As 5G mm-wave communication progresses toward higher frequencies to meet market demands, companies are being pushed to adopt cutting-edge technologies. Operating at these higher frequencies naturally leads to shorter wavelengths, which increases circuit density. This higher density, in turn, complicates heat management. A supplementary solution involving heatsink antennas has been proposed to address this issue. Recent research from the MS3 group introduces a cost-effective, dual-functional pin-fed shorted patch antenna designed to assist with heat dissipation. Although the integration of a heatsink significantly improves thermal management, it does so at the cost of electromagnetic (EM) performance, particularly affecting radiation pattern symmetry, cross-polarization levels, and bandwidth. This work builds upon the latest research by using the aforementioned design as a benchmark, exploring further improvements and addressing the challenges introduced by the heatsink. This work presents, for the first time, the use of differential feeding to address the performance degradation caused by the heatsink. Differential feeding has been shown to enhance radiation pattern symmetry, increase gain, and improve bandwidth in certain applications. Additionally, aperture coupling is employed to further enhance the impedance bandwidth of the shorted patch. The results at the element level indicate a 90% improvement in impedance bandwidth due to aperture coupling; however, this comes with a significant degradation in the radiation pattern. By integrating aperture-coupled differential feeding, this work demonstrates symmetric radiation patterns, an improved bandwidth compared to the benchmark, and reduced cross-polarization levels. At the array level, differential feeding leads to a higher degree of coupling than pin-fed patch arrays for spacing values less than or equal to 0.6λ. Furthermore, differential feeding improves gain, reduces cross-polarization levels, and enhances radiation pattern symmetry compared to the benchmark.

A novel new waveform is compared with two other existing waveforms to create a radar network. In this radar network one primary radar transmits a waveform for sensing and communication and one or multiple radars receive these waveforms. The constraint for the radars is that all radars can build up a radar picture and situation awareness. What also means that the performance of the primary radar does not degrade. This new radar type is called cooperative hitchhiker with communications. In this radar network the main task is sensing, therefore additional communication signals are used to increase the performance of the sensing task. The advantages of parallel sensing and communications are reducing interference, dual use of the scarce electromagnetic (EM) spectrum in a congested EM environment and it is possible in a bistatic configuration to detect a bistatic scattering object.
The three waveforms which are compared are phase coded frequency modulated continuous wave (PC FMCW) linear frequency modulated – minimum shift keying (LFM-MSK) and a new waveform time delay between radar bursts (TDBRB). TDBRB is a simple but effective communication method on top of the sensing waveform and can be used with staggered waveforms. Where the first two waveforms make use of binary phase shift keying (BPSK), TDBRB uses a time modulation with the inserted time delay between two successive radar bursts. To distinguish the two radar bursts the first burst has a linear frequency modulated (LFM) upchirp and the successive burst, after the inserted time delay, has an LFM downchirp. PC FMCW and LFM-MSK data are compared with simulations and an experiment of the TDBRB waveform.
The conclusions are that PC FMCW has the highest data rate, followed by LFM-MSK, and TDBRB has the lowest data rate for line-of-sight connections. The first two make use of the information in one pulse, or FMCW chirp. This results in a lower signal-to-noise ratio (SNR) and is therefore only suited for line-of-sight connections. With a matched filter and coherent integration of the TDBRB signals, this waveform is most suited for non-line-of-sight connections and communication over an object-of-interest at the cost of a lower bit rate. TDBRB has a similar performance for the different Swerling cases as the detection probability of a regular radar. The degradation in performance with the TDBRB waveform is only a fraction of the burst time because the time delay is added to an original radar burst. These results make TDBRB most suited for non-line-of-sight communication and for communications with a low signal-to-noise ratio.

