The goal of maritime situational awareness (MSA) is to provide a seamless wide-area operational picture of ship traffic in coastal areas and the oceans in real time. Radar is a central sensing modality for MSA. In particular, oceanographic high-frequency surface-wave (HFSW) radars are attractive for surveying large sea areas at over-the-horizon distances, due to their low environmental footprint and low power requirements. However, their design is not optimal for the challenging conditions prevalent in MSA applications, thus calling for the development of dedicated information fusion and multisensor-multitarget tracking algorithms. In this study, the authors show how the multisensor-multitarget tracking problem can be formulated in a Bayesian framework and efficiently solved by running the loopy sum-product algorithm on a suitably devised factor graph. Compared to previously proposed methods, this approach is advantageous in terms of estimation accuracy, computational complexity, implementation flexibility, and scalability. Moreover, its performance can be further enhanced by estimating unknown model parameters in an online fashion and by fusing automatic identification system (AIS) data and context-based information. The effectiveness of the proposed Bayesian multisensor-multitarget tracking and information fusion algorithms is demonstrated through experimental results based on simulated data as well as real HFSW radar data and real AIS data.

Ionospheric phase contamination deteriorates the coherence of high-frequency echoes, reducing the detection performance of multiple-input multiple-output over-the-horizon radar (MIMO-OTHR) on slow ships. A high-precision phase decontamination method approximately extracts a single-frequency reference signal from the echo in advance, called the calibration signal, which is completed in the Doppler domain. When severe ionospheric phase contamination causes the echo Bragg peaks to overlap, this operation is difficult to achieve. To solve this problem, the authors transform calibration signal extraction into a sparse decomposition of a short-time sequence vector set and solve it by an iterative method. This process is based on the sparsity of the short-time sequence in the Doppler domain, as well as the reversibility between the original data vector, accumulated after a long coherence time, and its corresponding short-time sequence vector set, obtained by sliding window segmentation of the original data vector. Then the ionospheric phase contamination is extracted from the calibration signal to compensate for the original echo. Compared with existing methods, the proposed method can adaptively extract a calibration signal to achieve high-precision phase compensation of a MIMO-OTHR echo while adding a more robust performance against noise when considering overlapping Bragg peaks formed by serious ionospheric phase contamination.

Synthetic aperture radar (SAR) image classification is one of the most important subjects in automatic target recognition. Therefore, identifying the correct class of targets has significant importance to take a decision. Recently, several deep learning techniques, especially the convolutional neural networks (CNNs), have improved the SAR images classification performance due to its powerful perspective of feature learning and reasoning. Yet, CNN's generally need a huge amount of data for training and do not accurately manage the transformations in the input data. These drawbacks are overcome using a relatively new deep learning approach called capsule networks (CapsNets). In this study, the authors propose a method that adapts and incorporates CapsNet for the SAR image classification problem and improve recognition accuracy through a dual convolution CapsNet framework. Results obtained while experimenting on the moving and stationary target acquisition and recognition data set prove the effectiveness and the robustness of the proposed framework. The proposed experimental results demonstrate the superiority of the employed method overcoming both CNNs and CapsNet separate methods in term of classification accuracy.

In order to realise the fast detection of ships in synthetic aperture radar (SAR) images, a detection method combining visual saliency and a cascade convolutional neural network (CNN) is proposed. First, based on visual saliency, a multiscale spectral residual model is designed for realising the fast detection of ship candidate regions. Then, a cascaded CNN is designed, which consists of two convolution networks, namely, the front-end shallow CNN, which is used to quickly exclude obvious non-ship candidates and classify the ship candidates according to the ship orientation, and the back-end deep CNN, which is used to detect high-probability candidate regions with rotatable boundary boxes. The whole structure can realise the fast detection and precise positioning of ships with an arbitrary orientation. Finally, the authors conduct detailed experiments on the SAR ship image dataset. The experimental results show that the proposed method can effectively improve the detection accuracy of ships, ensuring the detection efficiency in SAR images.

