New Methods and Processing Algorithms for Passive Radar Imaging

• Applicant: Dr.-Ing. Holger Nies, since 10/2022
• Subject Area: Electronic Semiconductors, Components and Circuits, Integrated Systems, Sensor Technology, Theoretical Electrical Engineering
• Term: from 2019 to 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 410851921

• Final Report Year: 2023

Final Report Abstract

The concept of passive radar offers the possibility of obtaining information about the environment, even with low­cost components, without actively transmitting a signal. The objective of this project was to demonstrate an operational measurement system for passive bistatic SAR and ISAR techni­ ques using geostationary satellites and DVB­T stations. Given the challenging conditions, such as the low bandwidth of the satellite transponders used (about 28 MHz) or the low receiving power due to the range of about 38,000 km to the illuminator, this seemed to be quite an ambitious project. A further complication is the complex motion profile of geostationary satellites compared to typical SAR platforms. The motion of the transmitter in azimuth and elevation complicates the model de­ scription for SAR processing in the frequency domain. In addition, the varying relative velocity of the satellite results in non­equidistant sampling, which affects the quality of the SAR images. Therefore, a BPA was used for SAR processing in this project. Although this is more computationally intensive compared to frequency domain processors, it is the only option due to the complex trajectory of the satellite. In our investigations of the DVB­S or DVB­S2 signal, which was originally developed for telecom­ munications services, they showed similar characteristics to typical radar signals. For example, the signal’s ambiguity function does not exhibit any strongly pronounced side lobes or artifacts, which is an important criterion for radar signals. Another feature of the DVB­S signal is that it has an almost uniform energy distribution over the bandwidth. In this project, the feasibility of passive radar with geostationary satellites was successfully demons­ trated with a low­cost measurement system for both SAR and ISAR applications. The passing of different ships with different system parameters was observed and analyzed. The pulse compressi­ on analysis also showed that the Wiener filter can reduce the masking effect in the close range better than the matched filter, provided that a suitable weighting parameter is chosen. However, due to the limited channel bandwidth of only 30 MHz, no significant impact on the ISAR image can be detected, since the resolution in the range is of the order of the ship’s width. Further­ more, it was shown that focusing in azimuth can be easily realized using the estimated parameters of a spectrogram. In principle, the Doppler response can be corrected by a chirp with an azimuth sweep rate k_az.The determined ship dimensions, such as ship length or velocity, roughly agree with the actual values. However, the backscatter spectrum is location dependent, so a specific azimuth filter must be determined for each ship as a function of range and speed to correct the Doppler response. Similarly, the feasibility of passive radar imaging using the signal from DVB­T transmitters in combination with a single­channel, low­cost receiver system has been demonstrated. Starting from some basic information about the OFDM signals used, the method of synchronization and channel estimation is described. As a main aspect, the reconstruction of the reference signal is realized. This allowed passive radar imaging with only one receiving antenna. The receiver system used was experiment­ ally tested using a ground­based moving platform. Due to the exact (noise­free) reconstruction of the transmitted signal, a better SNR could be achieved than would be possible with the direct signal. The signal processing was realized in MATLAB and published as open source software.

Publications

• Low-Cost Passive Radar Using OFDM Broadcast. 2019 IEEE Radar Conference (RadarConf), 1-6. IEEE.
  Behner, Florian; Reuter, Simon; Nies, Holger & Loffeld, Otmar
  
• “Passives Bistatisches Bildgebendes Radar und Zielverfolgung mittels geostationärer Satelliten”. Masterarbeit. University of Siegen, Juli 2023.
  Timur Mulic
  
• Ground-Based Passive Radar Imaging Using Geostationary Satellites—Perspectives and Constraints. IEEE Geoscience and Remote Sensing Letters, 22, 1-5.
  Nies, Holger; Behner, Florian; Reuter, Simon; Mulic, Timur & Ihrke, Ivo