Direction-of-arrival (DOA) estimation for incoherently distributed (ID) sources is crucial for Industrial Internet of Things (IIoT) applications operating in complex multipath environments, yet it remains challenging due to the combined effects of angular spread and gain-phase uncertainties in cost-sensitive antenna arrays. This paper presents a two-stage sparse DOA estimation framework, transitioning from partial calibration to full potential, under the generalized array manifold (GAM) framework. In the first stage, coarse DOA estimates are obtained by exploiting the output from a subset of partly-calibrated arrays (PCAs). In the second stage, these estimates are utilized to determine and compensate for gain-phase uncertainties across all array elements. Then a sparse total least-squares optimization problem is formulated and solved via alternating descent to refine the DOA estimates. Simulation results demonstrate that the proposed method achieves superior estimation accuracy compared to existing approaches, while maintaining robustness against both noise and angular spread effects in practical industrial environments.

Fluid antenna (FA) technology has emerged as a promising approach in wireless communications due to its capability of providing increased degrees of freedom (DoFs) and exceptional design flexibility. This paper addresses the challenge of direction-of-arrival (DOA) estimation for aligned received signals (ARS) and non-aligned received signals (NARS) by designing two specialized uniform FA structures under time-constrained mobility. For ARS scenarios, we propose a fully movable antenna configuration that maximizes the virtual array aperture, whereas for NARS scenarios, we design a structure incorporating a fixed reference antenna to reliably extract phase information from the signal covariance. To overcome the limitations of large virtual arrays and limited sample data inherent in time-varying channels (TVC), we introduce two novel DOA estimation methods: TMRLS-MUSIC for ARS, combining Toeplitz matrix reconstruction (TMR) with linear shrinkage (LS) estimation, and TMR-MUSIC for NARS, utilizing sub-covariance matrices to construct virtual array responses. Both methods employ Nyström approximation to significantly reduce computational complexity while maintaining estimation accuracy. Theoretical analyses and extensive simulation results demonstrate that the proposed methods achieve underdetermined DOA estimation using minimal FA elements, outperform conventional methods in estimation accuracy, and substantially reduce computational complexity.

This paper investigates a design framework for sparse fluid antenna systems (FAS) enabling high-performance direction-of-arrival (DOA) estimation, particularly in challenging millimeter-wave (mmWave) environments. By ingeniously harnessing the mobility of fluid antenna (FA) elements, the proposed architectures achieve an extended range of spatial degrees of freedom (DoFs) compared to conventional fixed-position antenna (FPA) arrays. This innovation not only facilitates the seamless application of super-resolution DOA estimators but also enables robust DOA estimation, accurately localizing more sources than the number of physical antenna elements. We introduce two bespoke FA array structures and mobility strategies tailored to scenarios with aligned and misaligned received signals, respectively, demonstrating a hardware-driven approach to overcoming complexities typically addressed by intricate algorithms. A key contribution is a light-of-sight (LoS)-centric, closed-form DOA estimator, which first employs an eigenvalue-ratio test for precise LoS path number detection, followed by a polynomial root-finding procedure. This method distinctly showcases the unique advantages of FAS by simplifying the estimation process while enhancing accuracy. Numerical results compellingly verify that the proposed FA array designs and estimation techniques yield an extended DoFs range, deliver superior DOA accuracy, and maintain robustness across diverse signal conditions.

This paper addresses reconfigurable intelligent surface (RIS)-assisted indoor localization and synchronization in the presence of multipaths. Considering the far field condition, direct range estimation becomes infeasible and there exists clock offset in the system, compounding the difficulty for a single RIS to accomplish user equipment (UE) positioning and synchronization. To tackle this challenge, a dual-RISs system is proposed. The extra degrees of freedom it offers can effectively resolve the problem It uses initial RIS phase design to separate components from dual RISs. Then, atomic norm minimization is employed to reconstruct the separated received signals, and 2D-MUSIC with single-snapshot estimates the angles-of-departure (AODs) of the UE and scatterers. After removing angle terms, root-MUSIC estimates the time-of-arrival (TOA). The UE’s position is derived via least squares using AODs from dual RISs. Combining the UE’s position with LOS path TOAs yields the clock offset. Scatterer positions are obtained using geometric relationships with the UE’s position, NLOS path TOAs, and clock offset. Channel parameters are refined via maximum likelihood estimation. Simulation results prove the method’s effectiveness.

Based on the exact spherical wavefront model, an under-determined estimation method for three-dimensional (3-D) parameters of near-field (NF) sources using L-shaped nested arrays is proposed, referred to as the cumulant algorithm. This algorithm leverages the temporal-spatial domain cumulants of NF sources by constructing virtual data through delayed fourth-order cumulant (FOC) calculations of the original received data. Subsequently, a spatial-spectrum-based subspace method is applied for 3-D NF localization, which involves a 3-D spectral search procedure. Additionally, the maximum number of identifiable NF sources of the proposed algorithm is analyzed. Simulation results demonstrate that, based on the exact spherical wavefront model, the proposed algorithm can achieve underdetermined 3-D parameter estimation of NF sources without any matching process, and it performs better in localization than existing methods.

This paper addresses target localization using a multistatic multiple-input multiple-output (MIMO) radar system with coprime L-shaped receive arrays (CLsA). A target localization method is proposed by modeling the observed signals as tensors that admit a coupled canonical polyadic decomposition (C-CPD) model without matched filtering. It consists of a novel joint eigenvalue decomposition (J-EVD) based (semi-)algebraic algorithm, and a post-processing approach to determine the target locations by fusing the direction-of-arrival estimates extracted from J-EVD-based C-CPD results. Particularly, by leveraging the rotational invariance of Vandermonde structure in CLsA, we convert the C-CPD problem into a J-EVD problem, significantly reducing its computational complexity. Experimental results show that our method outperforms existing tensor-based ones.

The wideband bistatic radar problem for automotive applications is addressed, and the localisation task is decomposed into three stages where Doppler is firstly estimated, followed by range, and direction. Group sparsity is applied to estimate the Doppler frequencies of all targets. Then, for each target, the estimated Doppler parameters are used to obtain the range parameters from sparse signals that have been compensated for range walk. By removing the effect of range walk, and introducing a CLEAN-like approach, the range and Doppler parameters for each target are then successfully paired. Finally, the direction-of-arrival is estimated through group sparsity and the target parameters are paired via a maximum-likelihood-based approach. Simulations show that significant enhancement in estimation performance can be achieved using the proposed method in comparison with a narrowband approach.

In this paper, a high-precision three-dimensional (3-D) near-field (NF) localization method is proposed under an underdetermined case based on a symmetric enhanced nested array (SENA). Firstly, the symmetry of the array and the fourth-order cumulant (FOC) are utilized to construct the equivalent virtual far-field (FF) reception data. Then, a gridless sparse and parametric approach (SPA), combined with an l1-SVD based pairing procedure, is used to obtain estimates for two paired angles. Finally, a one-dimensional (1-D) spectral estimator is applied to obtain the estimate of range parameter. Simulation results show the effectiveness of the proposed method.

Most existing direction of arrival (DOA) estimation methods are based on antenna arrays for line-of-sight (LOS) propagation. In this article, a different and challenging DOA estimation problem with a single receiving antenna in the non-line-of-sight (NLOS) scenario is addressed, where a reconfigurable intelligent surface (RIS) is combined with two robust array spatial covariance matrix (SCM) reconstruction schemes to solve the problem. In detail, a two-stage approach for high-efficiency RIS phase shifting is first designed, and then the Tikhonov regularization criterion and total least-squares (TLS) criterion are respectively exploited for SCM reconstruction with and without phase shift error (PSE), yielding an improved DOA estimation performance with reduced complexity. Theoretical analysis on the performance of SCM reconstruction is conducted, and simulation results are provided to show the effectiveness of the proposed solutions.

The digital elevation model (DEM) reconstruction accuracy of single-channel interferometric synthetic aperture radar (SC-InSAR) is limited by the SAR side-looking imaging geometry, decorrelations, phase unwrapping (PU), and so on. With the availability of increasing InSAR data, to overcome the limitations of SC-InSAR, a multi-source InSAR DEM reconstruction framework based on a complexity factor is proposed in this article. To simultaneously take the effects of noise level and terrain slope into account, a complexity factor for each interferometric pair is constructed. Next, to reduce the PU failure rate for each pair, this factor is used to guide the two-stage programming approach (TSPA) PU method. Then, to avoid the adverse effects of PU failure on elevation fusion, unreliable pixels of each pair are detected by exploiting the complexity factor. Finally, after multiple elevations from different side-looking directions are obtained, the elevation-weighted fusion is performed to reconstruct the final DEM in the map projection coordinate system. Experimental results on real multi-source InSAR data demonstrate that the complexity factor can effectively guide the steps of TSPA PU, detection of unreliable pixels, and elevation-weighted fusion in the proposed framework, thereby improving the DEM reconstruction accuracy for mountainous areas with complex and steep terrain.

Scene matching involves establishing correspondences between multiple images of the same location and poses significant challenges for synthetic aperture radar (SAR) images due to the anisotropic scattering prosperities of SAR targets; variations in looking and azimuth angles further complicate the matching process. A matching algorithm is proposed based on a coarse-to-fine framework to address these issues. First, a coarse matching employing normalized cross correlation (NCC) with a sliding window is applied to filter out irrelevant regions, reducing distractions, and shortening the processing time. Subsequently, a Siamese neural network (SNN), incorporating ResNet-50 and convolutional block attention module (CBAM) for enhanced feature extraction, is introduced to learn and discern differences between inputs. The effectiveness and robustness of the proposed method are validated through extensive experiments using a self-made dataset derived from Umbra Satellite.

Most of existing direction of arrival (DOA) based vehicle positioning techniques are established on array sample covariance matrix and multiple measurement data, which suffer from severe performance degradation in case of a single snapshot. In this paper, a challenging vehicle positioning scheme based on single-snapshot DOA and impinging signal magnitude-phase estimation is proposed. In detail, DOA is initially estimated by applying the generalized approximate message passing combined with belief propagation (GAMP-BP) algorithm under the assumption of complex discrete random variable with distinct phase information. Depending on the initial DOA estimates, two efficient approaches are respectively investigated for final DOA and signal magnitude-phase estimation, where the refine-grid GAMP combined with the least squares algorithm (GAMP-LS) and the special reweighted sparse total least-squares (SRE-STLS) are respectively adopted. With available DOA and magnitude-phase estimates, a principle for selecting reliable DOA sets is further designed, finally enabling improved vehicle positioning without ambiguity under multiple collaborative road side units (RSUs). Simulations are performed to show the effectiveness of the proposed solution.

