. In this paper, a semi-blind channel estimation method is proposed for pulse-shaped MIMO-OFDM systems. By utilizing the knowledge of pulse-shaping and matched Alters, a time domain semi-blind estimation method is developed for the pure multi-path channel. In order to reduce the computational burden of the time-domain algorithm, a frequency-domain alternative is derived. A number of computer simulation-based experimentations are conducted, and these simulations confirm the effectiveness of the proposed semi-blind method.

In this paper, a perturbation analysis of two subspace-based semi-blind MIMO channel estimation approaches is conducted. Our analysis shows that, in the noise-free case, the whitening-rotation (WR)-based algorithm is subject to a signal perturbation error, while the nulling-based algorithm is a signal perturbation free scheme with an ideal nulling constraint imposed on the channel matrix. This explains why the WR-based method is efficient only in the low SNR case, and concludes that the nulling-based approach is better for moderate to high SNRs. A novel closed-form mean square error (MSE) expression is also derived for the nulling-based blind estimation method, in which an appealing scheme for the determination of the weighting factor is presented. The nulling-based method with the proposed weighting scheme is validated via computer simulations, showing a very high estimation accuracy of our semi-blind solution in terms of the MSE of the channel estimate.

In this paper, the lattice structure of a class of filter banks that possesses both the linear-phase property and the mirror-image symmetry, referred to as MIS-LPPRFB, has been investigated. By combining the MIS and the LP constraints and proposing a new simplified scheme, a reduced lattice structure with fewer parameters has been developed. It is shown that the introduction of the MIS has imposed certain constraints on the invertible matrices in the lattice structure of the conventional LPPRFBs, reducing the number of free parameters by nearly one-half. Using the new lattice structure, the optimization design problem of MIS-LPPRFBs is also addressed with an objective of achieving a high performance filter bank for its use in image compression coding.

The second-order statistics of a time-domain signal are very often used in blind and semi-blind channel estimation. Considering that the received signal in MIMO-OFDM systems might be corrupted in the time-domain due to some adverse factors such as frequency offset and large peak-to-average power ratio (PAPR), an IFFT processor is required in the receiver to achieve a good-quality time-domain signal. This additional IFFT incurs extra computational complexity and probably a long time delay as well in real-time communication systems. In this paper, we propose a new algorithm for the computation of the time-domain correlation matrix directly from the received frequency- domain signal. The proposed frequency-domain correlation matrix estimation method is then used to develop a new semi-blind MIMO-OFDM channel estimation approach. A number of computer simulation based experiments are conducted, confirming the effectiveness of the proposed method.

In this paper, a novel signal perturbation free transmit scheme is proposed for MIMO channel estimation. A perturbation analysis of the WR-based method is first conducted, showing that the method is subject to a signal perturbation error and therefore, its performance is very poor under the moderate to high signal-to-noise ratios (SNRs). A new transmit structure is then proposed to cancel the signal perturbation error at the receiver in order to improve the performance of the WR-based method in the high SNR case. Computer simulations show that the WR-based method with the proposed signal perturbation free transmit scheme significantly outperforms the original WR-based method as well as the training-based LS method in terms of the MSE of the channel estimate.

In this paper, the conventional Grassmannian codebook for precoding is first analyzed, showing that the performance loss caused by linear receivers was not taken into account. To tackle the performance loss issue, a novel hierarchical codebook consisting of a Grassmannian subcodebook and a perturbation subcodebook is then proposed for precoding systems with linear receivers. A two-step codeword selection scheme that uses the product of two codewords selected from the subcodebooks as the precoder is also presented. Our analysis shows that the perturbation subcodebook is able to compensate for the performance loss from linear receivers. Compared with the Grassmannian codebook, the superiority of the proposed codebook in terms of search complexity as well as throughput/ BER is further confirmed by computer simulations.

A new approach for the formant frequency estimation of the voiced speech segments in the presence of noise is presented in this paper. A correlation model for the voiced speech is proposed considering the vocal-tract system as an autoregressive moving average (ARMA) model with a periodic impulse-train excitation. It is shown that the formant frequencies can be directly obtained from the model parameters. An adaptive residue-based least-squares optimization algorithm is proposed to estimate the model parameters, which overcomes the failure of conventional correlation based techniques in estimating formant frequencies at a low signal-to-noise ratio (SNR). The proposed algorithm has been tested on synthetic and natural vowels as well as voiced segments of some naturally spoken sentences from TIMIT database in presence of white Gaussian or babble noises. The experimental results show that the proposed method is more robust to noise than some existing methods even at a low SNR of 0 dB.

In this paper, we present a robust pitch estimation algorithm for noise-degraded speech. We propose a new circular average magnitude sum function (CAMSF) and a pseudo normalized correlation function (PNCF) both of which exhibit the periodicity at the pitch period of voiced speech. Exploiting the fact that CAMSF produces a peak while PNCF shows a notch, an integrated time-domain function (ITDF) is developed to enhance the pitch-harmonic-notches in presence of noise. Moreover, a frequency-frame relative smoothed noisy spectrum that acts as a harmonic spectral structure enhancer is utilized to accurately acquire a pitch-harmonic (PH) from noisy speech. We argued that employing the PH, pitch information can be effectively extracted through a variable-period impulse-train in conjunction with the proposed ITDF. It has been ascertained that the overall algorithm simulated using the Keele reference database is able to outperform some of the existing methods and well suited for a wide range of signal-to-noise ratios (SNRs) upto-10 dB.

