A helicopter may vibrate at high frequencies during its flight, which can give rise to paired echoes in the helicopter-borne synthetic aperture radar (SAR) imagery. This phenomenon substantially degrades the quality of SAR images since it may generate unwanted ghost targets in the images. In this study, the authors propose a method consisting of three steps to suppress paired echoes in helicopter-borne SAR imagery based on the characteristics of the helicopter vibration. The proposed method can suppress paired echoes with high efficiency and accuracy in the case of vibrations with varying amplitude. Furthermore, this method does not require the presence of isolated or strong points in a range cell. Both simulated and acquired data are processed to demonstrate the effectiveness of the proposed algorithm.

In this study, the authors propose an effective two-stage direction-of-arrival (DOA) estimation method for low signal-to-noise (SNR) signals in automotive radar systems. When antenna elements in the array receive low SNR signals, the performance of subspace-based DOA estimation algorithms is degraded. Concerning this case, they propose an enhanced DOA estimation method that offers better angular resolution and estimation performance. The authors’ proposed method is comprised of two stages. In the first stage, they roughly estimate the DOA using conventional subspace-based DOA estimation algorithms. Thereafter, the fine DOA estimation is performed in the next stage. The fine estimation includes a received signal calibration method using a priori information acquired from the previous stage. From simulation results, in terms of root mean square error and resolution probability, their proposed method exhibits a DOA estimation performance that is superior to that of the conventional method. Moreover, with actual measurement data, they verify that the proposed method can be applied to practical automotive radar systems.

Inspired by collaborative representation classifier (CRC), a Wishart distance-based joint CRC (W-JCRC) is proposed for polarimetric synthetic aperture radar (PolSAR) image classification. Since that neighbouring pixels usually belong to the same category with high probability, they can be simultaneously represented via a joint representation model of linear combinations of labelled samples. The joint collaborative representation of neighbouring pixels can overcome the influence of speckle noise at the same time. Considering the statistical property of PolSAR data, a weighted regularisation term with revised Wishart distance is designed to contain the correlations between unlabelled and labelled samples. The coefficients of representation are estimated by an -norm minimisation derived closed-form solution. In the experiments, three real PolSAR images are applied to evaluate the performance, and the experimental results demonstrate that the proposed method is able to improve classification accuracies compared with other state-of-the-art methods.

Four improved chirp scaling algorithms (CSAs) are proposed to reconstruct images from synthetic aperture radar (SAR) data received at high squint angles, in which a larger synthetic aperture is required to receive sufficient amount of data for image reconstruction, and the range migration also becomes more serious, demanding more computational load and larger memory size. The proposed methods reconstruct better SAR images with less computational load and memory than the conventional CSA, which are verified by simulations.

One-step passive localisation methods are verified to outperform the conventional two-step methods in terms of estimation accuracy and identifiability. However, the existing one-step algorithms still cannot work in the scenario where the number of emitters exceeds the total number of sensors from all the base stations. Based on a spatio-temporal processing procedure, the authors propose a novel localisation model, which has the form of multi-dimensional harmonic retrieval. By utilising multi-dimensional spectrum estimation techniques, the underdetermined localisation problem can be handled thanks to the increase of model's degree of freedom. To further improve the identifiability, a nested-array-based localisation model is also given based on the multi-dimensional processing framework. Simulation results demonstrate that the novel-model-based Capon algorithm can achieve higher localisation accuracy without the prior knowledge about the number of emitters, and is robust to the correlated noise.

Performance bounds on range resolution and range sidelobe level are built for waveforms with spectral notches. In this study, the authors use the integrated mainlobe energy (IME) and integrated sidelobe energy (ISE) of the autocorrelation function to assess the performance of range resolution and range sidelobe level, respectively. By considering the minimisation of the weighted sum of IME and ISE under given spectral constraints, they derive the optimal power spectrum density condition and subsequently calculate the best achievable IME and ISE. With this study, the radar system can (i) analyse the trade-off relationship between range resolution and range sidelobe level and thus better balance them in designing the transmit waveform; (ii) assess the fitness of potential working bands and thus select the best one to use.

