Andrei Anghel

Dr. Andrei Anghel


Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 January 2014 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
AAAndrei Anghel
Andrei Anghel
Gabriel VASILEGabriel VASILE
Gabriel VASILE
RCRemus Cacoveanu
Remus Cacoveanu
CICornel Ioana
Cornel Ioana
SCSilviu Ciochina
Silviu Ciochina
The frequency-modulated continuous-wave (FMCW) radar is an alternative to the pulse radar when the distance to the target is short. Typical FMCW radar implementations have a homodyne architecture transceiver which limits the performances for short-range applications: The beat frequency can be relatively small and placed in the frequency range affected by the specific homodyne issues (dc offset, self-mixing, and 1/f noise). In addition, one classical problem of an FMCW radar is that the voltage-controlled oscillator adds a certain degree of nonlinearity which can cause a dramatic resolution degradation for wideband sweeps. This paper proposes a short-range X-band FMCW radar platform which solves these two problems by using a heterodyne transceiver and a wideband nonlinearity correction algorithm based on high-order ambiguity functions and time resampling. The platform's displacement measurement capability was tested on range profiles and synthetic aperture radar images acquired for various targets. The displacements were computed from the interferometric phase, and the measurement errors were situated below 0.1 mm for metal bar targets placed at a few meters from the radar.

3 February 2023 in IEEE Geoscience and Remote Sensing Magazine
Verified
Journal Article
Closed access
MDMihai Datcu
Mihai Datcu
Zhongling HuangZhongling Huang
Zhongling Huang
AAAndrei Anghel
Andrei Anghel
JZJuanping Zhao
Juanping Zhao
RCRemus Cacoveanu
Remus Cacoveanu
The recognition or understanding of the scenes observed with a synthetic aperture radar (SAR) system requires a broader range of cues beyond the spatial context. These encompass but are not limited to the imaging geometry, imaging mode, properties of the Fourier spectrum of the images, or behavior of the polarimetric signatures. In this article, we propose a change of paradigm for explainability in data science for the case of SAR data to ground explainable artificial intelligence (XAI) for SAR. It aims to use explainable data transformations based on well-established models to generate inputs for AI methods, to provide knowledgeable feedback for the training process, and to learn or improve high-complexity unknown or unformalized models from the data.

16 September 2019 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Closed access
AAAndrei Anghel
Andrei Anghel
RCRemus Cacoveanu
Remus Cacoveanu
AMAdrian-Septimiu Moldovan
Adrian-Septimiu Moldovan
BRBjorn Rommen
Bjorn Rommen
MDMihai Datcu
Mihai Datcu
Monitoring large built structures, landslides, volcanoes, or glaciers implies relative height measurements and line-of-sight (LOS) displacement measurements, which are typically performed with monostatic spaceborne synthetic aperture radar (SAR) sensors. In this case, for each satellite that illuminates a given location, we can usually exploit only the information available from one ascending/descending orbit, and if the satellite uses multiple subswaths, the information is obtained only from one subswath per orbit. A bistatic configuration with spaceborne transmitter and fixed ground-based (GB) receiver opens the possibility to exploit the information available from more than one orbit and one subswath per orbit. Additionally, such a configuration offers new perspectives for target tracking and characterization (e.g., multiple LOSs and bistatic scattering signatures). This article presents an opportunistic C-band bistatic SAR differential interferometry architecture that uses a multichannel GB stationary receiver and a separate transmitter, which can be either the Sentinel-1A/B satellite, used opportunistically, or a specially designed GB sliding transmitter. The combination of the operation modes based on spaceborne and GB transmitter allows monitoring a small critical area of interest with many acquisitions triggered by the GB transmitter and surveying of the whole surrounding area with a small number of acquisitions corresponding to the satellites passes. The hardware platforms are presented along with the bistatic interferometric processing flow, and the potential of the proposed architecture is assessed in the context of monitoring large built structures.

