An improved dynamic programming track-before-detect (DP-TBD) algorithm is proposed in this study. A new relaxed DP-TBD test statistic containing a term of state transition probability is derived. The state transition probability is designed according to the one-step prediction of the target state. An asymptotic and recursive solution is developed to obtain the state prediction by the polynomial time series model under the framework of weighted least squares. The impact of the weight parameter on the performance of the proposed algorithm is also investigated. The proposed algorithm can efficiently integrate the energy back-scattered along the admissible target trajectory in that the designed state transition probability enables the relaxed test statistic to distinguish real targets from the false ones more effectively. The prediction needs no priori information of target state space model and can be embedded in the recursion of the DP-TBD. Numerical simulations are provided to assess and compare the performance of the proposed algorithm. It turns out that the proposed algorithm has better detection and tracking performance than the basic one and is resilient to various target motion forms.

This study proposes a new joint direction of arrival (DOA) and time of arrival (TOA) estimation (JDTE) scheme for multi-band orthogonal frequency-division multiplexing (MB-OFDM) ultra-wideband systems with a two-antenna receiver based on frequency-domain processing. Prior joint TOA and DOA estimation (JTDE) schemes first measure the TOAs from the received waveforms at all antennas and then obtain the DOA from the TOA measurements. The proposed JDTE method estimates parameters directly from the waveforms received at all antennas, where the channel frequency response associated with the two antenna elements can be considered as a 2D (frequency and space) ‘virtual array’. The DOA is estimated by exploiting the spatial dimension of the virtual array first, and then the TOA is estimated by exploiting the frequency dimension of the virtual array. The use of the virtual array enables the number of resolvable paths to exceed the number of antennas. More importantly, the Cramer–Rao bound (CRB) of DOA estimation that is derived from the received waveforms is significantly lower than that obtained from the TOA metric, and the CRB of the estimated TOA is also slightly lower than that obtained using the JTDE scheme. Numerical simulations verify the superiority of the proposed method.

Mobile indoor localisation has numerous uses for logistics and health applications. Current wireless localisation systems experience reliability difficulties in indoor environments due to interference and also require a large number of wireless access points to ensure position accuracy and resolution. Localisation using wireless channel propagation characteristics, such as radio-frequency (RF) receives signal strength are subject to wireless interference. The Fingerprint Context Aware Partitioning (FCAP) tracking model used received RF signal strength fingerprinting, combined with context aware information about the user's indoor environment. The authors show the use of context aware information in the FCAP model, reduces the effect of wireless interference and lowers the spatial density of access points required. The wireless localisation network consisted of reference nodes placed at locations in a building. Reference nodes are used by mobile nodes, to localise a user's position. The authors tested the FCAP model in a typical indoor environment and compared the performance and accuracy to other received signal strength indicator fingerprint localisation methods. They found the FCAP model had improved performance and was able to achieve a similar accuracy to other protocols, with fewer reference nodes.

This study presents a comparative analysis of amplitude and Doppler spectra statistics from monostatic and bistatic sea clutter data at S-band. These data cover a range of bistatic angles from 60° to 120° and include simultaneous monostatic and co-polarised and cross-polarised bistatic recordings, providing direct comparison of their statistical properties. The time series of amplitude data are fitted to the compound K and K + noise distribution in the range domain and in the Doppler domain, and the variation of the shape parameter is presented for monostatic and co-polarised and cross-polarised bistatic data as an indication of the spikiness of the clutter. From the analysis of this parameter, it is shown that cross-polarised bistatic data tend to be less spiky in the range and Doppler domain than the simultaneous co-polarised bistatic data, and than the simultaneous monostatic data, which could be advantageous in terms of radar detection performance, especially for low-visibility maritime targets.

Space time adaptive processing (STAP) is an excellent technique for improving ground moving target detection performance of airborne radar. However, covariance matrix estimation required by STAP is commonly corrupted by the presence of target-like signals (i.e. outliers), and thus resulting severe performance degradation. To overcome this problem, a robust non-homogeneity detector based on reweighted adaptive power residue is developed, where an adaptively reweighted scheme is employed to training data set. Therefore, the deleterious effect of outliers on the covariance estimation is eliminated and the robustness of the non-homogeneity detector is guaranteed. Performance analysis using the simulated and measured data validates that the proposed method can effectively remove outliers from the training data and improves the radar detection performance in a dense target environment.