Nowadays, accurate vehicle classification plays a critical role in Advanced Driver Assistant Systems (ADASs), autonomous driving systems, and traffic monitoring systems. The benefits of utilizing additional polarimetric information in road target classification have been revealed in the literature. This thesis investigates the polarimetric characteristics of multi-class vehicles and explores new features contributing to vehicle classification, using a labeled street-way database extracted from the PARSAX S-band polarimetric Frequency Modulated Continuous Wave (FMCW) radar. The vehicle classes involved in this thesis are sedan, sedan with extended luggage bin, mini-van, small truck, and large truck.

Three calibration algorithms are proposed and validated for calibrating the labeled street-way database to ensure feature quality. The first channel calibration algorithm removes the channel-specific amplification factors and biases due to the non-ideal and non-identical electronic devices in the four polarimetric channels of the PARSAX radar. The second phase compensation algorithm compensates the phase difference between the H- and V-polarized channels, which is caused by the time shift between the transmitted H- and V-polarized signals. The last antenna pattern compensation algorithm resolves the power degradation in the measurements due to the PARSAX radar beam width limitation.

Based on the calibrated labeled street-way database, multiple polarimetric features are extracted from the Polarization Scattering Matrices (PSMs), coherency and covariance matrices using the eigenvalues/eigenvectors decomposition methods. These matrices represent either the central bodies or the whole bodies of the vehicles. In each case, the eigenvalues and eigenvectors are analyzed to indicate the vehicles' reflection amplitudes/power and polarization basis. Furthermore, these features are evaluated, and most of the amplitudes/power-based features show great classification capabilities. However, all vehicles have a similar polarization basis, which does not have a contribution to vehicle classification. In addition, the target length and eigenvalues of the covariance matrix of detection cells are also extracted as potential features. The feature evaluation results show that the target length and the first eigenvalue of the covariance matrix of detection cells have classification capabilities, while the second eigenvalue does not contribute to vehicle classification.

Automatic Dependent Surveillance Broadcast (ADS-B) allows aircraft to broadcast their own position, speed, altitude, and other information to ground stations and other nearby aircraft. This information is then used by air traffic control for situational awareness, and collision avoidance. ADS-B spoofing is possible due to the lack of authentication and encryption in the ADS-B protocol. This can result in incorrect decision-making and potential safety hazards. Validation of the location of the ADS-B message is required for Luchtverkeersleiding Nederland (LVNL) such that it can maintain separation minima between civil aircrafts, whilst using ADS-B operationally.
Analysis of possible approaches for ADS-B validation has resulted a multilateration (MLAT) based approach. Time of Arrival (TOA) measurements of the ADS-B messages are used to validate the location. For validation, at least two Ground Stations (GS) are required instead of the four GSs required for a MLAT track, allowing for ADS-B validation in a larger area than it is currently used for in a tracking application. If an ADS-B message is considered validated, its content can be used by ATC. Therefore MLAT based validation results in an increased surveillance coverage. Validation is achieved in two steps, first a tracker is used to compute the state of the target using the TOA measurements, secondly this state is compared to the ADS-B location using a likelihood ratio test.
Tracking is done using a Sequential Importance Resampling (SIR) Particle Filter (PF). Classical PF issues as the degeneracy problem and sample impoverishment problem are mitigated by using a novel sampling method that samples directly from the measurement at the initialization of the SIR filter. Without this novel method a traditional SIR filter (where the proposal density is uniformly distributed) requires roughly a million particles to converge on the location of the target. Below this amount of particles the traditional SIR filter fails. The proposed SIR filter can converge on the location of the target using only 1000 particles.
To provide LVNL options and insights, three different likelihood ratio tests are proposed, namely the Minimum Bayes Risk, The Neyman-Pearson and the Minimax Hypothesis test.
Performance of the algorithm is investigated using a case study where data from LVNL’s Surveillance Data North Sea (SDNS) MLAT system is used. Results have found that each test is capable of correct ADS-B validation. The limiting factor in the validation algorithm is the quality of the state estimate. At lower altitudes (<FL20) state estimation can fail and therefore also the hypothesis test. Above this altitude, spoofed targets can be detected if the distance between the spoofing transmitter and the location inside the spoofed message is roughly 1000 to 2000 meters depending on the hypothesis test used. Horizontally, this falls within LVNL’s separation minima, vertically, this falls outside the separation minima.