In this paper, an identification technique for the minimum-phase autoregressive moving average (ARMA) systems using only the noise-corrupted observations is presented. In order to obtain a more accurate estimate of the AR parameters in the noisy environment, a repeated autocorrelation function (RACF) of the observed data is employed in the modified least-squares Yule-Walker equations. It has been found that at a very low signal-to-noise ratio (SNR), the effect of the additive noise can be significantly reduced if a twice-RACF is employed instead of the conventional ACF. Prior to the MA part identification, a noise-compensation scheme is proposed which operates on the noise-contaminated residual signal. The MA parameters are extracted from the noise-compensated power spectrum of the residual signal using the spectral factorization. ARMA systems of different orders and some natural speech signals are tested and computer simulations demonstrate a superior identification results even at a very low SNR.

A two-step algorithm for the voiced/unvoiced (V/UV) decision of colored noise-corrupted speech is presented in this paper. An effective noise-whitening process is first applied to the noisy speech to combat the adverse effect of colored noise. Then, a harmonicity measure is proposed which is derived from the LP residual of the pre-whitened speech. Integrating root-mean-square energy and zero-crossing rate, another composite measure is introduced. In the first step, signal-dependent initial-thresholds (SDITs) for both the measures which are capable of highlighting distinctive attributes of voiced and unvoiced frames, are determined analyzing their statistical properties. In the second step, based on the SDITs, a bi-feature logical target function is formulated to attain a preliminary score of V/UV decision. Additional voicing criteria are developed to conquer the artifacts that may exist due to the overlapping between decision regions. Simulation results demonstrate that the proposed algorithm yields superior performance in comparison with some of the existing V/UV decision schemes in the same interfering colored noise scenario.

In this paper, a semi-blind channel estimation method is presented for MIMO-OFDM systems. The new method uses the linear prediction for obtaining a blind constraint on the MIMO-OFDM channel matrix as well as the least-squares approximation for the training signal. The proposed method can be regarded as an extension of an existing semi-blind MIMO channel estimation algorithm. Yet the extension is nontrivial, since the formulation of the MIMO-OFDM signal and the related blind constraint cannot easily be obtained from the MIMO counterpart. The proposed algorithm is simulated using Monte-Carlo method and compared with the LS method in terms of the mean square error (MSE) of the estimation. Simulation results show that the proposed method consistently outperforms the LS method when the same training signal is used.

In this paper, several new cosine-modulated filter banks (CMFBs) are developed by using multiple prototype filters in conjunction with a proper modulation scheme. First, the conventional CMFB is formulated as a cascade of a modulation matrix and a polyphase matrix with a bidiagonal structure. Under this framework, it is then revealed that the proposed new CMFBs are a special class of paraunitary filter banks (PUFBs), whose polyphase matrices are of different sparse patterns. It is shown that when M prototype filters are used, a full polyphase matrix representing a general PUFB can be obtained, thus providing a bridge connecting CMFBs and PUFBs. It is also shown that with more free parameters involved in the CMFBs, one can achieve a tradeoff between the performance of the filter bank and its design/implementation complexity

The problem of composite hypothesis testing where the probability law governing the generation of the free parameter is not explicitly known is considered. It is shown that unlike the Neyman-Pearson (NP) approach, the competitive NP (CNP) approach models incomplete prior information about the source into the detector design by setting a variable upper bound for the probability of false-alarm term. Further, the CNP and NP approaches are employed to develop the CNP and NP detectors for voice activity detection (VAD), where the prior SNR is shown to be the free parameter of the composite hypothesis. We test the CNP and NP detectors using speech samples from the SWITCHBOARD database which are suitably corrupted using different noises and various SNRs. Our simulation results show that the CNP detector outperforms its NP counterpart and is comparable to the adaptive multi-rate (AMR) VADs

In this paper, a new model for the ramp-cepstrum of the one-sided autocorrelation function of a noise-free autoregressive moving average (ARMA) signal is presented. The proposed blind identification technique can estimate the parameters of ARMA systems in both noise-free and noisy environments without using the input observations. It is shown that, utilizing the proposed ARMA ramp-cepstrum model in accordance with a residue-based least-squares optimization technique, both AR and MA parameters of ARMA systems can be directly obtained. The proposed method is tested on synthetic ARMA systems of different orders and also on some natural speech signals. Simulation results demonstrate the efficacy of the proposed identification scheme at low to high SNR levels