Nyquist folding receiver (NYFR) can realize the wideband receiving with a small amount of equipment. As a tradeoff between the amount of equipment and the complexity of the signal processing, the NYFR output is usually a hybrid modulated signal. Therefore, it is necessary to investigate the parameter estimation method of the signal intercepted by the NYFR. Binary phase shift keying (BPSK) signal is a typical low probability of intercept radar signal and the authors focus on the parameter estimation of the BPSK radar signal intercepted by the NYFR. Firstly, matching component function is proposed to obtain Nyquist zone (NZ) index. Then, the Hankel-singular value decomposition (SVD) method is introduced to estimate the code rate of the BPSK signal. The efficacy of the Hankel-SVD method is proved in a mathematical way. Several numerical simulations demonstrate the merits of the proposed approach and finally, a brief investigation for case of the multiple BPSK signals intercepted by the NYFR is given. For NZ index estimation, the computational complexity of the matching component function is low and it has no need to use squaring process. In addition, the code rate estimation using the Hankel-SVD method is more accurate compared with the wavelet transform.

This study investigates the effects of antenna rotation on range, radial velocity, and azimuth estimation during coherent integration in linear frequency-modulated continuous wave (LFMCW) radars. When antenna scans during the coherent integration time, theoretical analysis indicates that the angle sensing with the traditional sum-difference amplitude-comparison (SDAC) monopulse method and the range, radial velocity estimation with interpolation method are different from that in staring mode. Simulation demonstrates that the estimation performance is heavily impaired by the antenna rotation. In this study, a generalised mathematical model of coherent integration for LFMCW radar is established. Furthermore, a high-precision estimation method, which consists of generalised SDAC monopulse method, radial velocity estimator with modified interpolation curves, and range estimator refined by Doppler frequency, is proposed. A complete signal-processing scheme is proposed for the accurate estimation of the target parameter in mechanical scanning LFMCW radar. A sub-optimal method is provided for practical implementation. Experimental and simulation results show that the proposed method achieves high accuracy and robustness.

The computational complexity of wideband sparse representation (SR) greatly restricts the application of SR-based wideband direction-of-arrival (DOA) estimation in practical system. Here, an efficient method for wideband direction finding based on SR is proposed. This method combines the focusing operation with weighted subspace fitting (WSF) to not only decrease the computational complexity but also improve the performance of DOA estimation. Exploiting the result of the focusing operation, the covariance matrix at the focusing frequency can be obtained and used as the data for sparse recovery to get wideband DOA estimates. The WSF is employed to reduce the sensitivity to the noise and the regularisation parameter is given by the asymptotic distribution of the WSF cost function. Simulations are provided to show the efficiency and performance of the proposed method.

To overcome the limitations of the conventional motion tracking and localisation methods: environmental constraints of the infra-based approaches and long-term inaccuracy of the wearable sensor-based approaches, an algorithm for simultaneous motion tracking and localisation (SMaL) of a person, namely multiple inertial measurement unit (IMU)-based SMaL (MIbS), is proposed. A filtering method for the multiple IMUs/depth camera integration (MDI) is also proposed to add long-term localisation accuracy to the MIbS. The performance of the proposed MIbS and MDI is verified experimentally. The experimental results show that the proposed MIbS can track the motion of the person with the wearable sensor system accurately, and can localise the person on the local coordinate frame of the testbed. The encouraging fact is that the localisation errors caused by the MIbS grow slowly with time. Also, the growing localisation errors can be compensated by the MDI.

Chaff is composed of small and thin pieces of glass fibres plated with aluminium or silver and is dispersed in the air like a cloud from air fighters or warships. The length of a chaff fibre is generally much smaller than the radar resolution and its orientation is random in chaff clouds. Therefore, chaff clouds can be viewed as anisotropic needle-like particles. For such kind of target, according to the polarimetric decomposition theorem, identifying the dominant physical scattering mechanism is the polarimetric scattering parameter, which is a roll-invariant parameter. As a result, this parameter is introduced to investigate the scattering characteristics of chaff clouds here. Theoretical analysis shows that this parameter of chaff clouds with three typical orientation distributions varies between and . Moreover, when the azimuth angles of chaff clouds meet the uniform distribution, the correlation of chaff clouds between the co- and cross-polarised channels vanishes. To this end, the polarisation scattering characteristics of chaff clouds mentioned above are validated by numerical simulation, and a study is also carried out to compare the chaff clouds with other targets, which confirms the feasibility and real-time of chaff jamming recognition by support vector machine classification method.