The main purpose of a radar is to detect, recognize, and track objects of interest. When it is known that only a single target is present, the matched filter is proven to be optimal detector. However, in practice, a radar scene often consists of multiple targets. For example, in air surveillance and monitoring applications, multiple aircrafts might be in the airspace. When multiple targets are to be detected the matched filter is not guaranteed to give the best results. This can happen when a strong reflector masks the signals reflected from weak reflectors, thereby resulting in missed detections. Furthermore, when the sensor resolution is low, targets that are spaced closely together may only result in a single target actually being detected. This research explores how the Relevance Vector Machine (RVM) framework may be used to achieve a better multi-target detector than the commonly used basic matched filter approach. RVM was selected to resolve the multi-target detection problem as it estimates the target locations iteratively. In this research it was shown how the RVM framework can be used to model the fluctuation of swerling I/II targets. Additionally, the RVM algorithm was modified to incorporate a notion of statistical thresholding. Simulations show that using RVM the false alarm rate can be reduced and target locations can be more accurately recovered compared to other existing methods in case of multiple swerling I/II fluctuating targets. Furthermore, the proposed approach is shown to have a much lower convergence time compared to a similar expectation-maximization based method, namely Enhanced Sparse Bayesian Learning.

Advancements in radar technology such as phased array antennas, digital beamforming, and adaptable waveform generation have led to the flexibility in controlling radar resources such as scan time, beamwidth and bandwidth during a radar mission. This new flexibility has led to a new research topic, radar resource management. Radar resource management involves the allocation of radar resources in order to achieve the highest performance in a radar mission. This thesis focuses on a specific scenario where a radar is tasked with deciding multiple object presence decisions located at multiple scan directions. The resource considered is the scan time allocated to each scan direction. To optimally allocate the scan time over the scan directions, a cost function is formulated, where the expected performance of an individual decision of an object being present or absent is formulated as the expected Bayes risk. The expected performance of all individual decisions in the same scan direction, are summed to obtain an expected task performance. The global performance at the system level is formulated using two approaches. The Sum approach formulates a cost function at the system level as the sum of the expected cost of each task. The Max approach formulates a cost function that minimizes the maximum of all expected task costs. Simulations have been performed to demonstrate the flexibility of the Sum and Max approach to adapt the scan time allocation based on different scenarios, including multiple object presence decisions per scan direction, sequential measurements, and a birth-death process regarding the presence of an object over time. Simulations demonstrate that using the Sum and Max approaches for the allocation of scan time results in an improved performance compared to the uniform distribution of the scan time resource over all scan directions.

For the development of automatic People Counting systems, radar is increasingly becoming a popular technology because of the increasingly stringent privacy requirements for people demographic information and the requirement to operate in a challenging environment. Because of the complexity of multi-target movement and the diversity of application scenarios, Radar-based People Counting methods are required to have sufficient robustness. However, based on the review of the current literature, the grouping phenomenon (i.e., multiple individuals moving close together as a single group) was not often considered in the experimental scenarios.

This thesis aims to study one of the most complex motions of individuals, grouping, and address the People Counting problem more in general, including the cases of grouping of multiple individuals. After studying the characteristics of Group People (defined as a group of people sharing neighboring, adjacent locations and moving together), with the help of multiple-input-multiple-output (MIMO) frequency-modulated continuous wave (FMCW) Radar, the combination of the Range-Azimuth map and spectrogram/cadence velocity diagram (CVD) is proposed to solve Group People Counting.