Multiple input multiple output (MIMO) beamforming is examined both theoretically and experimentally to highlight practical performance aspects that require a careful understanding when deciding if the MIMO concept is appropriate for a given application. Specifically, the time division multiplexing (TDM) and code division multiplexing (CDM) cases are compared. While both TDM and CDM form beams that agree closely with predictions, there are significant differences in system performance due to limits on the orthogonality of waveforms that occur when using CDM. These limits are manifested in the peak-to-mean sidelobe ratios, resulting in significant and extended range sidelobes consistent with a value approximately equal to the time-bandwidth product, and fairly independent from the number of transmitters or receivers used. In many cases, these sidelobes fundamentally limit the dynamic range of the radar. Simulations and experiments using a TDM/CDM MIMO radar validated these observations.

Phased-array provides range-independent beampattern and consequently it cannot detect/suppress range-dependent targets/interferences. Moreover, an M-element phased-array can resolve at most M − 1 targets simultaneously. This study proposes a nested array receiver using diverse time-delayers (DTDs) for jointly estimating the range and angle of targets. The essence is to construct a new array structure by systematically nesting two uniform linear arrays through DTDs instead of phase controlling in traditional nested array technique. Using the second-order statistics of the received data, an M-element array can resolve O(M ²) targets. Moreover, the nested DTD receiver design is exploited for minimum variance distortionless response beamforming and direction finding. The improvements offered by the proposed method, as compared with basic DTD receivers, are demonstrated by examining the output signal-to-interference-plus-noise ratio, the probability of detection and the Cramér–Rao lower bounds for estimating the range and angle of targets.

For different applications with diverse requirements, many synthetic aperture radar (SAR) modes have been developed in the literature, such as, spotlight SAR, terrain observation by progressive scans (TOPS) SAR, and sliding spotlight SAR. For highly squinted beam steering SAR (BS-SAR; spotlight SAR, sliding spotlight SAR, and TOPS SAR), the main challenge is to develop an efficient full-aperture focusing approach since the mostly traditional algorithms fail to process the highly squinted BS-SAR data without subaperture technique. In this study, a unified full-aperture focusing algorithm is presented for BS-SAR modes in the case of highly squint angle. The key of the algorithm lies in the linear multiplications applied in the range-wavenumber and azimuth-space domain, which can remove the azimuth spectral folding effect and the geometrical distortion caused by spectrum skewing. Moreover, there is no any approximation in the focusing procedure; thus, the presented algorithm can perform well. Since the subaperture is avoided, this improved focusing approach is highly efficient. Simulation results are presented to validate the proposed method.

This study presents the analysis of recent experimental data acquired using two radar systems at S-band and X-band to measure simultaneous monostatic and bistatic signatures of operational wind turbines near Shrivenham, UK. Bistatic and multistatic radars are a potential approach to mitigate the adverse effects of wind farm clutter on the performance of radar systems, which is a well-known problem for air traffic control and air defence radar. This analysis compares the simultaneous monostatic and bistatic micro-Doppler signatures of two operational turbines and investigates the key differences at bistatic angles up to 23°. The variations of the signature with different polarisations, namely vertical transmitted and vertical received and horizontal transmitted and horizontal received, are also discussed.

The authors develop a novel phase-modulation based dual-function system with joint radar and communication platforms. A bank of transmit beamforming weight vectors is designed such that they form the same transmit power radiation patterns, whereas the phase associated with each transmit beam towards the intended communication directions belongs to a certain phase constellation. During each radar pulse, a binary sequence is mapped into one point of the constellation which, in turn, is embedded into the radar emission by selecting the transmit weight vectors associated with that constellation point. The communication receiver detects the phase of the received signal and uses it to decode the embedded binary sequence. The proposed technique allows information delivery to the intended communication receiver regardless of whether it is located in the sidelobe region or within the main radar beam. Three signalling strategies are proposed which can be used to achieve coherent communications, non-coherent communications, and non-coherent broadcasting, respectively. It is verified that the proposed method provides improved bit error performance as compared to previously reported sidelobe modulation based dual-functionality techniques.