Most near-field (NF) localization algorithms cannot deal with the underdetermined case, while those which can are computationally expensive due to employment of fourth-order cumulants. In this work, a low-complexity solution is provided for underdetermined three-dimensional (3-D) NF localization, by employing second-order statistics with a tailored array configuration named two parallel centrally symmetric unfold coprime (TPSC) array. Its implementation can be divided into three stages. Firstly, the proposed algorithm constructs two cross-correlation matrices based on the received array data, which eliminates the non-linear range-related information of NF signals. Secondly, covariance and vectorization operations are applied to these two cross-correlation matrices to form a virtual array with extended aperture. Finally, the two-dimensional (2-D) angle parameters are estimated by the sparse and parametric approach (SPA) and a phase retrieval operation, and then the one-dimensional (1-D) range parameter is achieved by the multiple signal classification (MUSIC) algorithm. One specific feature is that the estimated angle and range parameters are matched automatically. An analysis of the properties of the TPSC array is provided, and an optimal parameter configuration is derived, given that the total number of array elements is fixed. Simulation results demonstrate that the designed TPSC array can achieve underdetermined 3-D NF localization, and deliver enhanced estimation capabilities, surpassing those of established algorithms.

In this paper, we analyze the effect of sensor placement to the achievable number of degrees-of-freedom (DOFs) when the sensors deviate from a half-wavelength grid. More specifically, we consider two variations of a uniform linear array (ULA), namely, when one or more sensors are shifted from half-wavelength grid positions and when the inter-element spacing of the ULA is smaller than a half-wavelength. The numerical rank and the rank-revealing QR factorization of the array data covariance matrix are examined and the number of DOFs of the array is studied in terms of the rank of the array data covariance matrix. A threshold based on the rank-revealing QR factorization is proposed to separate the eigenvalues respectively corresponding to the signal and noise subspaces, and thus the numerical rank of the array data covariance matrix is estimated. Simulation results are provided to justify the findings and provide insights on sensor placements to preserve the array DOFs.

In this paper, a deep learning (DL) based approach is investigated for direction-of-arrival (DOA) estimation, where large-scale uniform linear arrays (ULAs) and small number of samples are considered. Different from existing DL based DOA estimators, the proposed solution first exploits the Toeplitz prior of array covariance matrix and the linear shrinkage technique to obtain an enhanced sample covariance matrix (SCM), which is then formulated as a sparse linear representation (SLR) problem. Finally, a suitable deep convolution network (DCN) that learns such a SLR characteristic from large training dataset is designed. With aid of Toeplitz and sparse prior, the proposed solution can provide an increased resolution and estimation accuracy under the considered scenario, as verified by simulations.

Based on the exact source-sensor spatial geometry, a three-dimensional (3-D) spatial-temporal localization algorithm for multiple near-field (NF) sources is proposed without adopting the Fresnel approximation, which simplifies the spatial phase difference by Taylors polynomial. In addition, considering the propagation attenuation which varies from different sensors, the spatial and temporal information can be exploited to construct a third-order parallel factor (PARAFAC) data model and the array manifold matrices can be extracted by trilinear decomposition; then, estimation of the unambiguous range and angle parameters of the NF sources is achieved from the spatial amplitude-phase factors by the least squares method. The obtained 3-D parameters associated with each source require no additional pairing process, as also demonstrated by simulation results.

In this paper, based on sum-difference co-array analysis, a new fourth-order sparse array called sum-difference-FODC (SD-FODC) is proposed, allowing the construction of a fourth-order DCA with long consecutive lags using the continuous segments in the second-order DCA and SCA of the original array. It has a closed-form expression for sensor positions that can be generated by two arbitrary nonuniform linear arrays (NLAs) called generators. If the second-order SCA and DCA of the generators have long consecutive segments, the designed fourth-order sparse array can achieve a large number of uDOFs. Numerical results are provided to demonstrate the superior performance of the proposed design.

Hybrid precoding is considered in wideband mm-Wave massive MIMO-OFDM systems with beam squint. Traditional wideband hybrid precoding schemes cannot achieve near-optimal sum rate as digital precoding/beamforming (DBF) and may induce high hardware cost. Dynamic time-domain weighting hybrid precoding (DTW-HBF) updates the analog weights during an OFDM symbol, realizing equivalent frequency-dependent analog precoding and approximating DBF with a low cost. Directly reconstructing the time-domain signals, however, involves pseudo-inverse operations, which may cause numerical instability. In this work, the frequency-domain spectrum is reconstructed by introducing the optimal frequency-domain analog precoder and using the cyclic convolution property of Discrete Fourier Transform (DFT). The proposed method can approximately approach the performance of DBF while maintaining the hardware structure based on phase shifters (PSs). As shown by simulation results, an increased sum rate has been achieved.

In this paper, an underdetermined three-dimensional (3-D) near-field source localization method is proposed, based on a two-dimensional (2-D) symmetric nonuniform cross array. Firstly, the fourth-order cumulant of the near-field observations with multiple delay lags is exploited to construct virtual far-field pseudo-observations, leading to increased degrees of freedom (DOF); then, 2-D angles of the nearfield sources are jointly estimated by employing the recently proposed sparse and parametric approach (SPA) and Vandermonde decomposition technique, eliminating the need for parameter discretization. To estimate the range term, the one-dimensional (1-D) MUSIC algorithm is applied by resorting to the conjugate symmetry property of the signal's autocorrelation function. Numerical results are provided to demonstrate the superiority of our method.

Although multi-beam multiplexing can be implemented merely by phase shifters with hybrid beamforming configured by the sub-connected subarray architecture since all the antennas share the same magnitude, they cannot be set to a predetermined value. To tackle this issue, a non-convex constraint to enforce the magnitudes to a fixed value is first introduced in this design and then an iterative method is employed to relax it into a convex one. In doing so, the weighting magnitudes of all antennas can be preset in advance according to given requirements and a more flexible solution with phase-only excitation is obtained for multi-beam multiplexing. Numerical results are presented to verify the effectiveness of the proposed approaches.

Baseline errors is the main error source of airborne multi-baseline SAR tomography. P-band SAR can penetrate into deep vegetation layer even in tropical forests and therefore offers huge potentials in forest structure study. This paper introduces a novel method to estimate and compensate these baseline errors based on small baseline subset interferometric network, which is, compared to the existing methods, (1) less prone to heavy decorrelation noise induced by forest volume scattering and (2) easy to implement without pixel-by-pixel optimization. Numerical experiments conducted on real airborne P-band multi-baseline SAR dataset demonstrate that the proposed method can effectively estimate and correct the baseline errors.

Most DOA estimation techniques require phase information of the received array signals to accurately estimate the direction of arrival (DOA). Nevertheless, in some scenarios, the phase information may not be easily accessible or reliable due to various reasons such as hardware limitations or calibration issues. One way to tackle this challenge is to discard the phase information or only measure the magnitude of the received signals. In this work, a deep-learning (DL) based DOA estimation method is proposed for effective DOA estimation with magnitude-only measurements. To improve the generalization ability and robustness of the proposed solution, an attention mechanism is employed, and to avoid the implicit assumption that the number of signals is known in the training process, labels of the data are converted into the one-hot form. Simulation results show that the proposed solution has superior performance in terms of computational complexity, accuracy, and robustness compared to traditional DOA estimation algorithms.

A tensor-based wideband direction of arrival (DOA) estimation method is proposed in this paper. Virtual arrays are initially generated and extended into a unified ULA across all frequencies of interest. Next, the covariance matrices of these virtual array models are computed and stacked to form a three-dimensional tensor. Then, tensor completion with a denoising step is presented, and DOAs can be obtained after tensor decomposition. Simulations demonstrate that improved performance can be achieved by the proposed method in scenarios with varying source power ratios across frequencies.

Two-dimensional (2-D) direction of arrival (DOA) estimation with a circular array based on the difference co-array has attracted considerable attention in past years. In this paper, the difference co-array position set of a circular array with arbitrary sensor arrangement is derived, and condition under which the maximum number of virtual co-array sensors can be provided by a circular array is presented. Then, an augmented uniform circular array (AUCA) is proposed, providing the maximum number of DOFs for arbitrary number of physical sensors. Compressive sensing based focusing method for the one-dimensional case is extend to 2-D wideband DOA estimation, where focusing on the difference co-array is adopted for performance improvement. Simulations show that better performance can be achieved by our proposed AUCA.

Recently, one-bit direction of arrival (DOA) estimation has received significant attention due to its low cost and low implementation complexity, while still achieving high accuracy without the need of high-resolution measurements. In this work, we consider nonlinear estimation errors under finite number of snapshots in one-bit covariance reconstruction, and propose the two-step reconstruction approach, first using the arcsine law to reconstruct the unquantized covariance matrix and then incorporating the Toeplitz Hermitian structure as prior information to reconstruct the full-scale virtual uniform linear array (ULA) covariance matrix. It is shown that the error is smaller than the case when the two steps are swapped in order. Simulation results demonstrate the large difference in performance due to the order in which the two steps are applied, our proposed method has remarkably outperformed the current state-of-the-art solutions.

This article studies an intelligent reflecting surface (IRS)-empowered wireless-powered communication network (WPCN) in Internet of Things (IoT) networks. In particular, a power station (PS) with multiple antennas uses energy beamforming to enable wireless charging to multiple IoT devices, in the downlink wireless energy transfer (WET) phase; then, during the uplink wireless information transfer (WIT) phase, these IoT devices utilize the harvested energy to concurrently transmit their individual information signal to a multiantenna access point (AP), which equips with multiuser decomposition (MUD) techniques to reconstruct the IoT devices’ signal. An IRS is deployed to improve the energy collection and information transmission capabilities in the WET and WIT phases, respectively. To examine the performance of the system under study, we maximize the sum throughput with the aim of jointly designing the optimal solutions for the active PS energy beamforming, AP receive beamforming, passive IRS beamforming, and time scheduling. Due to the multiple coupled variables, the resulting formulation is nonconvex, and a two-level scheme to solve the problem is proposed. At the outer level, a 1-D search method is applied to find the optimal time scheduling, while at the inner level, an iterative block coordinate descent (BCD) algorithm is proposed to design the optimal receive beamforming, energy beamforming, and IRS phase shifts. In particular, the receive beamforming part is designed by considering the equivalence between sum rate maximization and sum mean square error (MSE) minimization, thereby deriving a closed-form solution. Furthermore, we alternately optimize the energy beamforming and IRS phase shifts using Lagrange dual transformation (LDT), quadratic transformation (QT), and alternating direction method of multipliers (ADMMs) methods. Finally, numerical results are presented to showcase the performance of the proposed solution and highlight its advantages compared to some typical benchmark schemes.