Detecting low-speed quiet targets such as divers and underwater-unmanned-vehicles in littoral environment by using active sonar is becoming increasingly attractive. When using a Doppler insensitive pulsed linearly frequency modulated signal, the high-level clutters which might arise from underwater physical scatterers will lead to excessive false alarm rates and limit the detection performance. However, Doppler sensitive waveforms such as binary phase shift keying have the capability of filtering clutters and degrading false alarms. Thus, the question how to estimate the clutter-suppressing performance of waveforms is essential for sonar system design. In this study, the clutter-rejecting principle of waveform is theoretically introduced firstly. Then, based on the reverberation statistical features, this study proposes two methods, the Doppler-statistic method and envelope-statistic method, separately, to estimate and evaluate the clutter suppressing performance of waveforms. Finally, the methods are verified by lake experiments. It is proved that the first method has the capacity of calculating the confidence probability of suppressing clutters by a waveform, and through the second method, the clutter rejecting performance of waveforms can be evaluated and verified. The methods can be used for selecting and designing waveforms to reduce false alarms and improve detection performance.

The multiple-input multiple-output (MIMO) radar combines ultra-wide angular illumination on transmit and narrow transmission angular resolution after processing on receive. Orthogonal waveforms are needed in theory for all independent transmitters, but this property is usually not achievable in practice, therefore leading to a degraded ambiguity function. In this study, the authors focus on phase codes waveform, which gives quasi-optimal angle and range resolution. On the other hand, they induce a harmful ‘floor’ of high-level sidelobes in the range–angle domain. The authors propose to apply the orthogonal matching pursuit (OMP) to the MIMO radar signals. Therefore, it becomes theoretically possible to detect targets initially hidden by high-level sidelobes. Nevertheless, the authors identify phenomenon (grid granularity, neighbour influence) that lead to possible localisation error resulting in ghost targets. Considering this, the authors propose an extension of the OMP rejection step that includes, in addition to the expected target position, the neighbour positions within the uncertainty area. The authors show that this extension allows them to suppress the sensibility of the OMP to localisation error, even in case of very close targets. Finally, the authors demonstrate the performance of the proposed method on simulations as well as on experimental MIMO radar signals.

There exist few rapid reconstruction methods for applications that use multistatic radar for targets in the near field such as vehicle navigation and personnel screening. Fourier accelerated multistatic imaging (FAMI) is a method that combines the rapid reconstruction speed of Fourier-based imaging algorithms with the flexibility of algebraic algorithms that allow arbitrary radiation patterns and antenna placement. Many of these new applications image in the near field of the individual antennas of a multistatic array, however, FAMI contains approximations that require the target to be in the far field of each antenna. Stretched-phase FAMI (SP-FAMI) overcomes this limitation and allows imaging in the Fresnel region of the antennas so that targets may be very close to the array while retaining the computational benefits of Fourier reconstruction. As radar wavelengths decrease with a corresponding increase in resolution, SP-FAMI allows the imaging of nearby targets that would otherwise be too close for FAMI to image accurately.

Classical stretch processing (SP) obtains high range resolution by compressing large bandwidth signals with narrowband receivers using lower rate analogue-to-digital converters. SP achieves the resolution of the large bandwidth signal by focusing into a limited range window, and by deramping in the analogue domain. SP offers moderate data rate for signal processing for high bandwidth waveforms. Furthermore, if the scene in the examined window is sparse, compressive sensing (CS)-based techniques have the potential to further decrease the required number of measurements. However, CS-based reconstructions are highly affected by model mismatches such as targets that are off-the-grid. This study proposes a sparsity-based iterative parameter perturbation technique for SP that is robust to targets off-the-grid in range or Doppler. The error between reconstructed and actual scenes is measured using Earth mover's distance metric. Performance analyses of the proposed technique are compared with classical CS and SP techniques in terms of data rate, resolution and signal-to-noise ratio. It is shown through simulations that the proposed technique offers robust and high-resolution reconstructions for the same data rate compared with both classical SP- and CS-based techniques.