New detection manifolds for radar signal processing

The spatial distribution of blood flow following spinal cord injury (SCI) represents a promising biomarker for improving patient outcomes. This work follows up on a previously reported algorithm, FlowMorph, that enables measurement of blood flow through individual microvessels in the rat spinal cord. Building upon FlowMorph in this study, spinal cord blood flow was quantified pre- and post-injury, and single-vessel changes were related to distance from the injury. Two-dimensional probability densities were calculated using kernel density estimation in order to determine the extent of injury throughout the cord and measure the borders of vascular injury. Using these borders, vessel density and average flow was calculated at the site of injury as well as in the surrounding tissue. Significantly fewer vessels had measurable flow at the site of injury after trauma, and flow in the remaining vessels was highly variable. Further experiments with different injury severities will delineate the effect of injury severity on blood flow. Additionally, experiments with the addition of contrast-enhanced measurements will validate this technique.

We propose a two stage beamforming technique for localisation of moving target in a passive radar environment. In order to perform localization and tracking in Cartesian plane, in addition to detection in the range Doppler domain, it is necessary to have angle of arrival of target return, which is obtained using beamforming. The beamforming technique can also enhance signal to disturbance ratio. However, performing constant false alarm rate detection (CFAR) in this case requires storing beamformer output statistics for every range-Doppler-azimuth bin which significantly burdens the system and hinders the real time processing and tracking of moving targets. Using experimental measurements, we demonstrate real time detection and tracking of an airborne target using a two stage beamforming approach, where a smaller set of beams is used in the first stage for CFAR detection and a larger set of beams is used in the second stage for angle of arrival estimation with higher accuracy.

This paper summarizes the application of software packages such as Mathcad and MATLAB in system simulations of communication and radar systems. With the help of the software packages, little time was spent on computer programming; nearly all the energy was concentrated on the analyses of the fundamental theory behind the simulations.

The use of space-based bistatic radar has many attractive features (particularly related to covert observation and survivability) over that of conventional monostatic systems. However (if detection performance is to approach the monostatic case) these advantages come at the price of inherently more complex hardware and processing algorithms. In particular, when considering a bistatic system that utilizes a space-based transmitter/ground-based receiver architecture, obtaining detection ranges of practical value is problematic without having unrealistic requirements for space-based transmitter ERP or ground-based receiver antenna size. A predetection signal processing scheme is postulated which adapts to the nature of the bistatic signal, thus allowing for longer coherent integration times and correspondingly larger signal processing gain.

Traditionally, moving target indication (MTI) modes tailored to wide-area surveillance (WAS) have employed waveforms with bandwidths matched to the range extent of targets of interest. Matching range resolution to expected vehicle lengths serves to minimize interfering clutter returns while preserving available target power, thereby maximizing target probability-of-detection. Recent advances in waveform generator technology and high-speed digital signal processors for matched filtering have encouraged radar designers to consider wideband waveforms for WAS MTI needs. While providing information for identification and imaging, wideband waveforms complicate the detection process, as an extended target will be over-resolved; that is, returns from a vehicle will be distributed over multiple, fine-resolution range bins. Algorithms for combining these signal contributions are required for detection and angle estimation. Given such algorithms, one wonders whether over-resolving targets leads to detection and angle estimation performance losses as compared to narrowband operation. In this paper we evaluate methods for collecting energy from an over-resolved target for detection and angle estimation processing. We bound the performance of these techniques on typical ground targets and show that the probability-of-detection and angle accuracy is improved through the use of wideband waveforms at high signal-to-noise levels.