Synthetic aperture radar (SAR) working at the terrain observation with progressive scan (TOPS) mode can obtain wide-coverage images. The improvement of image coverage makes further demands for higher radiation accuracy. However, traditional imaging algorithms do not take into account the effects of errors introduced by beam rotation, resulting in poor radiometric deterioration and distortion of image quality. This letter focuses on the specific manifestations of this phenomenon and the reasons for its formation. To improve the TOPS SAR image quality, a method based on the generalized cross correlation (GCC) of subsignals is presented to estimate the parameters of beam rotation in this letter. Experimental results with real spaceborne TOPS SAR data are provided to validate the analysis of this phenomenon and show that the proposed method improves the radiation accuracy for more than 0.2 dB.

Traditionally, synthetic aperture radar (SAR)-based ship target detection is performed in the image domain, where SAR imaging processing has to be applied first. However, SAR imaging processing is complex and time-consuming, especially in the wide-swath working mode. Actually, for open sea scenes, most echoes are sea surface signals with no ship targets, and there is no need for imaging processing in those areas. Therefore, non-image domain ship target detection is studied in this letter, and a novel Incept-text convolutional neural network (TextCNN) model is proposed for ship target detection in the SAR range-compressed domain (RCD). In the proposed method, the SAR echo data are converted into a 1-D range profile signal first by range compression and mean pooling, and then, the Incept-TextCNN model is proposed and applied, and information about existence of ship targets in relevant range cells will be its output. Finally, the effectiveness and efficiency of the proposed method is testified by simulation and real spaceborne SAR data, and the results demonstrate that the proposed model can filter out the invalid range-compressed data of the sea surface area, which can significantly reduce the amount of data for subsequent SAR imaging and ship classification.

Processing high-squinted spaceborne SAR data in the sliding-spotlight mode is a challenging task due to azimuth spectral aliasing and range-azimuth coupling for frequency-domain imaging algorithms, and most critically, the variation of Doppler parameters causes significant reduction in the depth-of- azimuth-focus (DOAF). In this paper, a novel imaging algorithm is proposed for focusing high-squinted spaceborne SAR data in the sliding-spotlight mode. First, linear range walk correction (LRWC) and range frequency-dependent de-rotation are applied to remove the coupling of range frequency with the Doppler parameters. Then, a modified range migration algorithm (RMA) is derived for accurate focusing. The de-ramp operation combined with improved nonlinear chirp scaling (INCS) is employed for solving the aliasing problem of azimuth time and extending the depth-of-azimuth-focus in the third step. Finally, geometry distortion caused by LRWC is corrected. Simulation results are provided to demonstrate the effectiveness of the proposed algorithm.

Multibaseline interferometric synthetic aperture radar (MB-InSAR) has attracted widespread attention as it can improve the measurement accuracy of elevation or deformation by exploring baseline diversity. However, the interferometric phase is normally contaminated by phase noise, which directly affects the measurement accuracy. In this article, an MB-InSAR interferometric phase noise suppression method based on BM4D (MBInSAR-BM4D) is proposed. To increase the number of similar cuboids for grouping, a topographic phase compensation strategy is introduced, which can reduce fringe density in complex topography. In addition, to accurately select similar cuboids from residual interferometric phase stack and collect them into 4-D groups, the generalized likelihood-ratio (GLR) test is applied, in which the observed amplitude, coherence, and phase are utilized simultaneously for improving the grouping accuracy. After performing collaborative filtering and aggregation on 4-D groups, the MB-InSAR interferometric phase stack noise suppression results are obtained by adding the reference phase back to the corresponding filtered residual interferometric phase. All interferometric phases are fully exploited to facilitate MB-InSAR phase stack filtering performance enhancement. Experimental results on both the simulated and real MB-InSAR data demonstrate that the proposed MBInSAR-BM4D provides superior noise suppression and fringe detail preservation for the MB-InSAR interferometric phase stack.

The ambiguity function (AF) is a crucial tool in characterizing the range-angle response of a multiple-input multiple-output (MIMO) radar system, which is intricately influenced by the transmit waveforms, receiving filters and also antenna configurations. Notably, the role of antenna configurations is less explored compared to the well-studied areas of waveforms and filters. In this article, we incorporate antenna positions as an additional design parameter alongside waveforms and filters to optimize the AF in a specific range-angle bin. We employ the mainlobe-to-integrated-sidelobe-level-ratio (MISLR) as a quantitative metric to assess the performance. The resulting optimization problem is inherently nonconvex, encompassing binary and unimodular constraints. To address this challenge, we reformulate the problem, enabling an alternating optimization approach for antenna positions and waveforms. Each iteration involves solving a sequence of quadratic constrained quadratic programming problems for the binarily constrained antenna position optimization and updating the waveforms iteratively via an analytical expression. Our simulation results validate the effectiveness of the proposed method as it achieves higher MISLR with the same number of antennas compared to conventional approaches. Moreover, the optimized antenna configurations notably enhance the balance between angular ambiguity and resolution.

The carrier frequency and bandwidth are vital parameters of radar transmit signals. In this article, a joint design of frequency and bandwidth in multifrequency synthetic aperture radar (SAR) is proposed to improve the target information acquisition capability. First, the target detection mutual information (MI) is considered as the performance metric, and its mathematical expression is derived using the example of decision-level fusion and hypothesis testing. The design is formulated as an optimization problem with multiple engineering constraints based on MI maximization. Then, a modified genetic algorithm (GA) is proposed to find the optimal solution satisfying the constraints via a code adjustment operator. It is shown by simulation results that for a specific scene, the multifrequency SAR with its frequencies and bandwidths designed by the proposed method has higher target information acquisition capability and more accurate target detection performance than existing spaceborne SAR systems.

The ScanSAR mode can change the antenna angle during operation and obtain a wide swath by scanning multiple strips at one time. However, due to discontinuous working in azimuth, the scalloping effect in ScanSAR will degrade the image quality. In this paper, a novel adaptive scalloping suppression method is proposed by analyzing the complex scene as well as the scalloping distribution. First, the images of the various types of scenes are pre-processed so that the distribution of sub-images satisfies the Kalman filter conditions. Then, the problem of space-variant property is solved by performing adaptive blocking in the range direction. Finally, the Kalman filtering algorithm is introduced to process the scalloping in each sub-block separately, and the processed sub-blocks are fused to obtain the final result. The proposed method is verified by the real ScanSAR images of GF-3. Experimental results show that the proposed method is more efficient for scalloping suppression than the existing ones for both general and complex scenes, and has clear improvement for large-scale images with strong scalloping, which fully verifies the robustness and adaptability of the proposed method.

Baseline error correction is critical for airborne synthetic aperture radar (SAR) tomography as the actual flight trajectory often deviates from the designed one due to turbulence, which may lead to large sidelobes or even complete defocusing in the tomograms. Current baseline error correction methods, however, are susceptible to heavy decorrelation noise. To mitigate the adverse effect of decorrelation noise, in this article, a novel method for airborne SAR tomography baseline errors correction driven by small baseline interferometric phase is proposed. In this method, a novel mathematical model that relates interferometric phase to the baseline error differences is first derived; then, a small baseline interferometric pairs selection strategy is employed to estimate the baseline error differences through an alternate iterative algorithm, and finally, the baseline errors are obtained through accumulating summation of the baseline error differences. The use of small baseline interferograms can avoid the phase linking processing and thereby greatly alleviate the heavy decorrelation effect. Both simulated and real airborne P-band SAR tomography experiments have demonstrated that the proposed method can achieve more accurate and robust estimation of baseline errors and is more tolerant to decorrelation noise than the well-known phase center double localization (PCDL) method.

Polarimetric synthetic aperture radar (SAR) tomography (Pol-TomoSAR) can be used for global forest digital terrain model (DTM) and canopy height model (CHM) mapping with high spatial and temporal resolution at low economic cost. However, the performance of DTM and CHM inversion in current Pol-TomoSAR methods is often compromised when the ground-to-volume ratio (GVR) is low, which usually happens in complicated terrain where large negative slope angles are commonly present, or in dense tropical forest where the ground visibility is low due to strong attenuation by the dense vegetation layer. In this work, a novel method for the separation of ground and volume scattering, aiming at robust and accurate DTM and CHM inversion in dense tropical forest and complicated terrain, is proposed. By fully exploiting multibaseline polarimetric SAR data, the proposed method can perform a more effective separation of ground and volume scattering. Subsequently, by applying the SAR tomography technology on the separated ground and volume scattering, the proposed method can retrieve accurate DTM and CHM information even at low GVR areas. Numerical experiments conducted on both simulated data and P-band airborne F-SAR data show that, compared to the most commonly used two-component algebraic synthesis method, the proposed one has a much better performance in terms of separation of ground and volume scattering. Furthermore, the inversed DTM and CHM present better agreement with light detection and ranging (LiDAR) measurements, especially in large negative slope angle terrain where the GVR is rather low.

An imaging radar is expected to provide high-quality images for interesting targets. To this end, an information-theoretic approach is used in this article to jointly optimize waveform and receive filter with desired cross-correlation properties. First, the problem formulation is achieved by maximizing the mutual information (MI), subject to constant modulus, high resolution, and low peak sidelobe ratio (PSLR) constraints. Second, to solve the resultant problem with a fractional quadratic objective function and various nonconvex constraints, four customized iterative loops are performed to transform the problem into a series of solvable subproblems, via minorization-maximization (MM), alternate direction penalty method (ADPM), and feasible point pursuit successive convex (FPP-SCA) approximation. Convergence of every iterative loop is proved, resulting in guaranteed convergence of the whole procedure with polynomial-time complexity. Finally, numerical examples are presented to demonstrate that the proposed method can construct a unimodular waveform and filter with better information acquisition ability and more desirable cross-correlation function.

A non-orthogonal multiple access (NOMA)-based spatial modulation system operating over multiple subchannels is investigated. For scheduled users of each subchannel, a mixed multicast and unicast transmission is delivered. The multicast content is transmitted via the transmit antenna domain, while unicast contents are transmitted through the amplitude-phase modulated symbols using NOMA via the active antenna. Firstly, the unicast rate for each user and an upper bound for the multicast rate are derived. Secondly, a joint subchannel and power allocation problem for weighted sum rate maximization is formulated. To solve this challenging mixed-integer non-linear problem, we decompose it into three subproblems, namely the decoding order design, the subchannel assignment, and the power allocation. A heuristic scheme is developed to solve the first one by investigating the characteristics of the decoding order constraint. To avoid the high complexity caused by exhaustive search, the subchannel assignment is reformulated as a many-to-one matching with peer effect, and the Gale-Shapley method and swap operation are designed to solve it. The power allocation is solved by employing the successive convex approximation. Moreover, a joint subchannel and power allocation algorithm is proposed to further boost the performance, and a robust power allocation algorithm is proposed under channel uncertainties.

