The design of periodic binary sequence sets for multiple-input multiple-output (MIMO) phase-modulated continuous wave (PMCW) automotive radar is crucial to its performance. While the corresponding optimization problem is NP-hard and high-dimensional, and many existing computational algorithms are not suitable for designing binary sequence sets with long lengths due to their high computational complexities. In this paper, we develop a new algorithm for designing periodic binary sequence sets with good auto- and cross-correlation properties for MIMO PMCW radar. Fast Fourier transform (FFT) can be used to implement the proposed algorithm, making the algorithm computationally efficient and hence can be used to design binary sequence sets with quite long lengths. Several numerical examples are provided to demonstrate the performance of the proposed algorithm.
Automatic modulation classification (AMC) plays an important role in the development of cognitive radio and cognitive radar systems. Due to the distinct aims of radar and communication systems, their commonly used modulation techniques may differ significantly. However, AMC in the radar and communication domains can be closely aligned. Considering the widespread applications of deep learning architectures, particularly, convolutional neural networks (CNN) in classification problems, we propose a CNN-based framework for the joint classification of communication and radar signals. In the proposed framework, a CNN is first trained by signals in the in- phase and quadrature domain and then another CNN is trained by using constellation diagrams to differentiate between close modulations. A publicly available benchmark dataset, which consists of different modulation schemes in the communication domain, has been augmented with our simulated linear frequency modulated radar signals under the same noise condition to validate the accuracy of the proposed framework.
Radars are commonly used in automotive applications to provide information on a target's location and velocity simultaneously. The multiple-input and multiple-output technology has been applied to improve spatial awareness, but impacts the Doppler sampling rate and causes velocity ambiguity. An incorrect velocity indication can pose significant safety risks in the automotive industry. To address this, we propose a complex-valued neural network to retrieve the actual velocity from aliased Doppler response by parsing signal magnitudes and phases. On the artificial data, it achieves an accuracy of 99.3%, outperforming the conventional methods on the same data.
This paper presents a method to model monolithically integrated photonic radar transceiver (TRX) with optical local oscillator (LO) distribution in silicon germanium (SiGe) electronic photonic integrated circuits (EPICs). The model proposed approximates the behavior of the nonlinear scattering (S)-parameters and noise Figure of each building block of the TRX chipset by Laplace polynomials and hyperbolic tangent functions. The modular approach of the model allows to optimize hardware components with respect to the entire TRX system, and fault identification with reduced computational effort. The proposed method is validated using the first monolithically integrated photonic radar transceiver chipset and shows excellent agreement with the post layout simulation results and, including the photodiode (PD) bandwidth (BW) degradation, also with the measurements.
This paper describes the design of a high-gain ultra-wideband (UWB) linearly tapered slot PCB antenna for mm-wave radar applications. The antenna design was focused on matching differential microstrip to antenna input feed for the best VSWR. For better antenna parameters PTFE lens was added.The antenna is designed as a planar structure on a four-layer PCB on substrate Isola Astra® MT77 with dielectric constant D k = 3.0. The dimensions of the antenna without a lens are 18.9x9 mm. The basic simulation of the antenna was performed. The results show the proposed antenna directivity, gain, and low VSWR over the operating frequency. The VSWR of the antenna is under 2, and the designed operation frequencies of an antenna are from 25 GHz to 85 GHz. The maximum achieved gain is 17.6 dBi at 65 GHz.
To achieve higher resolution, radar systems are penetrating ever higher frequency ranges. Recently, the terahertz frequency range has become the focus of researchers because it offers a high frequency bandwidth. Here, we present a synthetic aperture radar (SAR) approach enabled by terahertz time-domain spectroscopy (THz-TDS) for 2D and 3D imaging. Due to the high bandwidth of more than 2 THz, a resolution finer than 500 μm can be realized.
This paper is about spectrum analyzer measurements in situ of different antennatype radiation patterns, including radar, using an Unmanned Aerial Vehicles (UAVs) platform. This measurement method offers a 3D complex solution for mapping the emissions of emitters such as radar, Wi-Fi communication antennas, and electrosmog of automobiles, smart cities, which subject gets increasing attention today and in the near future. Based on these facts, the goal of this research is to analyze design issues and create a drone-based platform that can autonomously "in situ" measure these electromagnetic spectrum components and then draws conclusions from the data obtained.