Algorithm-wise, there are two categories of existing Radar-based People Counting methods, namely, tracking for counting methods and feature-based counting methods. It was found that these two categories of methods have complementary strengths. Thus, the proposed method combines the tracking for counting approach and feature-based counting approach into a new processing pipeline to estimate the number of people in each group in the scene while tracking each group. Based on it, the proposed method achieves "Beyond classification", which is the output the unlabeled classes not defined at the training stage. Moreover, compared with other state-of-the-art (SOTA) Radar-based People Counting methods, the proposed method outperforms them, and thus it is proved that the grouping problem can be solved in the Radar-based People Counting field.

The Fifth Generation (5G) of mobile networks exploit both sub-6 GHz and millimeter-wave (mmWave) spectrum. The sub-6 GHz spectrum comprises frequencies up to 6 GHz and provides large geographical coverage for radio signal in 5G. The mm-wave spectrum on the other hand, comprises higher frequencies ranging from 24 GHz -100 GHz and plays a major role in serving high data rates for the 5G technology. The evolution of 5G plays a pivotal role in the realization of challenging applications like Social XR conferences, which requires the network to deal with heavy traffic while maintaining low end-to-end latencies. The right configuration of the radio access network becomes crucial for such applications. The introduction of Massive Multiple Input Multiple Output (MIMO) technology in the radio network, concentrates the signal energy to the target user, which significantly improves the throughput and efficiency of the system. The quality and capacity of the radio channels also depend on the downlink Channel State Information (CSI). The CSI when obtained accurately at the Base Station (BS), plays a significant role in reaping the best benets out of Multiple Input Multiple Output (MIMO) technology. This thesis explores (or assesses) the different options and configurations of CSI feedback using an indoor Social Extended Reality (XR) conference application scenario. The performance analysis of the radio downlink while using the latest 5G New Radio (NR) Types I and II CSI which use a DFT-based codebook are detailed. The impact of the codebook-related configurable parameter of Rotation Factor (RF), the performance variations while using a `fixed-RF' for all the UEs compared to the more flexible `adaptive-RF', different beamforming technologies (Single-User MIMO (SU-MIMO) and Multi-User MIMO (MU-MIMO)), transmission ranks and co-scheduling parameter values are assessed using the key performance metric of Packet Loss Ratio (PLR). The frequency bands of 3.5 GHz (sub-6 GHz spectrum) and 26 GHz (mmWave spectrum) are chosen for the thesis and performance variations between the two bands are studied. The key insight from the thesis research is that the `adaptive-RF' case gives the optimal performance for the considered Social XR scenario when we set the right co-scheduling parameters (which balance the encountered interference and frequency of co-scheduling).

Nowadays, recent Integrated Circuit Technology demands a minimum metal fill inclusion across the entire die of each functional layer to have uniform planarity. It is challenging to numerically simulate the real structure using Electromagnetic (EM) solvers because the mesh size of an EM solver is inversely proportional to the number of vertices in the layout and the inclusion of these metal fills has a significant impact on these vertices and increases mesh size. It is important to understand the metal fills do have an impact on the RF performance of the component and the impact becomes stronger as the frequency increases due to the parasitics and hence we cannot neglect it. Therefore, a Design Flow Methodology is studied and analyzed to take into account the effect of metal fills without penalty in accuracy and simulation time.

The second part of the thesis focuses on two different approaches that can be used to extract the complex parameters providing the electromagnetic response of the layers, i.e. electrical permittivity and magnetic permeability. First, we describe the Quasistatic approach to extract the epsilon parameters from the capacitance and inductance. After, describing some of the limitations of this approach we focus on the Oblique Incidence Method as reported in the literature to extract the complex epsilon tensor parameters.

The third part of the thesis looks into the Design Flow Methodology that embeds the Oblique Incidence Method into the design flow for 3D EM solvers to take into account the effect of metal fills. The 2.5D EM solver takes into account a scalar number which is less accurate and hence an insight of how the solution from the Oblique Incidence Method can used to extract a scalar number for the effective parameters and therefore be implemented in 2.5D EM tools is discussed.