For direction of arrival (DOA) estimation based on graph signal processing (GSP), it has been assumed that there is a phase shift between adjacent snapshots of the received signals. However, this assumption does not hold for narrowband signals and thus affects the performance of the corresponding algorithms. To improve the performance, a new GSP-based DOA estimation method is proposed. By building a periodic directed graph based on a graph shift operator and computing the spectrum using the Kronecker product, the relationship between the input narrowband signals and the graph adjacency matrix of different direction coefficients is constructed. Simulation results show that this method performs better than existing algorithms based on GSP.

The problem of non-coherent direction of arrival (DOA) estimation of wideband signals with magnitude-only measurements is studied in this paper. Unlike the traditional coherent DOA estimation methods, where discrete Fourier Transform (DFT) can be applied to sensor measurements in order to formulate the problem into a narrowband form, the non-coherent model processes wideband signals in time domain directly by exploiting the convolutional sparse coding (CSC) framework. As shown by simulations, the proposed solution has the advantage of being robust against frequency independent sensor response errors.

As an important branch of Internet of Vehicles (IoV) systems, autonomous vehicle (AV) positioning based on direction-of-arrival (DOA) estimation has received extensive attention in recent years. In this article, an AV positioning method under unknown mutual coupling is proposed within the framework of a large-dimensional asymptotic theory (LAT). First, enhanced and closed-form DOA estimation is achieved by jointly exploiting large-scale uniform linear arrays (ULAs), Toeplitz rectification and the phase transformation result associated with the sample covariance matrix; second, a more reliable subset/set of DOAs is constructed according to the signal-to-noise at receivers; finally, robust AV positioning is achieved with the reliable subset/set. Motivated by satisfactory DOA estimation performance, an AV contour extraction scheme is developed with the aid of two antennas installed on an AV. The proposed method shows several salient advantages compared with existing methods, including improved resolution and accuracy, reduced computational complexity, robustness to mutual coupling and unreasonable DOA estimates, as well as the ability to effectively extract AV contour information.

To achieve accurate interferometric synthetic aperture radar (SAR) phase estimation, it is essential to select appropriate high-coherence interferometric pairs from massive SAR single-look complex (SLC) image data. The selection should include as many high-coherence interferometric pairs as possible while avoiding low-coherence pairs. By combining coherence and spectral clustering, a novel selection method for SAR interferometric pairs is proposed in this paper. The proposed method can be adopted to classify SAR SLC images into different clusters, where the total coherence of interferometric pairs in the same cluster is maximized while that among the different clusters is minimized. This is implemented by averaging the coherence matrices of representative pixels to construct an adjacency matrix and performing eigenvalue decomposition for estimating the number of clusters. The effectiveness of the proposed method is demonstrated using 33 TerraSAR-X and 38 dual-polarization Sentinel-1A data samples, yielding improved topography and deformation monitoring results.

Multitemporal synthetic aperture radar interferometry (MTInSAR) is an efficient geodetic tool for Earth surface displacement measurement, and the polarimetric capability of current and upcoming SAR satellites offers a new opportunity to further improve MTInSAR phase series estimation. However, none of the existing estimators for multipolarimetric MTInSAR phase series of distributed scatters (DSs) is derived under the minimum root-mean-square error (RMSE) criterion. In this work, a maximum likelihood estimator for multipolarimetric phase linking (MLE-MPPL) is proposed and the corresponding Cramer–Rao lower bound (CRLB) is also derived by modeling the polarimetric interferometric coherence matrix as the Kronecker product of polarimetric coherence matrix and interferometric coherence matrix. In addition, a new metric called Pol-detR is proposed for the performance evaluation of multipolarimetric MTInSAR phase series estimation in practical scenarios where the RMSE is not feasible any more. The experimental results based on both simulated and real data show that the proposed MLE-MPPL achieves the best estimation performance and is more robust against interchannel interference than existing methods.

A new near-field source localization method is proposed for two-dimensional (2-D) direction-of-arrival (DOA) and range estimation based on a symmetrical cross array. It first employs the conjugate symmetry property of the signal auto-correlation at different time delays to construct a conjugate augmented spatial-temporal cross correlation matrix, then the extended steering vector is decoupled to avoid the usual multiple-dimensional (M-D) search based on the properties of the Khatri-Rao product, and finally three one-dimensional (1-D) MUSIC type searches are employed to obtain the results. The proposed method can realize automatic pairing of multiple parameters associated with each source and it also works in the underdetermined case.

In this paper, we consider the underdetermined two dimensional (2-D) source localization problem for wideband sources based on a distributed sensor array network, where a sparse sub-array is placed on each observation platform and the source number is larger than the sensor number of each sub-array. The received signals are first decomposed into different frequency bins via discrete Fourier transform (DFT), followed by the vectorization process to obtain the virtual array model with a larger aperture. Then, focusing is applied to the virtual array instead of the physical array for performance improvement, and a group sparsity based 2-D localization method exploiting the difference co-array is proposed, with increased DOFs for localization. Improved performance is achieved as demonstrated by computer simulations.

Multi-beam multiplexing design based on the interleaved subarray architecture can be achieved by uniform linear arrays (ULAs) consisting of isotropic antennas in previous works. By considering the polarisation information of a signal, the crossed-dipole array is employed, where each antenna is associated with two complex-valued weighting coefficients. To reduce the implementation complexity of the system, when a large number of crossed-dipole antennas are available, we may not need to employ all of them and a subset of the available antennas may be sufficient for a specific beamforming scenario. For this purpose, two design methods are proposed to select the best set of crossed-dipole antennas or dipoles. Designed examples are provided to verify the effectiveness of the proposed methods.

Abstract Radar and communication are two key applications of radio technology, and they occupy a large portion of the frequency spectrum. Traditionally, radar and communication systems are operated at different frequencies, owing to their different functions and application areas. For instance, radar was mainly employed for sensing (target detection, localization, recognition, imaging, etc.) in the military field, while wireless communication was mainly for information delivery. However, along with the fast development of radio technologies and huge demand for information, the radio frequency (RF) spectrum is becoming increasingly congested, and the spectra of the radar system will be overlaid with those of wireless communication devices. Moreover, radar and communication are becoming increasingly merged in both technologies and applications. Besides the military field, radar has been widely employed in daily life including weather service, air traffic control, autonomous driving and security monitoring. Meanwhile, these applications rely Largely on information transmission through wireless communications. In this regard, integration of radar and communication (IRC) has proved to be a very promising development to address the spectrum congestion issue between radar and communications devices. This also brings us a number of key challenges in signal processing for both implementation of IRC and joint optimization between the two systems.

In this article, a novel vehicle positioning system architecture based on direction-of-arrival (DOA) estimation of incoherently distributed (ID) sources is proposed employing massive multiple-input–multiple-output (MIMO) arrays. Such an architecture with the associated signal model is more consistent with the actual array application and multipath transmission scenarios. First, an end-to-end two-dimensional (2-D) DOA estimation of ID sources utilizing a dual one-dimensional (1-D) convolutional neural network (D1D-CNN) under the deep learning (DL) framework is performed, where the normalized covariance matrix data is used for both offline training and online estimation. Then, the received SNR information is exploited to select a set of DOA estimates provided by multiple collaborative BSs for positioning. Moreover, transfer learning and an attention mechanism are employed to promote its generalization ability and achieve robustness against array perturbations. Simulation results are provided to show that the proposed method outperforms the state-of-the-art methods in terms of computational complexity, positioning accuracy, and robustness against array perturbations.

The Global Navigation Satellite System (GNSS)-based passive bistatic radar (PBR) which uses the GNSS signal as the illuminators of opportunity is studied for moving target detection (MTD). GNSS-based PBR has many advantages due to the removal of the transmitting device; however, its fundamental limitation is the low power density of the GNSS signal. Therefore, the integration time should be sufficiently long to obtain a promising maximum detectable range. On the other hand, the integration time is limited by the range migration and Doppler migration of the echo caused by target motion. In this letter, a novel MTD algorithm is proposed for the GNSS-based PBR, by employing a modified radon Fourier transform (MRFT) to achieve the required long-time integration for moving targets. The MRFT integrates the echo energy via joint searching of range, Doppler, and Doppler rate of the target, which can handle not only the range migration but also the Doppler migration problems, and significantly improves the signal-to-noise ratio (SNR) of the echo signal. An experiment using the GPS L5 signal as the illumination source is conducted and a moving car is successfully detected by the proposed algorithm, although significant range migration and Doppler migration are present due to variation of its speed.

The wideband direction of arrival (DOA) estimation problem can be formulated into a narrowband form by applying discrete Fourier Transform (DFT) to sensor measurements; however, a large number of temporal snapshots are required in order to meet the narrowband assumption in the frequency-domain. To reduce the number of snapshots required, a convolutional sparse coding (CSC) based wideband signal model is proposed for direct time-domain DOA estimation, and a group sparsity based minimization problem is formulated. Simulation results indicate that the proposed time-domain CSC (TD-CSC) based method has a better performance than the frequency-domain method, but with a higher computational complexity.

A generalized sum-difference expansion scheme is proposed to construct sparse arrays based on the fourth-order difference co-array with increased degrees of freedom (DOFs). Different from existing structures, both the second-order sum and difference co-arrays are exploited in array construction under this scheme, leading to a large consecutive fourth-order difference co-array with its number of uniform DOFs (uDOFs) derived. To optimize the provided uDOFs, required design properties of the initial prototype arrays are discussed. Three examples are then provided to demonstrate its superior performance over existing structures in both resolution capacity and estimation accuracy.

For azimuth multi-channel synthetic aperture radar (SAR), unavoidable inconsistency errors between channels can degrade SAR image quality severely, leading to possible ghost targets and image defocusing, etc. To address this issue, a novel channel inconsistency estimation method is proposed based on maximum normalized image sharpness. First, channel amplitude and time delay errors are corrected in the coarse compensation step. Then images of each channel are attained by azimuth spectrum recovery and imaging processing. Next, range-variant channel phase errors are estimated via optimizing normalized image sharpness, which reaches the maximum value when the image is focused well or ghost targets are suppressed completely. The Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm is employed to get the optimal solution based on the derived gradient of objective function. Finally, the ultimate image is formed through adding up phase compensated images of each channel. By optimizing the focused image quality, the proposed algorithm achieves high estimation accuracy. Simulated data and real multi-channel SAR data are processed to demonstrate the effectiveness of the proposed method.

