3D-GBInSAR

Structural displacement monitoring using ground-based synthetic aperture radar (Implementation of 3D displacement vector)

This repository demonstrates how to retrieve 3D displacement vector by interferometric processing of time-series observations of ground-based SAR (GBSASR) system.

Usage

First, run "threeD_dsplcmnt_Part1_rawdata.m" to generate the simulated data. Then, run "threeD_dsplcmnt_Part2_processing_Sim.m" to process the data and estimate the 3D displacement vectors through the time! Moreover, if you want to see the algorithm's performance on a simulated beautifl bridge, run "threeD_dsplcmnt_BridgeSimulation.m"!

Results [Point scatterers]

Here are the results you can obtain by running the codes.

In this example, two point scatterers are simulated: One is fixed and another one is moving!

Note that in order to resolve the displacement vector in a 3D space first we need to acquire data with both horizontal and vertical baselines!

In order to easier demonstration of 3D objects I simply convert them into point clouds, while ignoring the unstable scatterers (belonging to background).

Finally! Here is the estimated displacements in 3D space for the fix and moving scatterers!

Results [Bridge Simulation]

Here, we conducted a numerical simulation of a bridge as a moving object. The backscattering signal was modeled by dividing the bridge's surface into small plate reflectors. Subsequently, we estimated the power of the received signal by calculating the normal vector of each surface and determining its angle in relation to the radar antenna. Below illustrates the process of simulating a surface. Additionally, we randomly placed several corner reflectors at various locations on the bridge for strong reflections.

Here, you can see the resulting image. The reason for displaying the iso-range areas is that strong reflections from points at the same range to the radar could potentially impact the accuracy of displacement estimation results. Therefore, it may be optimal to deploy a corner reflector in each of these iso-range areas to ensure reliable estimations.

The next two figures show the simulated displacement behaviours on the bridge's surface.

Now let's see some results and error analysis:

Paper

Link to the paper:

Elsevier

ResearchGate

Citation

If you find this work useful, please cite us in your work:

@article{3D-GBSAR,
    title = {Structural displacement monitoring using ground-based synthetic aperture radar},
    volume = {116},
    issn = {1872826X},
    doi = {10.1016/j.jag.2022.103144},
    pages = {103144},
    journaltitle = {International Journal of Applied Earth Observation and Geoinformation},
    author = {Hosseiny, Benyamin and Amini, Jalal and Aghababaei, Hossein},
    date = {2023-02},
    note = {Publisher: Elsevier},
    keywords = {{MIMO} radar, Vibration, {mmWave} radar, {SAR} interferometry, Structural health monitoring},
}