This study proposes a new ballistic missile warhead tracking algorithm for enhancing the anti-ballistic capability of a wideband frequency-modulated continuous-wave (FMCW) radar seeker. Under the assumption that the target has a single dominant scattering point, the existing FMCW radars usually adopt the peak tracking algorithm which tracks the scattering point with maximum power in frequency spectra in order to calculate its range/range rate information. However, this method is not applicable for ballistic target tracking using wideband FMCW with high range resolution. This is because the return echoes of ballistic target obtained from wideband FMCW radar seeker are generally represented by many scattering points and, moreover, the most significant scattering point often does not appear at the warhead section of concern. To overcome this limitation, an inventive data association scheme is newly devised by exploiting the geometric constraints. This constraints is derived from the observation that the geometric relation between the scattering points corresponding to the warhead and body sections tends to be roughly determined by the velocity vector of the ballistic missile. Since it could be used for evaluating association hypotheses, the proposed geometric data association tracking filter can drastically improve the warhead tracking performance in practice. Through computer simulations, the effectiveness of the proposed method is demonstrated.

Frequency diverse array (FDA) can provide a range-angle-dependent beam by controlling the frequency increment between adjacent elements. However, in conventional FDA, the frequency increment is fixed, resulting in a significant restrain of jamming performance in various environments. With the above finding, the authors propose an optimal frequency increment selection method by maximising the output signal-to-interference-plus-noise ratio in each coherent processing interval for multiple-input–multiple-output-FDA radar. In addition, the discrimination method for the target and the jamming is studied. Simulation results demonstrate that jamming can be effectively discriminated and suppressed with the radar configuration, and better performance is achieved with an optimal frequency increment.

The multi-path phenomenon can degrade the direction estimation of arrival in radar systems. Without considering this phenomenon, it is possible to consider the multi-path signal in place of the true signal. In this study, a practical model is developed for radar signals received in flat areas such as deserts. After the modelling process including the presentation of received signals and their formulation, the authors suggest the maximum likelihood (ML) decision to estimate the elevation angle of the targets. Afterwards, by considering the relevant issues, they promote the ML method over the maximum a posteriori decision and the windowing decision. This model and its estimation are very useful in practical applications, particularly in low peak power and low probability of intercept radars.

The applicability of interferometric inverse synthetic aperture radar (InISAR) techniques to images reconstructed via compressive sensing (CS)-based algorithms is investigated. Specifically, the three-dimensional (3D) reconstruction algorithm is applied after exploiting CS for data compression and image reconstruction. The InISAR signal model is derived and formalised in a CS framework. A comparison between conventional CS reconstruction and global sparsity constrained reconstruction techniques is performed for different compression rates and different signal-to-noise ratio conditions. Performances on the 2D and 3D reconstructions are evaluated. Results obtained on real data acquired during the NATO-SET 196 trial are shown.

An innovative approach based on the variational sparse Bayesian learning method is presented for the optimal design of interleaved linear arrays with a fewer number of subarray elements. In this study, with a consideration of practical array assembling, the acceptable minimum element spacing can be freely adjusted and used to construct the constrained regions for subarray interleaving. The design of interleaved sparse arrays is then recast as a sparse constrained optimisation problem under region constraint, and a fast relevance vector machine is utilised to estimate the sparse weight vectors, with which the number of the selected antenna elements as well as their positions and excitations can be obtained. A set of representative numerical simulations validate the high accuracy and computational efficiency of the proposed method.

Through-the-wall radar systems are typically plagued by harsh clutter which confounds the detection and classification of targets. This study approaches the problem of separating humans from indoor targets through machine learning techniques by implementing support vector machine (SVM) classifiers. Simulated and experimental data are used to evaluate the suitability of SVMs for target classification via their spectral characteristics. Numerical modelling of humans and indoor clutter was accomplished through the finite-difference time-domain (FDTD) method. The radar cross-sections for the various objects are acquired over a wide range of frequencies, polarisations, aspect angles, and material properties. Furthermore, the spectral properties of humans and clutter targets are measured using a network analyser. The potentials of the acquired simulated and experimental spectral characteristics at different frequencies and polarisations are explored for target classification using SVMs. Feature selection algorithms are also investigated to reduce the redundancy and model complexity. Finally, the classification performance is assessed in the presence of additive white Gaussian noise and through various wall materials.