A joint design of the transmit weight sequence and receive filter is proposed to improve target information acquisition in high-resolution radar. First, using the criterion for target information acquisition maximization, the design is cast as a nonconvex fractional quadratically constrained quadratic problem (QCQP). Then, by employing a bivariate auxiliary function introduced in Dinkelbach’s algorithm to decouple the fractional objective function, an algorithm with polynomial computational complexity is developed to solve the QCQP using a cyclic maximization procedure alternating between two semidefinite relaxation (SDR) problems. Through exploiting a suitable rank-one decomposition, it is verified that the optimal solution obtained from the alternative iterative process is also optimal to the original QCQP. Finally, numerical examples are presented to demonstrate the performance of the proposed design.

In massive multiple-input multiple-output (MIMO) systems, accurate direction-of-arrival (DOA) estimation is important for the base station (BS) to perform effective downlink beamforming. So far, there have been few reports on DOA estimation considering gain-phase perturbations in massive MIMO systems. However, gain-phase perturbations indeed exist in practical applications and cannot be ignored. In this paper, an efficient method for two-dimensional (2-D) DOA estimation of incoherently distributed (ID) sources considering array gain-phase perturbations is proposed for massive MIMO systems. Firstly, a shift invariance structure is established in the subspace framework, and a constrained optimization problem is formulated to estimate the nominal azimuth and elevation DOAs as well as gain-phase perturbations with closed-form expressions, under the assumption that some of the BS antennas are well calibrated; secondly, the corresponding angular spreads are obtained with the aid of the estimated gain-phase perturbations. Theoretical analysis and an approximate Cramér-Rao bound are also provided. An improved estimation performance is achieved by the proposed method as demonstrated by numerical simulations.

Recently, some non-coherent DOA estimation methods are presented under a sparse phase retrieval framework, where DOAs of incident signals are assumed to be on the predefined grid points. However, this may not be correct in practice; in order to address this issue, an off-grid model involved with a bias vector is proposed and an efficient two-step method based on this model is developed. In addition, instead of using ULAs, uniform circular arrays (UCAs) are employed in order to overcome the ambiguities arising in non-coherent measurements, as analysed in detail. Numerical simulations show that, compared to on-grid model with a denser grid points, the off-grid model with a coarse grid can achieve a better performance with a lower computational complexity.

We present extended Cantor arrays based on fourth- order difference co-arrays (E-FO-Cantor). These arrays result from extending the recently proposed fractal arrays to fourth- order difference co-arrays, and lead to fourth-order difference co-arrays that are hole-free. The set of sensor positions of the E-FO-Cantor is expressed in a simple and recursive form. The proposed Cantor arrays lead to O(N2log23) ≈ O(N3.17) degrees of freedom compared to O(N2) that can be achieved by existing sparse arrays with the hole-free property. Compared with other sparse arrays with the hole-free property in their fourth-order co-arrays, the proposed Cantor arrays provide a longer uniform linear array with more virtual sensors, leading to better DOA estimation performance.

Directional modulation (DM) is an important technology for physical layer security in wireless communications. Recently, a symbol-independent magnitude constraint for all antennas was proposed in DM design to reduce the design complexity of its analogue implementation. However, a limitation of the method is that it can only set the magnitude to a certain value, and all the antenna coefficients have the same magnitude. In this article, a more flexible solution is provided and the challenge of the design is the nonconvex constraint enforcing an arbitrary symbol-independent magnitude for all coefficients. To solve the problem, a convex iterative method is proposed, based on which the magnitudes of weight coefficients for all antennas can be chosen by designers in advance according to the specific requirements, allowing more freedom in the design process, which is the major difference between the previously proposed design and the newly proposed one. Two design examples are provided to demonstrate the effectiveness of the proposed design. One is is a general example, where coefficient magnitudes for different symbols are the same for the same antenna, but different for different antennas; the other one is a special case where magnitudes for all antennas are the same.

The azimuth multichannel technique has been widely used in synthetic aperture radar (SAR) systems for improving the resolution and expanding the illumination area. However, due to phase inconsistency (PI) of different channels, the image quality deteriorates significantly, including resolution loss and appearance of ghost targets. In this letter, by exploiting the rotation-invariant property of the steering vector of the multichannel SAR signal, a PI error compensation method is proposed based on the estimation of signal parameters by rotation invariance technique (ESPRIT). Experimental results are presented using both simulated and real data to demonstrate the performance of the proposed method.

The Cramér-Rao bound (CRB) for direction of arrival (DOA) estimation exploiting both auto-correlation and cross-correlation information within multiple frequencies of the received array signals is derived. It provides a tighter bound than the existing CRB for the dual-frequency scenario. For the multiple frequencies, it is much lower than its dual-frequency counterpart, and also exists for a greater number of sources, thereby validating that exploiting multiple frequency pairs can improve both estimation accuracy and target resolvability.

The generalized linear moving sampling scheme (MSS) exploiting the second-order statistics and also the high-order cumulants is studied, where the set of MSS is defined as the shifted distance offsets involved in estimation based on a moving platform. Then, sparse physical arrays (SPAs) with nonuniform linear moving sampling schemes (NL-MSS), referred to as SPA-NL-MSS, are proposed to optimize the consecutive difference co-arrays. For the same number of sensors and data samples, better performance in terms of both the number of degrees of freedom (DOFs) and estimation accuracy can be achieved by SPA-NL-MSS than existing array structures exploiting array motions at the second order level.

Scanning synthetic aperture radar (ScanSAR) mode is widely used in Earth observation because of its capability of acquiring wide-swath images with moderate resolution. However, due to the operation mechanism of ScanSAR mode, the acquired images often suffer from the scalloping problem, resulting in significant deterioration of image quality. In this article, a novel scalloping suppression method is proposed for ScanSAR images based on the modified Kalman filter with preprocessing. First, an image model is built to analyze the effect caused by scalloping. Then, a modified Kalman filter is proposed to estimate the intensity of scalloping. However, if the scene is complicated or the scalloping effect is strong, the Kalman filter works with poor performance. Therefore, an innovative preprocessing operation is introduced, involving image segmentation and pixel value filling. Finally, the proposed method is verified by the GaoFen-3 (GF-3) and TerraSAR-X satellite images with different scenes. The results demonstrate that the proposed method can accommodate the complex scene well and achieve effective scalloping suppression.

The scalar sensor array (SSA) is generally assumed insensitive to the polarization of impinging signals, and only diversely polarized arrays, such as the vector (crossed-dipole or tripole) sensor array (VSA), can be used for polarization estimation. However, as shown in this paper, with the mutual coupling effect, the SSA can become partially sensitive to polarization of the impinging signals and therefore can be used for polarization parameter estimation. The polarization sensitivity model of an SSA is first established and then as an example, a dimension-reduction method based on multiple signal classification (MUSIC) is employed to jointly estimate the direction-of-arrival and polarization parameters. Computer simulations based on a planar array of circularly polarized microstrip antennas are provided to demonstrate the performance of the proposed method.

By adjusting the adjacent antenna spacing in terms of the relationship between the two required directions, previous techniques can multiplex two beams by changing the phase of different antennas. In this work, to maintain the adjacent antenna spacing as a fixed value and reduce the implementation complexity, one novel design with an equal magnitude constraint for all antennas, together with the inter-subarray coding scheme is proposed, which can achieve dual-beam multiplexing for arbitrary directions to serve two users via merely changing the analogue phase shift of each antenna. Following a most recent development in this area, the idea can be easily extended to multiple beams. Designed examples are provided to demonstrate the effectiveness of the proposed method.

The target localization problem for distributed sensor array networks where a sub-array is placed at each receiver is studied, and under the compressive sensing (CS) framework, a group sparsity based two-dimensional localization method is proposed. Instead of fusing the separately estimated angles of arrival (AOAs), it processes the information collected by all the receivers simultaneously to form the final target locations. Simulation results show that the proposed localization method provides a significant performance improvement compared with the commonly used maximum likelihood estimator (MLE).

The co-prime-sampling SAR can effectively alleviate the contradictory relationship between spatial resolution and swath width by applying co-prime sampling in azimuth. In order to achieve accurate imaging of co-prime-sampling SAR on the condition of non-negligible range cell migration, an imaging method based on 2D sparse signal reconstruction is proposed. The 2D observed signal is intercepted and the corresponding sparse dictionary is constructed both according to the Doppler parameters of each range gate. An improved 2D signal sparse reconstruction algorithm is used to perform sparse scene reconstruction. Simulation results are provided to demonstrate the effectiveness of the proposed method for sparse scene imaging.

A new method is proposed to achieve millimeter-wave two-beam multiplexing with arbitrary directions based on the interleaved subarray architecture. Beam interference can be mitigated and beam gain augmented by multiplexing multiple beams. Previous techniques can only multiplex two beams whose directions satisfy a specific relationship. By the proposed design and the associated inter-coding technique, two-beam multiplexing for arbitrary directions to serve two users is achieved. Design examples are provided to demonstrate the effectiveness of the 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.

The performance of wideband array signal processing algorithms is dependent on the noise level in the system. A method is proposed for reducing the level of white noise in wideband linear arrays via a judiciously designed spatial transformation followed by a bank of highpass filters. A detailed analysis of the method and its effect on the spectrum of the signal and noise are presented. The reduced noise level leads to a higher signal‐to‐noise ratio for the system, which can have a significant beneficial effect on the performance of various beamforming methods and other array signal processing applications such as direction of arrival estimation. Here the authors focus on the beamforming problem and study the improved performance of two well‐known beamformers, namely the reference signal based and the linearly constrained minimum variance beamformers. Both theoretical analysis and simulation results are provided.

An effective method is proposed to estimate the desired-signal (S) subspace by the intersection between the signal-plus-interference (SI) subspace and a reference space covering the angular region where the desired signal is located. The estimated S subspace is robust to steering vector mismatch and overestimation of the SI subspace, capable of detecting the relative strength of the desired signal. And even the basis of the estimated S subspace can serve as an effective estimation of the steering vector of the desired signal. With these properties, the estimated S subspace can help to select a more accurate narrow area for searching for the steering vector of the desired signal in mismatch cases. The proposed method is applied for robust adaptive beamforming with an improved performance, as demonstrated by simulation results.

The multi-azimuth-angle (MAA) UAV SAR is introduced in this paper. Compared with the traditional TomoSAR that rebuilds the 3-D scene only from the boresight aspect, the MAA UAV SAR can rebuild the 3-D scene from different multi azimuth angles and then provide multiple or even full aspects 3-D image by the novel 3-D image formation algorithm presented in this paper. Firstly, the MAA UAV SAR is introduced, followed by the 2-D imaging algorithm to focus the raw data. Then, a novel squinted tomography method is provided to obtain squinted 3-D image using multiple 2-D images with the same azimuth angle. Finally, the squinted 3-D images are geolocated and fused to construct a full 3-D image. Imaging results are provided to demonstrate the better performance of the MAA SAR.