In this paper, a 3D chest detection framework for a driver seated in a vehicle mock-up is proposed. The aim of the project is to predict the 3D chest position of the driver while performing specific tasks in a fully autonomous vehicle. The proposed framework is implemented using convolutional neural network (CNN) forked-architecture with two frequency-modulated continuous wave (FMCW) radars and a Microsoft Azure kinect sensor. A dataset of 8,659 frames is built using the aforementioned sensors. Two CNN models are trained, tested and analysed using the dataset, demonstrating the possibility of using only radar data as input to predict human body key point in a vehicle compartment.
This paper presents a first example of experimental results of the application of Doppler beam sharpening (DBS) to enhance the resolution and detectability of maritime targets at a) 77 GHz using multiple-input multiple-output (MIMO) radar and b) 150 GHz using a real-aperture radar. The performance of DBS and MIMO-DBS beamforming have been evaluated with varied platform dynamics and test conditions to highlight the advantages of higher frequencies for the next generation of radar sensors. The applicability of DBS in maritime conditions has also been assessed.
In this paper, we present an approach for supervised training of a neural network operating on camera images to estimate the receive power of an automotive radar. The training requires the association of pixels from camera images and radar spectra via an image-warping layer. We evaluate the influence of forward and backward image warping, known from classical image processing, on the application of receive power estimation. It is found, that the implementation via backward image warping results in a biased estimation of the receive power whilst the forward image warp results in an unbiased estimation.
This paper presents a reinforcement learning approach for automatic adaptation of the process noise covariance (Q). The Q value plays a crucial role in estimating future state values within a Kalman filter tracking system. Proximal Policy Optimization (PPO), a state-of-the-art policy optimization algorithm, was employed to determine the optimal Q value that enhances tracking performance, as measured by Root Mean Square Error (RMSE). Our results demonstrate the successful learning capability of the PPO agent over time, enabling it to suggest the optimal Q value by effectively capturing the policy of appropriate rewards under varying environmental conditions. These outcomes were compared with those of a feed-forward neural network learning, the Castella innovation/ Q values mapping, and fixed Q values. The PPO algorithm yielded promising results. We employed the Stone Soup library to simulate ground truths, measurements, and the Kalman filter tracking process.
Human sensing has long been an actively studied problem due to its potential in a wide range of applications. However, most existing approaches depend on cameras that require proper light condition and also pose quite a challenge for privacy protection due to native nature of easy understanding by eyes in the visual data. As a promising alternative, mmWave radars have been explored recently as a viable alternative to provide precise human sensing capability without aforementioned concerns. In this paper, we propose to use commercial low-cost mmWave radars to perform multiple human sensing functionalities concurrently and accurately based on a multi-task learning architecture. We show that the proposed system is able to classify activities and identify human individuals at the same time such that question of “Who is doing what” can be asked. Our experiments show that an average accuracy of 98.4% and 97.6% are achieved respectively for the two tasks by the proposed method, implying its great potential for effective human sensing using mmWave radars.
High resolution Three-Dimensional (3D) image of space targets can be achieved under large relative rotational angle, providing precondition for high-precision recognition of space targets. Under large relative rotational angle, the large Migration Through Resolution Cells (MTRC) and the highly Doppler Frequency Modulation (DFM) arise in the Two-Dimensional (2D) Inverse Synthetic Aperture Radar (ISAR) imagery, which results that the existing 3D Interferometric ISAR (InISAR) imaging methods suffer from poor imaging quality. To address this issue, we propose a high resolution InISAR imaging method under large relative rotational angle with single baseline, which is excellent in providing high resolution 3D image of space targets and simplifying system structure compared with the traditional InISAR image methods. Firstly, Polar Format Algorithm (PFA) is performed to eliminate the impact of MTRC, and the second-order Local Polynomial Fourier Transform (LPFT) is applied to reduce the azimuth defocuing. Then, the cross-range scaling technique is adopted to obtain the accurate vertical and horizontal coordinates of the scatterers. Finally, the height coordinates of the targets can be obtained by interferometric processing. Simulational experiments demonstrate that the proposed method can achieve high resolution 3D imaging for space targets under large relative rotational angle with low hardware complexity.