In this study, the optimum coherent detector in homogeneous K-distributed clutter is derived. It is named the Hom-OKD to show its relationship with the optimum K-distributed detector (OKD) under highly heterogeneous K-distributed clutter. The Hom-OKD has a different expression from the OKD because the received vectors at the reference cells are used as a part of the hypotheses. The Hom-OKD is shown to be constant false alarm rate (CFAR) with respect to speckle covariance matrix, Doppler steering vector, and the scale parameter of the gamma texture. Moreover, the adaptive Hom-OKD using a non-iterative maximum likelihood estimate of speckle covariance matrix is given and is proved to have the same CFAR property as the Hom-OKD. Under homogeneous K-distributed clutter environments, the adaptive Hom-OKD provides better detection performance than the adaptive OKD owing to the fact that the texture information conveyed by the secondary data at the reference cells is fully exploited. Simulated experiments are used to verify the optimality of the Hom-OKD. Experiments using real sea clutter data are made to show that the adaptive OKD and adaptive Hom-OKD are complementary. The former surpasses the latter for highly heterogeneous sea clutter and the situation is converse for homogeneous sea clutter.

In this study, the authors propose a new broadband target detection algorithm for FM-based passive bistatic radar systems, which simultaneously exploits multiple FM radio channels transmitted by the same transmitting station. It is shown that the joint exploitation of the signals of opportunity received at multiple carrier frequencies improves target detection capability as well as the target range resolution. In addition, the proposed detection algorithm exploits all available information making the detection performance robust against time-varying program content broadcast by the individual FM radio channels. Therefore, after formulating the broadband target detection problem as a composite hypothesis test, they derive a broadband uniformly most powerful invariant (B-UMPI) test, together with a closed-form expression for a statistical threshold that allows for automatic detection. To get a better insight into the detection performance of the proposed detector, a close-form expression for the probability of detection is also derived. In addition, they analytically show that how the range resolution of the proposed broadband target detection algorithm improves. They also analytically obtain an integrated sidelobe level ratio at the output of the B-UMPI statistics to show an improvement in the target detection quality. Finally, they provide some simulation examples to validate the authors’ theoretical analysis as well as to show the improvement in the target detection capability and the target range resolution in the broadband FM-based passive bistatic radar systems.

In this study, a novel adaptive algorithm, termed extended recursive maximum correntropy (EX-RMC), is proposed to estimate the time delay of automatic dependent surveillance-broadcast (ADS-B), an up-to-date surveillance technology in civil aviation. The proposed algorithm is designed to deal with both the random time-varying delay and the impulsive noise. The superiority of EX-RMC in tracking time-varying state of an unknown plant has been verified by simulation experiments. On-site experiments are also carried out to demonstrate the practical value of the ADS-B time delay estimation method based on EX-RMC.

Skywave over-the-horizon radar (OTHR) plays an important role in the out-of-sight surveillance. However, the operational performance will severely degrade due to transient interferences. Previous works concentrate on time-domain methods to achieve transient interference suppression. However, time-domain methods are usually time-consuming and suffer performance loss when several discontinuous transient interference segments occupy parts of coherent processing interval. In fact, most transient interferences are spatially structured and can be cancelled at adaptive beamforming stage. In this study, the adaptive beamforming is adopted and a space-time cascade processing procedure is proposed for transient interference suppression. In the procedure, the sidelobe transient interferences are cancelled by adaptive beamforming and the mainlobe transient interferences are suppressed by a modified time-domain method. The proposed procedure is in accordance with the regular processing procedure of skywave OTHR and can provide reliable performance with small extra computational costs. The performance of the proposed procedure is evaluated using experimental data recorded by a trial skywave OTHR and the results verify the feasibility and effectiveness of the proposed procedure.