Wind profile prediction at different scales plays a crucial role for efficient operation of wind turbines and wind power prediction. This problem can be approached in two different ways: one is based on statistical signal processing techniques and both linear and nonlinear models can be employed either separately or combined together for profile prediction; on the other hand, wind/atmospheric flow analysis is a classical problem in computational fluid dynamics (CFD) in applied mathematics, which employs various numerical methods and algorithms, although it is an extremely time-consuming process with high computational complexity. In this work, a new method is proposed based on synergy's between the signal processing approach and the CFD approach, by alternating the operations of a quaternion-valued least mean square (QLMS) algorithm and the large eddy simulation (LES) in CFD. As demonstrated by simulation results, the proposed method has a much lower computational complexity while maintaining a comparable prediction result.

Since the landmark work of R. E. Kalman in the 1960s, considerable efforts have been devoted to time series state space models for a large variety of dynamic estimation problems. In particular, parametric filters that seek analytical estimates based on a closed-form Markov–Bayes recursion, e.g., recursion from a Gaussian or Gaussian mixture (GM) prior to a Gaussian/GM posterior (termed ‘Gaussian conjugacy’ in this paper), form the backbone for a general time series filter design. Due to challenges arising from nonlinearity, multimodality (including target maneuver), intractable uncertainties (such as unknown inputs and/or non-Gaussian noises) and constraints (including circular quantities), etc., new theories, algorithms, and technologies have been developed continuously to maintain such a conjugacy, or to approximate it as close as possible. They had contributed in large part to the prospective developments of time series parametric filters in the last six decades. In this paper, we review the state of the art in distinctive categories and highlight some insights that may otherwise be easily overlooked. In particular, specific attention is paid to nonlinear systems with an informative observation, multimodal systems including Gaussian mixture posterior and maneuvers, and intractable unknown inputs and constraints, to fill some gaps in existing reviews and surveys. In addition, we provide some new thoughts on alternatives to the first-order Markov transition model and on filter evaluation with regard to computing complexity.

In this letter, an accurate mathematical model for azimuth ambiguity in stripmap synthetic aperture radar (SAR) images is first constructed, with an azimuth ambiguity factor (AAF) defined as the residual amplitude and phase terms of ambiguities. Next, a novel framework for reconstructing and suppressing azimuth ambiguity is proposed based on the analysis of the AAF. In this framework, azimuth ambiguities are accurately reconstructed by applying reconstruction filters in the range Doppler and 2-D frequency domain, and then, the reconstructed signal is used for suppressing azimuth ambiguities. Moreover, the proposed framework does not depend on the statistical characteristics of a SAR image and is capable of reducing the space-variant ambiguities. As verified by both simulated data and real TerraSAR-X data, the proposed method is capable of suppressing azimuth ambiguities in SAR images.

An expanding and shift scheme for efficient fourth-order difference coarray construction is proposed. It consists of two sparse subarrays, where one of them is modified and shifted according to the analysis provided. The number of consecutive lags of the proposed structure at the fourth order is consistently larger than two previously proposed methods. Two effective construction examples are provided with the second sparse subarray chosen to be a two-level nested array, as such a choice can increase the number of consecutive lags further. Simulations are performed to show the improved performance by the proposed method in comparison with existing structures.

A two-dimensional (2-D) direction of arrival estimation method for L-shaped array with automatic pairing is proposed. It exploits the conjugate symmetry property of the array manifold matrix to increase the effective array aperture and the number of virtual snapshots simultaneously, and then applies the principle of MUSIC to construct an angle cost function and transforms the conventional 2-D search into 1-D via a Rayleigh quotient, which can greatly reduce the computation complexity. Finally, the azimuth and elevation angles are estimated without pair matching. Simulation results show that the proposed method has a better performance and can resolve more sources than some existing computationally efficient methods.

Group-sparsity-based method is applied to the 2qth-order difference coarray for underdetermined wideband direction of arrival (DOA) estimation. For complexity reduction, a focused compressive-sensing-based approach is proposed, without sacrificing its performance. Different from the conventional focusing approach, in the proposed one, focusing is applied to the virtual arrays and no preliminary DOA estimation is required. Simulation results are provided to demonstrate the effectiveness of the proposed methods.

There are two challenges for efficient processing of both the sliding spotlight and terrain observation by progressive scans (TOPS) data using full-aperture algorithms. First, to overcome the Doppler spectrum aliasing, zero-padding is required for azimuth up sampling, increasing the computation burden; second, the azimuth deramp operation for avoiding synthetic aperture radar (SAR) image folding leads to azimuth time shift along the range dimension, and in turn the appearance of ghost targets and azimuth resolution reduction at the scene edge, especially in the wide-swath case. In this paper, a novel three-step algorithm is proposed for processing the sliding spotlight and TOPS data. In the first step, a modified derotation is derived in detail based on the chirp z-transform (CZT), avoiding zero-padding; then, the chirp scaling algorithm kernel is adopted for precise focusing in the second step; and in the third step, instead of the traditional range-independent deramp, a range-dependent deramp is applied to compensate for the time shift. Moreover, the SAR image geometry distortion caused by range-dependent deramp is corrected by employing a range-dependent CZT. Experimental results based on both simulated data and real data are provided to validate the proposed algorithm.

To reach a higher number of degrees of freedom by exploiting the fourth-order difference co-array concept, an effective structure extension based on two-level nested arrays is proposed. It increases the number of consecutive lags in the fourth-order difference coarray, and a virtual uniform linear array (ULA) with more sensors and a larger aperture is then generated from the proposed structure, leading to a much higher number of distinguishable sources with a higher accuracy. Compressive sensing based approach is applied for direction-of-arrival (DOA) estimation by vectorizing the fourth-order cumulant matrix of the array, assuming non-Gaussian impinging signals.

A new model for three-dimensional processes based on the trinion algebra is introduced for the first time. Compared to the pure quaternion model, the trinion model is more compact and computationally more efficient, while having similar or comparable performance in terms of adaptive linear filtering. Moreover, the trinion model can effectively represent the general relationship of state evolution in Kalman filtering, where the pure quaternion model fails. Simulations on real-world wind recordings and synthetic data sets are provided to demonstrate the potential of this new modeling method.

The gradients of a quaternion-valued function are often required for quaternionic signal processing algorithms. The HR gradient operator provides a viable framework and has found a number of applications. However, the applications so far have been limited to mainly real-valued quaternion functions and linear quaternionvalued functions. To generalize the operator to nonlinear quaternion functions, we define a restricted version of the HR operator, which comes in two versions, the left and the right ones. We then present a detailed analysis of the properties of the operators, including several different product rules and chain rules. Using the new rules, we derive explicit expressions for the derivatives of a class of regular nonlinear quaternion-valued functions, and prove that the restricted HR gradients are consistent with the gradients in the real domain. As an application, the derivation of the least mean square algorithm and a nonlinear adaptive algorithm is provided. Simulation results based on vector sensor arrays are presented as an example to demonstrate the effectiveness of the quaternion-valued signal model and the derived signal processing algorithm.

To process squinted sliding-spotlight synthetic aperture radar data, the azimuth preprocessing step based on the linear range walk correction (LRWC) and derotation operations is implemented to eliminate the effect of 2-D spectrum skew and azimuth spectral aliasing. However, two key issues arise from the azimuth preprocessing. First, the traditional chirp scaling (CS) kernel is not suitable for data focusing because the property of the 2-D spectrum is changed significantly; second, the spatial variation of the targets' Doppler rates along the azimuth direction due to the LRWC operation limits the depth-of-azimuth-focus (DOAF) seriously. In this letter, a modified accurate CS kernel is derived to realize range compensation. Then, an azimuth spatial variation removing method based on the principle of nonlinear CS is proposed to equalize the Doppler rates of the targets located at the same range cell, which can extend the DOAF and improve processing efficiency. Finally, a novel imaging algorithm is proposed, with its effectiveness demonstrated by simulation results.

An effective sparse array extension method for maximizing the number of consecutive lags in the fourth-order difference co-array is proposed, leading to a novel enhanced sparse array structure based on co-prime arrays (CPAs) with significantly increased number of degrees of freedom (DOFs). One method to exploit the increased DOFs based on nonstationary signals is also proposed, with simulation results provided to demonstrate the effectiveness of the proposed structure.

Recently, a robust adaptive beamforming (RAB) technique based on interference-plus-noise covariance (INC) matrix reconstruction has been proposed, which utilizes the Capon spectrum estimator integrated over a region separated from the direction of the desired signal. Inspired by the sampling and reconstruction idea, in this paper, a novel method named spatial power spectrum sampling (SPSS) is proposed to reconstruct the INC matrix more efficiently, with the corresponding beamforming algorithm developed, where the covariance matrix taper (CMT) technique is employed to further improve its performance. Simulation results are provided to demonstrate the effectiveness of the proposed method.

A class of low-complexity compressive sensing-based direction-of-arrival (DOA) estimation methods for wideband co-prime arrays is proposed. It is based on a recently proposed narrowband estimation method, where a virtual array model is generated by directly vectorizing the covariance matrix and then using a sparse signal recovery method to obtain the estimation result. As there are a large number of redundant entries in both the auto-correlation and cross-correlation matrices of the two sub-arrays, they can be combined together to form a model with a significantly reduced dimension, thereby leading to a solution with much lower computational complexity without sacrificing performance. A further reduction in complexity is achieved by removing noise power estimation from the formulation. Then, the two proposed low-complexity methods are extended to the wideband realm utilizing a group sparsity based signal reconstruction method. A particular advantage of group sparsity is that it allows a much larger unit inter-element spacing than the standard co-prime array and therefore leads to further improved performance.

Two challenges have been faced in signal processing of ultrahigh-resolution spaceborne synthetic aperture radar (SAR). The first challenge is constructing a precise range model, and the second one is to develop an efficient imaging algorithm since traditional algorithms fail to process ultrahigh-resolution spaceborne SAR data effectively. In this paper, a novel high-order imaging algorithm for high-resolution spaceborne SAR is presented. First, a modified equivalent squint range model (MESRM) is developed by introducing equivalent radar acceleration into the equivalent squint range model, and it is more suitable for high-resolution spaceborne SAR. The signal model based on the MESRM is also presented. Second, a novel high-order imaging algorithm is derived. The insufficient pulse-repetition frequency problem is solved by an improved subaperture method, and accurate focusing is achieved through an extended hybrid correlation algorithm. Simulations are performed to validate the presented algorithm.