This paper proposes a multi-agent technique for modeling and simulating multiple objects and multiple sensor tracking for drones. The goal is to achieve efficient resource allocation based on different preset cost functions that take into account several local and global attributes. The target dynamics, the sensor selection, and the measurements are simulated by way of an agent-based modeling framework called MESA. The results show an optimum resource allocation with an information-based cost function that is optimized by an Auction Algorithm.
Real-time, high-resolution radar measurements resulting from a system bandwidth of several GHz are important for many applications nowadays. Additionally, wireless data transmission of system - and scenario information with low rate plays an important role in such sensor systems. Therefore, it makes sense, especially for military applications such as synthetic aperture radar, to integrate a communication link into such a broadband radar system e.g. for friend-foe-identification. This publication describes the investigation of an efficient dual function system for broadband FMCW radar and single-carrier QPSK communication in frequency division multiplexing (FDM) for high-range applications. The transceiver architecture combines a 35 GHz FMCW radar with 2 GHz bandwidth and a QPSK modulator centered at 37.4 GHz with a data rate of 0.2 Mbit/s in FDM mode. The investigation of the influence of the radar chirp to the communication receiver and the QPSK-signal to the radar measurement is presented. The successful operation of an efficient joint broadband FMCW radar and single-carrier QPSK communications system for high-range applications is shown for a peak-to-average power ratio of theoretically 6 dB and a maximum bandwidth of 3.4 GHz of the transmit signal.
The classical noise radar is most often considered to be a kind of continuous wave sensor, transmitting the sounding waveforms and receiving the target echoes simultaneously. Thus it suffer from the near-far problem and the detection range is limited mainly by antenna crosstalk and receiver dynamic range. The paper presents the limitation caused by this problem and the alternative solutions leading to almost unlimited distance of noise radar working in pulsed waveform mode.
The paper proposes an implementation of the high-frequency technique for simulating the electromagnetic-wave reflectivity of electrically large objects at sub-THz frequencies, with the aim of producing ISAR images of space objects and building a database for automatic classification using semantic classifiers. A simulator has been built based on the graphical computer software Blender, which can provide precise and efficient ray-tracing calculation for geometrical optics (GO) implementation, and physical optics and impedance boundary conditions (PO - IBC) for wave effects computation. The simulator is capable to produce images which account for the frequency dependent effects, such sensitivity to surface roughness and materials' electrical properties. The results indicate that the tool has a great potential for fast generation of high-resolution images at extremely high frequencies where the computational load is only spent on the electromagnetic part and the ISAR processing, while the large amount of CPU required by the geometric manipulations of models for ray-tracing calculations are made with GPU accelerators.
Coordination of radars can be performed in various ways. To be more resilient radar networks can be coordinated in a decentralized way. In this paper, we introduce a highly resilient algorithm for radar coordination based on decentralized and collaborative bundle auctions. We first formalize our problem as a constrained optimization problem and apply a market-based algorithm to provide an approximate solution. Our approach allows to track simultaneously multiple targets, and to use up to two radars tracking the same target to improve accuracy. We show that our approach performs sensibly as well as a centralized approach relying on a MIP solver, and depending on the situations, may outperform it or be outperformed.
Recently, Joint Communication and Sensing (JCAS) gained a lot of attention due to its relevance for the upcoming 6G standard. Sensing can supplement communication. For example, when the sensing module senses the presence of an obstacle and its motion direction, it then passes the sensing result to a network orchestrator module. In this way, the communication module can take reasonable measures to maintain communication quality. This paper describes 2D and 3D instantaneous velocity estimation methodologies for a target tracked using coordinated OFDM radar nodes. With our approach, the 2D and 3D velocities of objects can be found accurately. A precise estimation of the instantaneous velocity improves the prediction of the target's trajectory, which is crucial for many potential applications, such as automated guided vehicles and robotic arms in factories.
Bi- and multistatic radar systems have larger sensitivity when compared to the monostatic configuration however, at the expense of increased system complexity. Because the different parts that compose the resulting radar system are not co-located, a synchronisation method for remote nodes is mandatory. For moving platforms, the synchronisation link requires a clock steering algorithm which tolerates ever-changing baselines as well as a method to measure the relative positioning throughout the time between a transmitter-receiver pair. It is intended to research, develop and build a versatile networked radar system which allow moving nodes to take advantage of the fact that its different nodes may adapt their relative geometry to the convenience of the situation. This paper reviews frequency and time transfer methods, and introduce the problematic associated with a dynamic two-way time transfer.