The fourth part of the thesis shows the validation of the Oblique Incidence Method for the 3D EM tool for two test cases which are wave propagating structure and lumped structure namely Grounded Coplanar Waveguide (CPWG) and Stacked Transformer. Validation of the Scalar Number Approaches for the 2.5D EM Tool was done for the CPWG structure.

Finally, a brief study on the Design Based Approach is conducted with different metal fill shapes and alignment by using the Oblique Incidence Method by making use of 3D EM tool. This approach provides a better understanding of the epsilon variations for different metal fill shapes and that for less epsilon variations, an accurate scalar number value can be extracted to enable 2.5D EM simulations with improved accuracy.

With modern multi-function radars becoming more flexible, handling the limited amount of resources of these radars becomes increasingly important. In this thesis the radar resource management (RRM) problem in a multi-target tracking scenario is considered. Partially observable Markov decision processes (POMDPs) are used to describe each tracking task. By comparing the future effect of radar actions using model predictive control (MPC), the POMDPs are solved in a non-myopic way. The model predictive control problem can be decoupled into sub-problems using Lagrangian Relaxation to reduce the computational complexity of the solution method. An algorithm based on golden section search is employed to find the Lagrange multiplier. An interacting multiple model filter is used to allow the method to be effective in RRM problems involving the tracking of targets performing a broad number of maneuvers.
The novel approach is compared to an existing solution method based on policy rollout and Monte Carlo sampling. Through simulations of dynamic multi-target tracking scenarios in which the cost and computational complexity of different approaches are compared, it was shown that the computational complexity is greatly reduced while the resulting resource allocation results remain similar.

In order to ensure safety and prevent collisions on road, automotive radars must be fault proof and have to be tested on reliability and performance, which requires proper diagnostic of well-functioning of the radar so that the car may participate to traffic. One way of the diagnostic of well-functioning of the automotive radar is by means of calibration in service stations. In contrast to offline calibration in service stations, one another way of testing the automotive radar on well-functioning would be by means of monitoring the state, or with other words the healthiness, of the radar in real-time. One possible solution to monitor the radar state is to use a massive set of calibration targets in road infrastructure and thereby the problem lies in optimally estimating the state based on the RCS information provided by the radar. As a result, in the first part of this thesis, based on the defined target selection criteria, selection of the most appropriate calibration target among possible candidates is discussed. Using massive set of targets is required to overcome the uncertainty in production and installation accuracy in one-target measurements. However, this method brings randomness which is caused by two error sources being the non-ideal shapes of the calibration targets originating from mass production errors and orientation errors either from installation or maintenance errors. Second part of the thesis investigates how this randomness affects the self-diagnostics performance of automotive radar. Thereby, the model of target orientation and RCS loss due to orientation errors, and, the model of target RCS and its RCS loss due to mass production errors are developed. For both error sources, the statistical characteristics of the loss factors are determined by means of corresponding probability distribution functions which are derived analytically. In case of orientation errors, analytical results are validated by Monte-Carlo simulations as well as Kullback-Leibler Divergence. In case of mass production errors, analytical results are validated by Monte-Carlo simulations only. Together with the results obtained in the first part, results of the statistical characteristics of the loss factor due to non-orthogonality help to find a balance between the size, quality and number of targets to be deployed in a certain range in a given road configuration. To finalize the project, the measurement model is determined according to which the approach for self-diagnostics is developed under certain assumptions and considering a set of measurements of the same realization of a single target only. By the developed approaches, the relation of statistical parameters on self-diagnostics performance is determined by means of three different estimation methods and the results are validated by Monte-Carlo simulations.