By forging global navigation satellite system (GNSS) signals similar to authentic ones, a spoofer can make receivers track forged signals (spoofing signals) and generate wrong position, velocity and time results. Receiver autonomous integrity monitoring (RAIM) can be extended to the field of spoofing detection and exclusion (SDE). However, it is well known that when there are six or more signals and only one spoofing signal among them, RAIM can effectively exclude the spoofing signal. In this study, based on maximum likelihood estimation (MLE) theory and the idea of the traverse, one-dimensional traversal MLE-RAIM (TMRAIM) is proposed, which can exclude multiple spoofing signals. Theoretically, the influence of spoofing biases will be reflected in pseudorange residuals, and then affect the probability distribution of the parity vector. Through MLE deduction, the authors can find corresponding spoofing signals which are relevant with the maximum probability of the parity vector only once under the supposed number of spoofing signals. By just traversing the number of spoofing signals, TMRAIM can run effectively on real-time GNSS receivers with low complexity. The SDE ability and time complexity are analysed in detail and two field experiments are constructed. Experimental results demonstrate the method is feasible and effective for anti-spoofing applications.

The traditional method considers the Sun as a rigid sphere and diffuse radiator, and neglects the solar characteristics. However, this leads to high bias in the measurement model. In this study, to build a more realistic bias model and a statistical model of noise, the authors consider three solar characteristics, including the solar differential rotation, Lambert radiator and small-scale solar activities. After mathematical derivation, the bias model of the solar Doppler difference measurement considering the solar differential rotation and Lambert radiator is built. Simulation results demonstrate that the difference between the proposed bias model and the traditional one is on the order of 0.1–1 m/s, which is slowly-varying and non-ignorable. Therefore, the bias model considering the solar differential rotation and Lambert radiator is meaningful. After mathematical analysis, they also find that small-scale solar activities cause an extremely low error in the solar Doppler difference measurement. Moreover, the noise of the Doppler difference is within 2 m/s. Due to the randomness of small-scale solar activities, the extremely low errors caused by activities can be regarded as noises. These results provide a reference for the measurement modelling of the solar Doppler difference velocimetry navigation.

Costas signals are widely used in modern radar systems as frequency coded. The Costas-code-based waveform can achieve a nearly thumbtack shape of the ambiguity function (AF). However, the highest AF's sidelobe levels (SLLs) of Costas waveform equals just 1/ N times of its main lobe level height, where N is the length of the Costas code. In this study, the authors suggest a generalised Costas waveform as a ‘burst’ of weighted pulses having variable time spacing between them, calling it as weighted variable time spacing Costas (WVTSC). They computed the AF analytical formulas of the designed WVTSC. They developed an adaptation model to adapt the WVTSC to target scattering coefficients, using the interior-point algorithm. The adaptive WVTSC yields much better Doppler cut and improved delay cut in the main lobe area of the AF, in the sense of SLL reduction. Moreover, they constructed an optimisation model resulting in optimal variable time spacing between the sub-pulses. The optimised WVTSC also yields considerable improvement in the AF's recurrent SLLs, using the genetic algorithm. The adapted optimised WVTSC design is proven, through simulation, to have better sidelobe performance than that of the initial waveform, without influencing the original delay-Doppler resolution, and without need to increase N .

The electronic support measure has a pivotal role in electronic warfare. Radar signal deinterleaving achieves complex radar pulse streams to deinterleave by analysing the intercepted radar emitter signals and estimating the pulse repetition interval (PRI) values. At present, most of the deinterleaving algorithms use the classical model of pre-sorting combined with the PRI deinterleaving algorithm, which is challenging to adapt to the current radar environment. The agility of pulse parameters increases the error rate; the existence of PRI jitter and pulse missing leads to the sub-harmonic problem and low estimation accuracy; the overall algorithm real time is mediocre. To overcome these disadvantages, the authors propose a synthetic algorithm incorporating the statistical histogram into pre-sorting, combining the improved sequence difference histogram algorithm with the multi-width PRI transform in PRI deinterleaving. The simulation results show that the algorithm can accurately estimate the PRI values in a short time and detect various radar signals. C language experiments validate the algorithm in real time.