In synthetic aperture radar interferometry, past studies of interferometric phase statistics are mainly based on the assumption that the interferogram cell size is much larger than the wavelength of the incident radiation and the scene is a homogeneously distributed scatterer. However, strong scatterers are often present in the scene, and in this work, the interferometric phase statistics are studied for this case for single-look interferograms. Its closed-form probability density function is first derived by approximating the complex interferogram signals to two correlated Gaussian random variables with nonzero-mean values. The closed-form mean value and variance are then derived with the assumption that the intensity of the strong scatterer is much larger than that of its background. It is shown that the phase statistics of strong scatterers are related not only to the correlation but also to the intensity of the dominant point and the phase difference variation between the dominant point and the remaining points.

Sparse wideband array design for sensor location optimization is highly nonlinear and it is traditionally solved by genetic algorithms (GAs) or other similar optimization methods. This is an extremely time-consuming process and an optimum solution is not always guaranteed. In this work, this problem is studied from the viewpoint of compressive sensing (CS). Although there have been CS-based methods proposed for the design of sparse narrowband arrays, its extension to the wideband case is not straightforward, as there are multiple coefficients associated with each sensor and they have to be simultaneously minimized in order to discard the corresponding sensor locations. At first, sensor location optimization for both general wideband beamforming and frequency invariant beamforming is considered. Then, sparsity in the tapped delay-line (TDL) coefficients associated with each sensor is considered in order to reduce the implementation complexity of each TDL. Finally, design of robust wideband arrays against norm-bounded steering vector errors is addressed. Design examples are provided to verify the effectiveness of the proposed methods, with comparisons drawn with a GA-based design method.

A simple scheme for direction of arrival (DOA) estimation of coherent targets in a multiple-input multiple-output (MIMO) radar is proposed. It is based on the idea of joint transmission and reception diversity smoothing. Compared to the existing transmission diversity smoothing (TDS) method, the major advantage of the new scheme is that more covariance matrices are available for averaging to decorrelate the coherent signals, leading to a better estimation result. Moreover, it is able to identify much more coherent targets than the TDS method when sparse arrays are used.

A robust adaptive beamforming scheme based on two-component electromagnetic (EM) vector-sensor arrays is proposed by extending the well-known worst-case constraint into the quaternionic domain. After defining the uncertainty set of the desired signal's quaternionic steering vector, two quaternion-based constrained minimization problems are derived. We then reformulate them into two real-valued convex quadratic problems, which can be easily solved via the second-order cone (SOC) programming approach. Numerical simulations show that our quaternion-based robust beamformer significantly outperforms the sample matrix inversion minimum variance distortionless response (SMI-MVDR) beamformer and the quaternion Capon (Q-Capon) beamformer in the presence of steering vector mismatches.

A novel design of oversampled generalised discrete Fourier transform filter banks is proposed, with application to subband‐based convolutive blind source separation (BSS), where either instantaneous BSS algorithms or joint BSS algorithms can be applied. Conventional filter banks design is usually focused on elimination of the overall aliasing error and the perfect reconstruction (PR) condition, which are required by traditional subband adaptive filtering applications. However, because of the unknown scaling factor, the traditional PR condition is not necessary in the context of subband BSS and can be relaxed in the design. Owing to the increased degrees of design freedom, the authors can introduce an additional cost function to enhance the mutual information between adjacent subband signals. Together with a reduced subband aliasing level, it leads to an improved subband permutation alignment result for instantaneous BSS and an overall better performance for the joint BSS.

Recently, a sensor delay-line (SDL) based array structure was proposed for broadband beamforming. It is similar in form to the traditional narrowband beamformer, the only difference being a re-arrangement of the sensor positions according to the SDL concept. However, it is well-known that the narrowband beamformer can perform broadband beamforming to some degree, which gives rise to the question of whether this new arrangement is superior to the narrowband approach or not. The aim of this work is to answer this question by providing a detailed study of sensor positions for broadband beamforming. We show that when the bandwidth exceeds some threshold value, the SDL beamformer significantly outperforms the narrowband beamformer, which well justifies the new structure and gives a clear endorsement to the SDL concept.

A novel robust fixed frequency invariant beamformer (FIB) against arbitrary array manifold errors is proposed, by employing the worst-case performance optimization technique with norm-bounded error matrices. A closed-form solution is provided by finding the minimum generalized eigenvector of a pair of matrices. Design examples show that our proposed design significantly outperforms previously proposed FIBs in terms of both frequency invariant property and sidelobe attenuation in the presence of array manifold errors.

Subband-based blind source separation has great potential in solving the complicated convolutive mixing problem. However, its performance is largely affected by the permutation ambiguity problem during the synthesis stage. Researchers have suggested methods to correct the permutation by exploiting the correlation information between adjacent frequencies/subbands. An improved solution to this permutation problem is proposed based on a novel filter banks design method, which is based on a model that includes inter-subband correlation as part of the optimisation criterion. Simulation results show that a better subband permutation alignment result has been achieved, leading to improved separation performance.

An approach for blind adaptive wideband beamforming is proposed based on a uniform circular array. The received array signals are first transformed into different phase modes and each phase mode output is then processed by a filter to achieve a frequency independent response. As a result, a set of instantaneous mixtures of the original source signals is obtained and the original wideband beamforming problem can be readily solved using the standard instantaneous BSS algorithms and the original source signals can be recovered one by one or simultaneously, depending on the specific requirement.

In this treatise, we investigated the application of singular value decomposion (SVD) assisted multiuser transmission in a multicell scenario. The SVD based scheme is capable of completely removing the cochannel interference, similarly to the classic zero forcing (ZF) based and block diagonalization (BD) aided schemes. Two different power allocation schemes are investigated for both SVD, ZF and BD based multicell transmission. The SVD scheme achieves a suboptimal performance, but at a reduced complexity. Nonetheless, it always outperforms the ZF based scheme due to the joint reception of the transmitted symbols.

In this contribution we propose and study a novel multiuser detection (MUD) scheme for multicarrier direct-sequence code-division multiple-access systems employing both time (T)-domain and frequency (F)-domain spreading, which are referred to as the TF/MC DS-CDMA systems. Specifically, a reliability-aided MUD scheme is proposed, which consists of a linear MUD and a so-called L-level maximum-likelihood (ML)-MUD. The linear MUD is either a joint TF-domain linear MUD or constituted by two linear MUDs, one of which is operated in the T-domain and the other one in the F-domain. The L-level ML-MUD is a reduced ML-MUD, which only searches in a space of size 2Lin order to find the optimum solutions for the L most unreliable data bits detected by the linear MUD. In this contribution the bit error rate (BER) performance of the TF/MC DS-CDMA using the reliability-aided MUD is investigated, when communicating over additive white Gaussian noise (AWGN) channels or over frequency-selective Rayleigh fading channels. Our simulation results show that the reliability-aided MUD is capable of significantly outperforming the corresponding linear MUD. It can be shown that, provided that the signal-to-noise ratio (SNR) is sufficiently high, the reliability-aided MUD can readily obtain several decibels of SNR gain over the linear MUD at the cost of slight or moderate increase of the detection complexity.

The bandwidth performance of a two-element linearly constrained minimum variance beamformer with sensor delay-lines (SDLs) attached is studied in terms of the directions of the interference signals, the inter-spacing between delay-line sensors and the length of the SDL. Compared with broadband beamformers with tapped delay-lines (TDLs), the SDL-based structure performs better in two ways: its output SINR drops less as the inter-delay within delay- lines increases and with the same number of delays and weights it can achieve a better performance than the TDL one.

A novel singular value decomposition (SVD) aided uplink (UL) multiuser MIMO system is proposed. In contrast to the traditional minimum mean square error (MMSE) or zero- forcing (ZF) multiuser detection (MUD) technique, the proposed method exploits the specific characteristics of the individual users' channel matrix, instead of treating all the users' channels jointly. Furthermore, two different power allocation schemes are investigated in the context of the proposed structure. One of them was designed for achieving the maximum information rate, while the other for maintaining the maximum signal-to-noise ratio (SNR). We demonstrate that the capacity of the proposed scheme using the maximum information rate based power allocation policy is higher than that of the classic ZF receiver for the UL.

The exploration of ionosphere is significant for satellite communication and navigation etc. Spaceborne HF-SAR is utilized for the observation of topside ionosphere, in order to acquire higher spatial resolution, global scale ionospheric electron density map and irregularities distribution. The operation mode and system parameters are introduced. The echo signal of spaceborne HF-SAR has long synthetic aperture time, large range migration, and small depth of focus due to the low carrier frequency. Considering these characteristics of spaceborne HF-SAR, a two dimension time-frequency domain correlation image formation algorithm is presented. The effectiveness of the algorithm is validated by computer simulation results.

In this contribution an adaptive detection scheme based on least mean square (LMS) principles is proposed and investigated in the context of the hybrid direct-sequence time-hopping ultrawide bandwidth (DS-TH UWB) systems. The bit-error-rate (BER) performance of the hybrid DS-TH UWB system is investigated when communicating over the UWB channels modelled by the Saleh-Valenzuela (S-V) channel model. Furthermore, since both the pure DS-UWB and pure TH-UWB constitute special examples of the hybrid DS-TH UWB, their BER performance is also investigated in this contribution for the sake of comparison with that of the hybrid DS-TH UWB. Our study and simulation results show that the LMS-aided adaptive detection can be a feasible detection scheme for deployment in practical DS-, TH- or hybrid DS-TH UWB systems. It can be shown that, with the aid of a training sequence of reasonable length, the considered UWB schemes are capable of achieving a BER performance which is close to that achieved by the minimum mean-square error (MMSE) detector with perfect channel knowledge.

In this treatise, we propose a novel singular value decomposition (SVD) based downlink (DL) multiuser MIMO system, which takes into account the specific characteristics of the individual users channel matrix, instead of treating all the users channels jointly, as in the traditional minimum mean square error (MMSE) or zero-forcing (ZF) multiuser transmission (MUT) technique. Furthermore, two different power allocation schemes are investigated in the context of the proposed structure. One of them was designed for achieving the maximum information rate, while the other for maintaining maximum signal-to-noise ratio (SNR). The simulation results demonstrate that the BER performance of the proposed structure combined with the maximum SNR based power allocation policy is better than that of the traditional equal-power policy employed in the DL.