Nowadays, many practical radar applications require an automatic interpretation of the received data, including data processing algorithms and target classification. The exploitation of additional polarimetric information is a very promising concept to improve the performance of automotive target classification. In this thesis work, we aim to identify target features that can be used for the classification and definition of sub-classes of moving automotive vehicles driving on a highway. This analysis is based on a multi-dimensional polarimetric feature database, created from real observations from a fully polarimetric-Doppler S-band FMCW radar (PARSAX). The polarimetric information of the vehicles is extracted while tracking the targets in a multi-target environment in the range-Doppler domain. Therefore, a multi-target tracking algorithm, based on an OS-CFAR detector, polarimetric data fusion algorithm, and a classical Kalman filter, is used. In order to cope with Doppler ambiguity, a novel MHT-based approach has been introduced.

The feature extraction analysis shows that the polarimetric features of the observed targets provide well-defined reliable statistical relations between physically related features, but that blind classification based on our target feature database does not provide new insights that are useful for classification. Reliable clusters that are useful to describe the polarimetric signatures of the targets have not been found, except the polarimetric correlation coefficients, which, unfortunately, despite their physical clear sense, were not supported by the other analyzed features. Nevertheless, from similar feature analysis, it has been shown that the features originating from the incoherent polarimetric H/A/α-decomposition form compact and well-separated clusters corresponding to target scattering and clutter scattering. Therefore, it can be concluded that these features can be used to accurately distinguish moving vehicles from static clutter.

In this thesis, a design of a passive radar for use on a drone used in inhospitable areas where air traffic control is not available due to circumstances. The thesis focusses on the receiver chain and angle of arrival algorithm.

In this thesis signal processing techniques for the Ultra-wide-band impulse radar (UWB-IR) radar are dis-cussed. To context wherein these algorithms are used is in a car, with the aim of detecting a child left behind in the car. This thesis also touches on algorithms to obtain vital information of the child inside the car. To this end a mathematical model for chest movement causes by breathing and heartbeat is made and explained. From this mathematical description data is derived in the form of a range-time matrix in MATLAB. The functioning of the algorithms is then showcased using this data. Regarding the algorithms. This thesis discusses 3 potential filters that can be implemented in the signal processing scheme. These are the adaptive, average, and exponential filters. Their hardware implementation is spoken of as well as their implementation in MATLAB. The algorithm for detecting of humans are also covered with emphasis on the threshold used for this. Finally the last step in the signal processing scheme, the measuring of breathing- and heart-beat frequency ,is worked out in the form of four additional algorithms which include: Fast Fourier transform, Hilbert Huang transform, The centroid algorithm, Cumulative summation. This thesis then finally tests these four algorithms in form of a complete signal processing algorithm using real world experimental data. Then by considering all aspects of the scheme, from hardware implementation, efficiency, functionality and precision, the algorithm that works best for our goal of detecting a child in a car is chosen.

This thesis analyses different radar systems for sensing chest movement to detect human presence inside a car. The detection in this thesis is limited to sensing breathing rate, although heartbeat detection is considered a possibility. A comparison is made between CW-, FMCW-, pulsed- and UWB pulsed radar. UWB pulsed radar was chosen to be a suitable technology for this application. As a benchmark, the Xethru X4 SoC by Novelda is used. The parameters and the working of Xethru X4 are investigated. A suitable antenna for Xethru is designed and simulated. Different detection techniques are discussed. The feasibility of the Xethru X4 is tested by means of calculations in different settings. Based on these tests, the Xethru X4 is evaluated and concluded to be suitable for detecting human presence inside a car by means of vital sign detection.

Wings for Aid is a company which develops an airborne aid-delivery system in emergency-struck areas and wants to use the radar system of Selfly to make an autonomously flying system. The goal of this project is to optimise the current ground-based radar system of Selfly for this application. This is done through the design of a Bi-static radar system, which can be mounted on, and used for drones to detect and avoid other obstacles in the airspace.\\
This thesis describes the design process and results of a sub-system of the Bi-static Sense and Avoid System for Drones (BiSAD). This a Bi-static radar system, consisting of three sub-systems that together form the base of the radar system. There will be one transmitter drone and several smaller receiver drones which are designed. Also, an optimal waveform is created for this purpose.\\
The report gives a proof of concept and will focus on a proof of concept for the transmitter chain as well as highlight some total system design choices and results.