14 April 2023 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Open access
Reza M. AsiyabiReza M. Asiyabi
Reza M. Asiyabi
MDMihai Datcu
Mihai Datcu
AAAndrei Anghel
Andrei Anghel
HNHolger Nies
Holger Nies
Deep learning models have achieved remarkable success in many different fields and attracted many interests. Several researchers attempted to apply deep learning models to Synthetic Aperture Radar (SAR) data processing, but it did not have the same breakthrough as the other fields, including optical remote sensing. SAR data are in complex domain by nature and processing them with Real-Valued (RV) networks neglects the phase component which conveys important and distinctive information. A Complex-Valued (CV) end-to-end deep network is developed in this study for the reconstruction and classification of CV-SAR data. Azimuth subaperture decomposition is utilized to incorporate physics-aware attributes of the CV-SAR into the deep model. Moreover, the correlation coefficient amplitude (Coherence) of the CV-SAR images depends on the SAR system characteristics and physical properties of the target. This coherency should be considered and preserved in the processing chain of the CV-SAR data. The coherency preservation of the CV deep networks for CV-SAR images, which is mostly neglected in the literature, is evaluated in this study. Furthermore, a large-scale CV-SAR annotated dataset for the evaluation of the CV deep networks is lacking. A semantically annotated CV-SAR dataset from Sentinel-1 Single Look Complex StripMap mode data (S1SLC_CVDL dataset) is developed and introduced in this study. The experimental analysis demonstrated the better performance of the developed CV deep network for CV-SAR data classification and reconstruction in comparison to the equivalent RV model and more complicated RV architectures, as well as its coherency preservation and physics-aware capability.

15 November 2021 in IEEE Access
Verified
Journal Article
Open access
NRNicolae-Catalin Ristea
Nicolae-Catalin Ristea
AAAndrei Anghel
Andrei Anghel
RIRadu Tudor Ionescu
Radu Tudor Ionescu
Radar sensors are gradually becoming a wide-spread equipment for road vehicles, playing a crucial role in autonomous driving and road safety. The broad adoption of radar sensors increases the chance of interference among sensors from different vehicles, generating corrupted range profiles and range-Doppler maps. In order to extract distance and velocity of multiple targets from range-Doppler maps, the interference affecting each range profile needs to be mitigated. In this paper, we propose a fully convolutional neural network for automotive radar interference mitigation. In order to train our network in a real-world scenario, we introduce a new data set of realistic automotive radar signals with multiple targets and multiple interferers. To our knowledge, we are the first to apply weight pruning in the automotive radar domain, obtaining superior results compared to the widely-used dropout. While most previous works successfully estimated the magnitude of automotive radar signals, we propose a deep learning model that can accurately estimate the phase. For instance, our novel approach reduces the phase estimation error with respect to the commonly-adopted zeroing technique by half, from 12.55 degrees to 6.58 degrees. Considering the lack of databases for automotive radar interference mitigation, we release as open source our large-scale data set that closely replicates the real-world automotive scenario for multiple interference cases, allowing others to objectively compare their future work in this domain. Our data set is available for download at: http://github.com/ristea/arim-v2 .

24 March 2025 in IEEE Geoscience and Remote Sensing Magazine
Verified
Journal Article
Closed access
Cheng HuCheng Hu
Cheng Hu
Yuanhao LiYuanhao Li
Yuanhao Li
ZCZhiyang Chen
Zhiyang Chen
FLFeifeng Liu
Feifeng Liu
QZQingjun Zhang
Qingjun Zhang
MDMihai Datcu
Mihai Datcu
As a crucial sensor for wide-area Earth observation, spaceborne synthetic aperture radar (SAR) plays a pivotal role in large-scale terrain mapping, ocean observation, disaster monitoring, and so forth. Driven by the increasing demands for diverse applications, enhanced performance, and the continuous advancement of satellite and radar technologies, the distributed configuration has emerged as a key developmental trend for spaceborne SAR. This review comprehensively summarizes the systems and typical applications of distributed spaceborne SAR. The system configurations encompass homogenous distributed SAR, formed by multiple identical or similar platforms, and heterogeneous distributed SAR, characterized by significant differences between the transmitting and receiving platforms. Typical applications of distributed SAR include intelligent target recognition, terrain mapping, deformation retrieval, atmosphere measurement, and ocean observation, among others. Finally, the review offers a prospective outlook on the future development of distributed spaceborne SAR.