Transmit preprocessing employed at the basestation (BS) has been proposed for simplifying the design of the mobile receiver. Provided that the channel impulse response (CIR) of all the BS to mobile station (MS) links is known in advance-even before the signal's transmission-it is plausible that the different users' signals may be differentiated with the aid of their unique, user-specific downlink CIRs. Naturally, this non-causal CIR knowledge is unavailable in practice. Hence a natural design option is to estimate the CIRs at the receiver after the BS's signal was received and convey it using side-information to the BS for its future use. Naturally, the resultant CIR has to be quantized before its transmission. In addition to this quantization error, it also becomes outdated and both imperfections result in an erosion of the achievable transmit preprocessing gain expressed in terms of either the attainable transmit power reduction or the number of users that may be supported. Another attractive design option is to avoid the CIR-signalling latency by invoking the previously received CIRs for predicting their future evolution using CIR-tap prediction. In this paper, Kalman filtering aided CIR prediction is combined with both minimum mean square error (MMSE) and zero forcing based preprocessing. Our simulation results show that the proposed scheme is capable of attaining 6.5 dB gain at a BER of 10-2, when using 6 antennas.

Invoking SDMA in the down-link (DL) has the potential of increasing the achievable throughput with the aid of linear transmit preprocessing, provided that the channel impulse responses (CIRs) of all users and all antenna elements are known at the DL transmitter. However, in a frequency division duplex (FDD) system, since these CIRs have to be transmitted by the mobile terminals (MTs) to the base station (BS), they are naturally out-dated. Hence, we proposed a periodical CIR update scheme employing a channel predictor at the DL transmitter for predicting the CIR taps for each future symbol transmission instant and hence to mitigate the performance degradation imposed by the associated signalling delays. Moreover, a predictive vector quantizer (PVQ) is used at the MTs for compressing the CIRs before their uplink transmission. Compared to a conventional vector quantizer (VQ), PVQ has significantly reduced the CIR feedback bit rate. Hence, with the aid of the same feedback bit rate, the new PVQ scheme can provide more accurate CIR information or support a channel having a higher Doppler frequency.

Existing blind source extraction (BSE) methods are limited to noise-free mixtures, which is not realistic. We therefore address this issue and propose an algorithm based on the normalised kurtosis and a nonlinear predictor within the BSE structure, which makes this class of algorithms suitable for noisy environments, a typical situation in practice. Based on a rigorous analysis of the existing BSE methods we also propose a new optimisation paradigm which aims at minimising the normalised mean square prediction error (MSPE). This makes redundant the need for preprocessing or orthogonality transform. Simulation results are provided which confirm the validity of the theoretical results and demonstrate the performance of the derived algorithms in noisy mixing environments.

Broadband beamforming, including frequency invariant beamforming, is often achieved by processing the received sensor signals through tapped delay-lines. Unlike most of the existing techniques, we propose a novel design method for three-dimensional frequency invariant beamformers without employing tapped delay-lines. The resultant beamformer, which can form a beam steerable along both the elevation angle and the azimuth angle, has a very simple implementation. A design example is provided to show the effectiveness of the proposed method.

It is demonstrated that iteratively decoded variable length space time coded modulation (VL-STCM-ID) schemes are capable of simultaneously providing both coding gain as well as multiplexing and diversity gain. The VL-STCM-ID arrangement is a jointly designed iteratively decoded scheme combining source coding, channel coding, modulation as well as spatial diversity/multiplexing. In this contribution, we analyse the iterative decoding convergence of the VL-STCM-ID scheme using symbol-based three-dimensional EXIT charts. The performance of the VL-STCM-ID scheme is shown to be about 14.6 dB better than that of the fixed length STCM (FL-STCM) benchmarker at a source symbol error ratio of 10-4, when communicating over uncorrelated Rayleigh fading channels. The performance of the VL-STCM-ID scheme when communicating over correlated Rayleigh fading channels using imperfect channel state information is also studied.

In this contribution projection approximation subspace tracking using deflation (PASTD) is investigated in the context of MIMO transmit preprocessing systems by exploiting the specific property of time division duplexing (TDD) techniques that the uplink and downlink channels are similar, since they both use the same carrier frequency. Hence the channel estimated from the received signal can also be used for transmit preprocessing. More explicitly, based on the received signal, the PASTD algorithm is used for tracking both the left and the right singular vectors of the MIMO channel matrix, which are required by eigenmode transmissions, instead of periodically reestimating the MIMO channel matrix and performing the singular value decomposition (SVD), which would impose a high computational complexity. A specific deficiency of the family of subspace tracking algorithms is their phase ambiguity imposed by the non-unique nature of the SVD, which is resolved in this treatise by employing differential encoding. The efficiency of the proposed subspace tracking scheme is demonstrated by our performance results, indicating that the advocated technique preforms within 1 dB from the BER curve of the perfect channel estimation aided benchmarker.

Eigen-beamforming is capable of attaining attractive performance gains in the context of multiple-input multiple-output (MIMO) systems, provided that accurate channel state information (CSI) is available. However, when a realistic pilot-based channel predictor is used for acquiring the CSI, a significant performance degradation may be imposed by the phase-ambiguity inherent in the estimated eigen-vectors. In this contribution, both higher-complexity coherently detected coded modulation (CM) schemes requiring channel information as well as their lower-complexity differentially encoded counterparts are employed for assisting the operation of the eigen-beamformer, when using a minimum mean square error based pilot-assisted channel predictor. It is shown that differentially encoded CM schemes are capable of assisting the eigen-beamformer in attaining a coding gain of about 6.5 dBs, when communicating over correlated Rayleigh fading channels.

A critical analysis of the canonical correlation analysis (CCA) approach in blind source separation (BSS) is provided. It is proved that by maximizing the autocorrelation functions of the recovered signals we can separate the source signals successfully. It is further shown that the CCA approach represents the same class of generalized eigenvalue decomposition (GEVD) problems as the matrix pencil method. Finally, online realizations of the CCA approach are discussed with a linear-predictor-based algorithm studied as an example.

We introduce an algorithm for blind source extraction (BSE) of independent sources in the presence of noise, without the need for initial prewhitening, for which the normalised kurtosis is used within the cost function. Unlike the previously proposed methods designed for noise-free mixtures, which is not realistic in practical applications, we address BSE for noisy mixtures and propose a novel cost function which caters for the effects of noise. The proposed method is justified by rigorous analysis and supported by simulations

An analysis of the canonical correlation analysis (CCA) approach in blind source separation is provided. In particular, it is proved that by maximizing the autocorrelation functions of the recovered signals we can separate the source signals successfully. We show that the CCA approach represents the same generalised eigenvalue decomposition problem introduced in the matrix pencil method. Finally, an adaptive blind source extraction (BSE) algorithm is derived as an online realisation of the CCA approach. Simulation results verify the proposed approach

The blind source extraction (BSE) problem for noisy measurements is addressed using the linear predictor method. Based on a previously proposed method for the noise-free case, we propose a cost function with the effect of noise removed. Two adaptive algorithms are next introduced, one of which is based on minimisation of the normalised mean square prediction error (MSPE), whereas the other minimises the MSPE using prewhitening followed by regularisation of the demixing vector. The successful operation of these algorithms requires the knowledge of the correlation matrix of noise

In this contribution, the application of fully connected recurrent neural networks (FCRNNs) is investigated in the context of narrowband channel prediction. Three different algorithms, namely the real time recurrent learning (RTRL), the global extended Kalman filter (GEKF) and the decoupled extended Kalman filter (DEKF) are used for training the recurrent neural network (RNN) based channel predictor. Our simulation results show that the GEKF and DEKF training schemes have the potential of converging faster than the RTRL training scheme as well as attaining a better MSE performance.

A novel method for separation of a class of convolutive mixtures is proposed, in which the received sensor signals are first transformed into instantaneous mixtures and then standard blind source separation (BSS) algorithms for instantaneous mixtures are applied. Since partial information about the mixing mechanism is required in the design of the transformation, the proposed method is strictly speaking semi-blind. From the beamforming viewpoint, the proposed approach represents a blind broadband beamforming method. As the separation is performed in fullband and only one separation is needed, the permutation problem associated with the frequency-domain BSS is avoided and the separation can be easily implemented online. Simulation results verify the usefulness of the proposed method.

In this paper, a new class of broadband arrays with frequency-invariant beam patterns is proposed. By suitable substitutions, the beam pattern of a continuous sensor array with continuous temporal processing can be regarded as the Fourier transform of its spatio-temporal distribution. Based on this principle, starting from the desired frequency-invariant beam pattern, and by a series of substitutions, a simple design method is derived. This method can be applied to one-dimensional (1D), 2D, or 3D broadband arrays, either with continuous arrays and signal processing or with discrete arrays and signal processing. A 2D discrete design example is presented.

In this paper, a novel subband-selective generalized sidelobe canceller (GSC) with cosine-modulation for partially adaptive broadband beamforming is proposed. The columns of the blocking matrix are derived from a prototype vector by cosine-modulation, and the broadside constraint is incorporated by imposing zeros on the prototype vector appropriately. These columns constitute a series of bandpass filters, which select signals with specific direction of arrivals and frequencies. This results in high pass-type bandlimited spectra of the blocking matrix outputs, which is further exploited by subband decomposition and discarding the low-pass subbands appropriately prior to running independent unconstrained adaptive filters in each non-redundant subband. By these steps, the computational complexity of our GSC implementation is greatly reduced compared to fully adaptive GSC schemes, while performance is comparable or even enhanced due to subband decorrelation in both spatial and temporal domains.

This paper proposes an efficient class of perfect reconstruction (PR) modulated filter banks (MFB) using sum-of-powers-of-two (SOPOT) coefficients. This is based on a modified factorization of the DCT-IV matrix and the lossless lattice structure of the prototype filter, which allows the coefficients to be represented in SOPOT form without affecting the PR condition. A genetic algorithm (GA) is then used to search for these SOPOT coefficients. Design examples show that SOPOT MFB with a good frequency characteristic can be designed with very low implementation complexity. The usefulness of the approach is demonstrated with a 16 channel design example.

A new family of two-channel low-delay filter banks and wavelet bases using the PR structure of Phoong, Kim and Vaidyanathan (1995) with sum of powers-of-two (SOPOT) coefficients are proposed. In particular, the functions alpha(z) and beta(z) in the structure are chosen as nonlinear-phase FIR and IIR filters, and the design of such multiplier-less filter banks is performed using the genetic algorithm. The proposed design method is very simple to use, and is sufficiently general to construct low-delay filter banks with flexible lengths, delays, and regularity. Several design examples are given to demonstrate the usefulness of the proposed method.

In this paper, a new family of multiplier-less modulated filter banks, called the SOPOT MFB, is presented. The coefficients of the proposed filter banks consist of sum of powers-of-two coefficients (SOPOT), which require only simple shifts and additions for implementation. The modulation matrix and the prototype filter are derived from the fast DCT-IV algorithm of Wang (1984) and a lattice structure. The design of the SOPOT MFB is performed using the genetic algorithm (GA). An 16-channel SOPOT MFB with 34 dB stopband attenuation is given as an example, and its average number of terms per SOPOT coefficient is only 2.6.