The main problem of special aerial reconnaissance is the contradiction between the dimensions of the survey area and the dimensions of elementary objects (targets). Since the viewing area is very large, the size of the target is small, and the instantaneous field of view of the on-board equipment is limited, the time for obtaining data is quite significant and the amount of information to be processed is very large. Also, due to technical limitations on the time spent in the air and the limitations of the reconnaissance equipment, one UAV is not able to cover the entire area in one flight. The severity of this problem can be partially removed by using a group of UAVs. but both for one person and for a group, the task of planning the route of movement arises. A simple comb survey of the reconnaissance zone is far from optimal, and in practice route planning is used by points with the involvement of a priori information about the most likely locations of targets. However, this method is rather inconvenient for the automatic route planning of the UAV itself, since the point on the map usually does not correlate with the flight height, the instantaneous field of view of the on-board surveillance equipment, the scale and detail of the picture. In order to increase the optimization possibilities of automatic flight planning, it is suggested to use the target distribution density function (TDF). The density function of the target distribution is a two-dimensional mathematical model that describes the conditional relative probability of finding targets at different points in space. It is created and set on the basis of a priori data about the terrain, previous observations or intellectual assessments, reflecting the distribution of possible targets in a certain area or for the entire reconnaissance area. TDF allows to model the space not as homogeneous, but as an area with different degrees of importance or probabilities of finding targets. It should be noted that obtaining mathematically justified or calculated by the theory of probability or mathematical statistics, two-dimensional discrete probability densities for randomly placed targets is actually impossible. Therefore, here the TDF is considered as an area with different degrees of review importance, which is reflected by conditional relative probabilities obtained from various sources, including expert evaluation.

Abstract. The direct benefit of a waterlogging control project is reflected by the reduction or avoidance of waterlogging loss. Before and after the construction of a waterlogging control project, the disaster-inducing environment in the waterlogging-prone zone is generally different. In addition, the category, quantity and spatial distribution of the disaster-bearing bodies are also changed more or less. Therefore, under the changing environment, the direct benefit of a waterlogging control project should be the reduction of waterlogging losses compared to conditions with no control project. Moreover, the waterlogging losses with or without the project should be the mathematical expectations of the waterlogging losses when rainstorms of all frequencies meet various water levels in the drainage-accepting zone. So an estimation model of the direct benefit of waterlogging control is proposed. Firstly, on the basis of a Copula function, the joint distribution of the rainstorms and the water levels are established, so as to obtain their joint probability density function. Secondly, according to the two-dimensional joint probability density distribution, the dimensional domain of integration is determined, which is then divided into small domains so as to calculate the probability for each of the small domains and the difference between the average waterlogging loss with and without a waterlogging control project, called the regional benefit of waterlogging control project, under the condition that rainstorms in the waterlogging-prone zone meet the water level in the drainage-accepting zone. Finally, it calculates the weighted mean of the project benefit of all small domains, with probability as the weight, and gets the benefit of the waterlogging control project. Taking the estimation of benefit of a waterlogging control project in Yangshan County, Guangdong Province, as an example, the paper briefly explains the procedures in waterlogging control project benefit estimation. The results show that the waterlogging control benefit estimation model constructed is applicable to the changing conditions that occur in both the disaster-inducing environment of the waterlogging-prone zone and disaster-bearing bodies, considering all conditions when rainstorms of all frequencies meet different water levels in the drainage-accepting zone. Thus, the estimation method of waterlogging control benefit can reflect the actual situation more objectively, and offer a scientific basis for rational decision-making for waterlogging control projects.

Condensation of R-134a in upward and downward annular flow was experimentally studied in a 5-mm ID transparent test section using a LED-based image processing technique. In addition to the flow orientation, the frequency of the film structures was quantified in terms of the following parameters: total mass flow rate, condensation heat transfer rates (outlet vapor quality) and test section inclination with respect to the horizontal. The image processing technique combined Fourier analysis and the Kernel density estimation technique to calculate the two-dimensional (amplitude–frequency) probability density function of the film flow structure as a function of the independent parameters. Although the tube inclination and the condensation rate showed some influence on the frequency of the interfacial waves, the most significant effect was due to the flow orientation, which resulted in abrupt frequency changes apparently linked to flow regime transitions dictated by shear and gravity forces.

In drone detection and short-range automotive radar, the detection system should have the ability to monitor a large area and to detect the distances and directions of multiple targets. Suffered from electronic bottleneck, it’s difficult to achieve a high-performance detection via conventional radar system. Microwave photonics has the advantages of low loss, large bandwidth and so on, and greatly improves the performance of radar system. To realize simultaneous distance and direction detection, microwave photonic phased array radar steers the beam for inertialess and squint-free scanning to detect targets from different directions. However, limited by the narrow angle scanning range, its field of view is inadequate. Photonics-assisted AOA (angle-of-arrival) detection method can detect the direction of target in a large-angle range. Nevertheless, it can hardly obtain the distance information of targets. It is a key issue to perform wide-angle simultaneous detection of distances and directions for multiple targets. We introduce a microwave photonic radar for distance and direction measurement of multiple targets. At the transmitter, a linear frequency modulation signal is generated by photonic frequency doubling and transmitted for detection. At the receiver, echo signals are received by a uniform linear array with three antennas. The intermediate antenna supplies overlapped reference spectra by polarization-multiplexing modulation. The de-chirped spectra possess specific symmetry, thus de-chirped frequencies of each antenna are clearly distinguished and distances and directions of multiple targets can be obtained. Two-target proof-of-concept experiments have verified that the average measurement errors of distance and direction are 2.4 cm and 0.6°, respectively.