In this study, a multi-polarised two-dimensional (2D) planar array of sparsely located vector antennas (VAs) is designed for the block sparse representation (SR)-based 2D direction finding and polarisation parameter estimation of wideband signals. In order to alleviate the inter-VA mutual coupling effect, the minimum inter-VA spacing of the 2D sparse array is constrained to be no less than one wavelength that corresponds to the highest signal frequency. To reduce the computational complexity of parameter estimation, the 2D block SR model for the 2D difference coarray output at a certain frequency bin is established, under which the two direction cosine terms for 2D direction-of-arrival (DOA) estimation are decoupled with the polarimetric terms. This enables separated but simultaneous 2D direction finding and polarisation parameter estimation with the newly developed joint 2D block orthogonal matching pursuit (Joint-2D-BOMP) subband fused sparse recovery algorithm. Moreover, with the use of two spatial only (polarisation independent) over-complete dictionaries, the representation dimension of the new 2D block SR model is greatly reduced as compared with the traditional space-polarisation joint SR model. The efficacy of the presented VA array geometry and the associated parameter estimation method is validated by computer simulations.

Robust object detection in sonar images is an important task for underwater exploration, navigation and mapping. Current methods make assumptions about the shape, highlight or shadow of an object, which may be invalid for some environments or targets. We focus on the area of feature extraction-based detection, which does not rely on information about the shape of the target, towards a robust framework for target detection for a variety of seabed structures and target types. The proposed framework first estimates the seabed type from the spatial distribution of features to determine the set of optimal parameters, and then obtains a set of features which are filtered according to intensity and distribution to yield a detection decision. The proposed method also provides a means to determine the seabed type, and a machine-learning based methodology to choose the feature detectors' parameters to match the evaluated seabed type. We report the performance of a variety of feature detectors for a simulated environment and of one feature detector for real sonar images. Results show the importance of choosing the parameters of the feature extractors based on the current environmental conditions and the proposed method obtains a favourable tradeoff between detection and false alarm rates.

A multi-sensor fusion approach for simultaneous localisation and mapping (SLAM) based on a bio-inspired polarised skylight sensor is presented in this study. The innovation of the proposed approach is that a newly designed bio-inspired polarised skylight sensor, which is inspired by the navigation principle of desert ant, is introduced to improve the accuracy of SLAM. The measurement equations based on a polarised skylight sensor and a lidar are derived to obtain the orientation and position of the mobile robot and landmarks. Three kinds of non-linear filters, extended Kalman filter (EKF), unscented Kalman filter, and particle filter, are adopted and compared to fuse the polarised skylight sensor, lidar, and odometry to estimate the position, orientation, and map in the experiments. Simulation tests and experiments are conducted to validate the effectiveness of the proposed method. The simulations show that the EKF–SLAM with the polarised skylight sensor reduces the error of localisation about 30% and the error of mapping about 25%. Experiments indicate that the proposed EKF–SLAM approach can reduce the error of position and the heading angle, which verifies the proposed EKF–SLAM method, can be used for the outdoor with the low-cost multi-sensor.

Aircraft wake is a pair of strong counter-rotating vortices generated behind a flying aircraft, which might be dangerous to the following aircrafts. The real-time and precise locating of aircraft wake vortex-core positions is essential in the aviation safety field, especially in airport traffic management. This study proposes a two-step vortex-core locating method which uses the Gabor filter to make a preliminary locating and the velocity range distribution to make a fine locating consequently. The combination of Gabor filter and Doppler velocity distribution can improve the method's adaptability to complex background wind field and mitigate the impact of noise. Numerical examples and field detection campaigns from Changsha Huanghua International Airport and Hong Kong International Airport have well verified the good performance of the method, in terms of both accuracy and real time.