This thesis compares several technologies to detect human vital signs in cars, in particular respiration and heartbeat, to avoid babies from being left behind in the car. The thesis starts with imposing mandatory and trade-off requirements. The thesis continues with a global overview of the possible technologies based on a literature study. The most promising technologies for our implementation: radar, sonar, infrared and optical camera are chosen to investigate further. Based on this further investigation, the technologies are compared based on the mandatory and trade-off requirements.

ISAR imaging is a well known technique which uses range and Doppler information from a radar to generate an image of a moving target. Research performed on the combination of a large receive array with airborne ISAR is scarce. This thesis is aimed to reach the highest possible image quality for AMBER, a radar with a 24 element receive array which was developed by TNO. Optimizing the image quality was split up in two parts, motion compensation and clutter suppression. A literature study is performed to research the existing methods. A performance evaluation is performed on different motion compensation techniques based on a simulation which is set up to resemble the radar parameters of AMBER. Based on the results of this evaluation, keystone formatting combined with image contrast maximization is concluded to be the best fitting approach. For clutter suppression a similar approach is followed with real measured data. Apart from the existing techniques, MVDR, DPCA and ODPCA, a new technique is proposed which filtered a target from clutter based on motion. It is referenced to as the IDPCA. The techniques show better image quality than regular beam forming. To enhance the image quality more, a method is searched to combine different clutter suppression techniques. This leads to a unique method which is proposed in this thesis. The new technique exploits the circular phase variance of pixels in the range Doppler image between different sub arrays. With this approach, strong clutter can be filtered in the range Doppler domain based on its angle of arrival. The newly proposed technique is applied as a mask, as it does not contain amplitude information of the target. Combined with the ODPCA method, the final image is generated. The combination of techniques shows clear improvements from the other discussed approaches.

Acquiring an accurate estimate of position is a challenging problem in coherent radar processing techniques such as Synthetic Aperture Radar (SAR). Even more so, for light and agile platforms such as multi-copters. Due to their unpredictable flight path, their motion must be accurately measured during data acquisition, and compensated for during processing. To obtain a focused SAR image, each pulse location must be accurately known to ensure that each pulse is added coherently to the imaging grid. Traditionally, this is achieved with an Inertial Navigation System (INS). While an INS can provide reasonable performance, its weight and size are often a constraint for agile platforms, limiting the available options and attainable accuracy.

In this study, we perform an analysis on the applicability of an omnidirectional radar array for explicitly estimating the motion of a multi-copter platform, and improving on the positioning accuracy achieved by the on board INS. Building on existing 1D SAR motion compensation techniques, we develop new methods for estimating the 3D motion of the radar platform by estimating its height and velocity. In addition, we also present a novel 3D autofocus technique termed multi-beam autofocus. This technique allows for the correction of 3D trajectory errors from pulse to pulse by exploiting the beamforming capabilities of the array, and focusing multiple regions as the image is created.

Using an Extended Kalman Filter (EKF), we obtain position estimates from the radar velocity measurements based on the last known INS position. We experimentally verify that using our velocity estimation method alone, the positioning performance is already improved compared to that of a state-of-the-art INS, allowing for INS-free imaging using arbitrary flight paths. This enables imaging in GNSS-denied environments, and has the potential to further reduce the weight of the platform. We also show that fusing the estimates obtained from our method with the existing INS output yields an additional performance increase in terms of SAR image focus, improving the resolvability and detectability of weak targets. The presented results open further avenues of research, not only in agile SAR imaging but also in autonomous GNSS-denied navigation.