Radio direction finding (DF) is a technique for estimating radiowave angle of arrival (AoA) and is based on prior knowledge of the antenna responses versus AoA.Traditional DF systems exploit rotating antennas or sampled antenna arrays for which the antenna response patterns are very carefully controlled to provide support for closed‐form DF solutions or for the development of steered beams. Sources contributing to DF errors include local scattering, instrumentation, and propagation. Some more recent DF systems exploit array calibration and digital beamforming techniques to provide reliable AoA estimates while accomodating local scattering. Both of these DF system categories are designed for operation against single component plane waves. More sophisticated DF systems now exploit sampled arrays and DF processing to decompose the complex wavefields created by multiple simultaneous signal components.

This on going seminal work presents a new methodology for the design and implementation of high performance artificial raw data generation algorithms for remote-sensed imaging applications. Special attention is given in this work to synthetic aperture radar (SAR) system modeling and simulation imaging applications in the geosciences. Of particular importance are processes such as soil moisture content, backscattering from crops, nearshore ocean surface currents, and subsurface imaging in hyperarid regions. Our computing approach is based on the successful use of cross-ambiguity functions, in a Weyl-Heisenberg computational framework, as surface point target response functions for nonlinearly modulated, time-frequency structured, artificially created transmitted signals for our SAR system raw data modeling and simulation efforts. The functions are correlated with prescribed target reflectivity density functions to produce the desired results.

In direction-of-arrival (DOA) estimation using millimeter waves or ultrasonic, the physical size of the sensors can require inter-element spacing of greater than half wavelength. The accompanied benefits of such spacing are larger array aperture and reduced mutual coupling, but the serious problem of DOA ambiguity arises. This paper proposes a method to suppress grating lobes by employing a MIMO radar array configuration. The resultant virtual array has sufficient half wavelength spacings to significantly suppress grating lobes while the transmit and receive array elements have greater than half wavelength spacing. The grating lobe suppression performance is evaluated by simulation of MIMO-OFDM radar system that also has frequency diversity properties against DOA ambiguity. The simulation results show that the grating lobe suppression performance increases with signal-to-noise ratio (SNR). The effectiveness of the proposed array configuration and frequency diversity for resolving DOA estimation is also demonstrated.

In recent years, Through the Wall Imaging (TWI) has become a topic of research interest due to its promising applications in home-land security situations.In this paper, the basic concept of Stepped-Frequency Continuous Wave (SFCW) for ranging the targets behind the wall is described.Imaging of targets is implemented using the back-projection algorithm.This paper also brings out the received signal model for detection of stationary human being behind the wall using SFCW approach.The detection is based on estimation of heart beat and respiratory motion.Simulation results of human detection through Doppler processing and the 2-D imaging of targets are brought out for illustration.MATLAB software has been used for the development of signal processing modules.

The baseband system serves as the core component responsible for signal processing in harmonic radar systems. However, designing a baseband system that meets requirements poses several challenges. In this study, we addressed the design and simulation of a baseband system for a 3.6GHz harmonic radar. A fundamental signal at a frequency of 3.3MHz was selected, with 90° phase-shift binary phase shift keying (BPSK) modulation employed as the modulation scheme for the baseband system. MATLAB was utilized for program development to simulate the reception of nonlinear target-reflected harmonic signals. Results of signal processing for the second and third harmonics were provided. Furthermore, an FPGA-based hardware circuit for the harmonic radar was designed, followed by hardware testing and target data collection. The practical feasibility of the design approach was validated. Comparative analysis between measured and simulated data revealed the system's proficiency in identifying harmonic signals during signal processing.

Research on Radar System Simulation and Non-Gaussian Mathematical Model under Virtual Reality Technology