In a vector tracking loop (VTL) architecture, non-linearities exist in discriminator functions and pseudo-range/pseudo-range rate measurement expressions. Generally, normalisation functions are used in discriminators to export the desired code phase or carrier frequency error and the extended Kalman filter is adopted to estimate receiver's states. This process could be accurate enough when the code phase or carrier frequency error approaches zero in the signal moderate environment but begins to distort due to non-linearity when the tracking errors become large in harsh situations. This finally narrows the applicable range of VTL. To overcome this issue, a square-root cubature Kalman filter (CKF)-based VTL is designed in this study. The discriminator functions are employed directly as measurements of navigation filter, and the non-linear expressions of discriminator functions in terms of the receiver's position, velocity, and time states are derived without normalisation. Then the CKF, which is competitive in high-dimensional non-linear systems, is employed in its square-root version to estimate the position, velocity, acceleration, and time states of the receiver. Comparison trial results between traditional and proposed VTL illustrate that the proposed algorithm can not only keep a superior tracking accuracy but also improves the tracking stability of VTL in <20 dB-Hz signal harsh circumstances.

Some new problems in distributed inverse synthetic aperture radar (ISAR) fusion imaging may be caused by constant frequency errors (CFE) due to the imperfect match of two independent oscillators. In this study, the effects of the CFE on distributed ISAR coherent fusion imaging are first analysed, which show that the CFE will produce false images (incorrect scatterer number and position) in distributed ISAR fusion imaging. To eliminate the effects of the CFE, a novel CFE calibration scheme is devised by exploiting the echo signal phase relationship between different sensors and making some reasonable approximations. The echo signal of the active radar (transmit and receive) is used as a reference signal. After some derivation, the phase relationship between the reference signal and other signals with CFE can be derived. Based on the derived phase relationship, the CFE of each radar will be estimated by multiple signal classification algorithm. Then, the estimated CFE value is further used to calibrate the CFE of each radar before signal coherent fusion. As a result, the false image is eliminated. Simulation results prove the authors theoretical analysis and the effectiveness of their method.

Space–time adaptive processor based on power inversion (PI) criterion can effectively suppress interference for the global navigation satellite system (GNSS) receiver when the steering vector of the GNSS signal is unknown. However, existing space–time PI processing methods will introduce meter-level pseudo-code tracking biases into the GNSS receiver measurements which cause several meter-level position errors. A distortionless pseudo-code tracking space–time PI algorithm is proposed. It could not only suppress interference, but also introduce no pseudo-code tracking biases. The major novelty of the proposed method is ensuring the symmetry of the output signal's cross-correlation function by constraining coefficients. Several experiments are implemented to test the performance of the proposed algorithm. For comparison, the results of the PI algorithm and the minimum variance distortionless response (MVDR) algorithm are also shown. Results show that for the PI algorithm and the MVDR algorithm, pseudo-code tracking biases are introduced in different experiments; however, for the proposed algorithm, there are no pseudo-code tracking biases in those experiments. As a result, the effectiveness of the proposed method is verified.

Air-to-ground ranging (AGR) is one of the important operation modes of an airborne monopulse radar system. Here, an approach for the detailed modelling and simulation of the AGR performance of such a radar system over complex terrain is proposed. Time-domain monopulse signals reflected by the terrain's surface were modelled using the dynamics of the airborne platform, clutter properties, and RF specifications of the radar, which included the operating frequency, antenna beam patterns, and pulse repetition time. The AGR performance was then numerically analysed by employing the modelled monopulse return signals and a digital elevation model of the terrain's surface. Several crucial factors affecting the accuracy of line-of-sight estimation, including shadowing effects and roll stabilisation, were addressed using numerical simulations of various scenarios.

