Hello,

Please visit the following link for detailed tutorial:

https://geospatialwarehouse.wordpress.com/2017/04/29/synthetic-aperture-radar-sar-remote-sensing-basics-and-applications/

Hope that will help.

Shahid

Synthetic Aperture Radar data are complex-valued usually.Their magnitude corresponds to "how strong the radar backscatter was" and their phase corresponds to "where the data come from", i.e. the position on a map. In order to assess the intensity of the radar backscatter it must be compared to objects on ground (so-called calibration targets, corner reflectors for example) whose backscatter (i.e. the strength of the signal) is known in theory. It corresonds to putting a 1 kg weight onto a weighing machine in order to check whether the weighing machine is showing "1 kg" as a result of this measurement. So, radiometric calibration is the adjustment of gray-values in SAR images such that their back scatter (gray-value) correponds to a defined value according to the theory. And the background of doing this is comparability. Radar images cannot be compared otherwise if they come from different SAR sensors.

The amount of radar echo from a target is characterized by its Radar Cross Section (RCS) typically in units of square-meters or dBsm.  For distributed targets (e.g. grass, dirt, etc.) this value is usually normalized per unit area, that is, square-meters per square-meters, or dBsm/sm...  RCS per unit area is often called clutter reflectivity.  Relating pixel values to either RCS or clutter reflectivity is called radiometric calibration.

why do we need to relate pixel values to either RCS or clutter reflectivity? 

Is it because of the introduced factors  such as the antenna gain, the system loss and the effective aperture of antenna, etc that introduce radiometric bias in the SAR image  

Radiometric calibration is not required if only (gray-scaled or colored) images are required (for example to see through clouds). But SAR images are not only images. They are measurements. They measure the intensity of radar reflections from ground. The more reflection we have, the brighter the image. The intensity of radar reflections from ground is a measure of moisture and roughness. Many researchers try to seperate these two physical entities in order get soil moisture maps or roughness maps. A SAR image is very dark if we have low backscatter and it is very bright it we have severe backscatter. Suppose you have two SAR image taken at Antarctica, one is taken in July and one is taken November, one is very dark and one is very bright. What can you say about reflectivity from ground? Nothing, unless you know that both images are radiometrically calibrated. It tells you that both images can be compared. Now you can say: Okay, the brighter image is telling me that something has changed on ground and this "change" can be quantified (measured) in numbers, given by the difference between both images.  

For many applications, a target RCS, or a calibrated area reflectivity, is a discriminant for characterizing or identifying the target.  For example, a slow-moving +10 dBsm reflector is probably not a person, but rather a vehicle.

now I understood the calibration is that you use calibration devices in order to make sure that their back-scatter values is the same at the receiver, and by doing that you guarantee that you have the right values of backscatter at the end.

how can we get complex output after performing radiometric calibration.

For radiometric calibration of Sentinel-1 Level-1 products please see the attached document prepared by Nuno Miranda, P.J. Meadows. They clearly explained the process.

I need to know how I can calibrate a SAR product (single look or detected) without the magic of available calibration tools (provided in SNAP or OrfeoToolbox). I need to know how exactly one can retrieve the respective information from the metadata and then use the LUTs provided in above-mentioned S1 manual to do the calibration. I would appreciate any comment in this regard.

SAR Radiometric calibration can be classified into system calibration and data calibration. Radar designers and calibration researchers focus on the former one to reduce SAR radar system radiometric errors by internal (fast time varied) and external calibration (constant, space varied, slow time varied) devices, such as transmitted power, antenna pattern， transmitted and receive gains, image processing gains error. Unfortunately, there are residual system errors, and other non-system errors, such as local atmosphere induced radiometric errors, topography induced errors, are still included in the calibrated images.

For SAR remote sensing data users, system calibration have been performed.

SAR "data calibration" can be treated as re-calibration or some kind of data validation.

Generally, data calibration is not necessary (depend on the required radiometric accuracy for their geophysical or other parameters inversion) or too complicated ( calibration devices and professional knowledge) for data users. Data users can order SAR image data for a specific area in advance. Before the SAR satellite fly over this area, data users need to place a set of reflectors with known RCS and well pointed to the SAR satellite. So users can get SAR images with point targets with known RCS. With these point targets, users can re-calibrate or validate the radiometric calibration accuracy of the specific SAR images ordered from the data suppliers.