This paper presents a new algorithm for the voiced/unvoiced (V/UV) decision of noise-corrupted speech. A speech periodicity-harmonic function (SPHF) is proposed to manifest distinctive characteristics between voiced and unvoiced regions. A composite feature vector is developed by combining a periodicity measure obtained from the SPHF with some energy measures such as zero-crossing rate-weighted RMS energy, Kaiser-Teager frame energy and the normalized low-frequency energy ratio. Unlike the conventional hard threshold, a signal-dependent initial-threshold (SDIT) for each feature is determined based on its statistical properties. The SDIT is exploited to develop a logical expression that returns an objective score regarding V/UV region. Additional voicing criteria are introduced to remove the artifacts that may exist due to the overlapping between decision regions. Simulation results of the proposed multifeature classification scheme, using the Keele reference database, show superior efficacy at a low SNR relative to some of the existing V/UV decision algorithms

In this paper, a generalized polyphase (GP) structure based design method for Mth-band FIR filters is presented. A few new transform matrices are proposed as seed matrix for generating a GP structure, which overcomes some limitations of the commonly used Hadamard transform. Under the framework of a general GP structure realization, the relationship between the original filter and the interpolator as well as the constituent filter is revealed from the frequency response perspective. This relationship facilitates the design problem of GP structure based FIR filters, since the original large-tap filter is reduced to a number of short-length constituent filters which can easily be designed according to the derived specifications. In particular, a closed-form frequency specification for the design of Mth-band filters is obtained. Design examples are given to illustrate the effectiveness of the proposed method.

In this paper, a realization scheme that combines the singular-value decomposition (SVD) and the generalized polyphase (GP) structure is proposed for 2D linear-phase FIR filters. With a small number of extra additions, a high-order 2D FIR filter is converted to several lower-order 2D subfilters. These subfilters are then realized using the SVD, yielding a parallel implementation structure for each 2D subfilter which consists of a number of 1D short-tap FIR filters. Due to the energy compaction of the SVD and the frequency-selective property of the GP structure, the number of parallel branches in each 2D subfilter is significantly reduced without introducing a large error. It is also shown that the various symmetries of 2D filters, such as the quadrantal symmetry and the central symmetry, are well preserved in the proposed GP-SVD structure.

A new approach for the identification of minimum-phase autoregressive moving average (ARMA) systems in the presence of heavy noise is presented in this paper. A damped sinusoidal (DS) model for the autocorrelation function of a noise-free ARMA signal is proposed to estimate the AR parameters, which overcomes the failure of conventional correlation based techniques in estimating the AR parameters of an ARMA system at a very low signal-to-noise ratio (SNR). The MA parameters of the ARMA system are then estimated by using Durbin's method along with an optimum order selection criterion. Both white noise and periodic impulse train excitations are considered for the application of the proposed method to system identification as well as to speech processing. Computer simulations are carried out based on both synthetic ARMA systems and natural speech signals, showing superior identification results even at an SNR of -5 dB for which most of the existing methods would fail.

In this paper, we present a joint time/frequency domain approach for pitch estimation of speech at a very low SNR. The kernel of this approach lies in introducing a new function for detecting the time-domain cue by modifying the circular average magnitude difference function (CAMDF). By using the new function in conjunction with the half-wave rectified version of the autocorrelation function, the pitch-peak can be emphasized and the non-pitch peaks suppressed. To guarantee a robust pitch detection in noisy speech, a priori frequency-domain estimate of the dominant pitch-harmonic is extracted as an additional cue and is utilized to optimally match the pitch-peak in time-domain. The proposed approach is simulated using the Keele reference database. It is shown that the proposed method using joint time and frequency domain cues is able to give a superior accuracy relative to some of the existing methods even at a very low SNR of -10 dB.

A method for using an arc array to resolve coherent sources is presented. A narrowband plane wave received by sensors of an arc array is analyzed. By introducing a sensor-dependent delay to each sensor, the arc array problem is approximately mapped to an equispaced linear array under certain constraints. Combining this with an existing eigen-assisted algorithm for high-resolution array processing, a method for arc array resolution of coherent sources is proposed. The algorithm needs no spatial smoothing and hence no reduction of array aperture is introduced. Computer simulation results are presented to show the performance of the method.>

Analytic results on the performance of eigen-assisted methods in the presence of two correlated sources are presented. Based on the assumption that two narrowband sources are incident on an equispaced array, the two largest eigenvalues of the array correlation matrix, and their corresponding eigenvectors, are derived. The effects of correlation of sources and signal-to-noise ratio on the eigenvalue distribution and signal detection are studied. The effects of these factors on the two corresponding eigenvectors and the resolution of the eigen-assisted processors are examined. When correlation between the sources is not strong, the performance of the processor is shown to be affected by correlation very slightly. As a special case, it is proved that those methods reduce to classical delay-and-weight (or DFT) methods when the two sources are coherent (if no preprocessing is applied). This analytical result is then used to explain the well-known experimental phenomenon that when sources are coherent, the performance of these processors degrades severely.>

A method is presented for the design of 3-D recursive digital filters with approximately spherically symmetric magnitude frequency response. The desired 3-D recursive filter is obtained by cascading a number of 3-D transformed filters which can be designed by rotating a 1-D analog filter and using the triple bilinear z-transformation. The stability of the 3-D filters is discussed in detail, and a theorem which gives the sufficient conditions for stable 3-D filters is derived.>