Compared with conventional phased array radar, the collocated multiple-input multiple-out (C-MIMO) radar is envisioned to offer greater freedom in waveform design and can simultaneously manage different beams to track multiple targets in the simultaneous multi-beam (SM) working mode. In this study, aiming at the multiple targets tracking (MTT) problem, a joint beam selection integrated with power and bandwidth allocation (JSPBA) scheme for the C-MIMO radar is proposed in the SM mode. By incorporating the modified particle filter, the predicted conditional Cramér–Rao lower bound (PC-CRLB) is calculated, which gives a measure of the best achievable performance for targets tracking. Then, the optimisation model is established with the aim of improving the worst case of PC-CRLB to achieve better performance of the worst-case tracking. Next, a three-step solution method is proposed to solve the JSPBA problem by converting the non-convex problem into a series of convex problems. At last, after the solutions are fed back to the control centre to guide the beam generation in the next tracking epoch, a cognitive tracking system is established. Simulation results confirm the superiority of the proposed method in improving the MTT performance in C-MIMO radar system.

This article considers autonomous scanning (coverage) control of a two-dimensional (2D) surface embedded in 3D environments. The author uses a robot with rigidly mounted range sensors. The robot can measure its depth (height for aerial robot) and has narrow aperture rays near the central ray which is normal to the longitudinal axis of the robot. The goal of the robot is to scan (cover) a 2D surface embedded in 3D environments within a given range of depths. To achieve robust scanning performance in 3D environments, this article presents sliding mode controllers as well as switching controllers. According to the author's knowledge, this article is unique in developing 3D reactive scanning controllers considering a robot equipped with narrow aperture rays near the central ray. Extensive simulation results are utilised to demonstrate the effectiveness of the proposed controllers.

An underwater acoustical navigation (UWAN) system is widely used on an autonomous underwater vehicle (AUV). The navigation error prediction is of great significance to the design, use, and capability upgrade for the UWAN system. Owing to the trend of faster AUV and the need for higher navigation accuracy, traditional navigation models are mismatched. Besides, they adopt fixed input errors (including measurement errors of sound velocity, time delay, and array position), which are inconsistent with the real complex acoustic channel. As a consequence, the prediction accuracy of the navigation error is greatly reduced. The authors proposed a novel navigation error prediction method for UWAN based on AUV. Firstly, they established the acoustical navigation model for moving AUV and deduced the error prediction formula. Secondly, considering the complex underwater acoustic channel, they applied the variable error models as the inputs of the prediction formula. Finally, simulations and experiments in different situations verified the effectiveness of the proposed method. The method will provide effective theoretical support for the performance evaluation of the UWAN system based on moving AUV.

Performance of global navigation satellite systems (GNSSs) mounted on aerial platforms could be degraded by the presence of jamming or spoofing threats. Detection of jamming and spoofing is essential considering practical applications of satellite navigation in passenger aircrafts, unmanned aerial vehicles (UAVs), helicopters and fighters. Different algorithms and methods have been proposed for detection of these threats; however, their usage has many limitations because of their demanding weight, size and computational complexity, when embedded on aerial systems. In this study, the authors develop a theoretical framework to detect the presence of the threat of UAVs. The idea is based on the fact that, due to the UAV motion, the samples of received signal power from a fixed threat and from a GNSS satellite have different empirical probability density functions. Moreover, by using two antennas (an omnidirectional and a down-tilted-directional), they introduce a new method to distinguish between aerial and ground-based threats. The proposed algorithms have a low-computational burden and can consider the fading loss as well. Simulation results show the superior performance of the proposed methods, in terms of detection and false alarm probability, compared to the existing methods.

In this study, the authors propose for the parameter estimation of chirp signals at low signal-to-noise ratios using the spectrum phase instead of the traditional temporal phase. First, the spectrum phase expression of chirp signals is derived, which demonstrates that the spectrum phase can be approximated as a quadratic function of the frequency. However, the spectrum phase is a wrapped phase modulo 2π, like the temporal phase. To obtain the unwrapped spectrum phase, the authors develop a robust phase unwrapping method and then perform linear regression on the unwrapped spectrum phase to obtain the estimates of the parameters. Finally, the estimates are refined by the O'Shea refinement strategy to reduce bias. The simulation results show that the proposed technique has a lower signal-to-noise ratio threshold and is less complex than the existing methods.