Research Rocket Flight Trajectory Estimation System

"On each of GRS units there is an application created with LabVIEW software. This application configures both USRP-2922 transceivers, records received data to an SSD, synchronizes data, and partially processes data in real time."

- Damian Duraj, Gdańsk University of Technology

The Challenge:

Estimating the flight trajectory of a suborbital research rocket without using typical high-power radar methods.

The Solution:

Using pairs of USRP-2922 transceivers to create a measurement system that can receive measurement signals transmitted from the rocket, store measured data on hard drives, and partially process signals in real time.


Damian Duraj - Gdańsk University of Technology
Marek Płotka - Gdańsk University of Technology
Karol Abratkiewicz - Gdańsk University of Technology
Mateusz Rzymowski - Gdańsk University of Technology
Krzysztof Nyka - Delft University of Technology
Łukasz Kulas - Gdańsk University of Technology


Introduction to WiComm Center of Excellence

The Minister of Science established the WiComm Center of Excellence in 2004 as a result of a contest that selected the 100 best research teams. WiComm is the only Center of Excellence in Poland and focuses its scientific and research work on wireless communication systems, radio-location systems, high-frequency communications, and information technology. The main research program focuses on devices in the newest wireless systems such as WiFi, WiMAX, Bluetooth, ZigBee, UWB, RFID, wireless sensors, wireless mobile telecommunication 3G and 4G (UMTS, HSDPA, EDGE, LTE), collision avoidance radar, GPS, and Galileo.


For one of our research projects, we worked on creating a wireless communication system that can estimate a rocket’s flight. Scientists have successfully used common radar technologies for large objects, such as airplanes, helicopters, or space shuttles, for decades. For smaller objects, like suborbital research rockets, legal regulations limit the power and bandwidth of the signal transmitted in non-licensed bands. This, in addition to small size and a poor reflective surface of the object, affect conventional radar detection with small range and high measurement uncertainty.


Most existing systems are based on collecting data from onboard sensors, writing them on some built-in memory, and reconstructing data after retrieving the rocket. When there is no possibility of retrieving the rocket or it was destroyed during the flight, we lose the collected data about the flight path. Sometimes data about the rocket flight trajectory could help diagnose causes of malfunction or destruction. We could also use this data during the process of finding a rocket vessel after landing, but sometimes that data cannot be sent using typical communication systems due to bandwidth limitations and the problem of communicating with fast-moving objects.


Application and System Overview

We created a system that includes one onboard rocket terminal (BRT), three to four ground reference stations (GRS), and one central unit. The BRT contains different modules, such as a computing unit, a high-frequency signal generation module, a clock synchronization module, a WSN module, an inertial measurement unit, and a ground-rocket communication link module (Figure 1). Each GRS consists of two USRP-2922 transceivers, a computer, and a Radioline station. One of the USRP-2922 is equipped with a GPS disciplined oscillator, which allows synchronization of the clock and the PPS input using a GPS signal. Both of the USRP-2922 devices connect using a multiple input, multiple output (MIMO) cable (Figure 2). The central unit controls all elements in the system and features a computer, a Radioline station, and a ground-rocket communication link module.


On each of GRS units there is an application created with LabVIEW software. This application configures both USRP-2922 transceivers, records received data to an SSD, synchronizes data, and partially processes data in real time. Due to the large amount of data, they are computed after the flight. We created an application to control all of the GRS and receive part of the processed data in real time using Radiolines and TCP/IP protocol for the central unit. We used NI-VISA drivers to communicate with a ground-rocket communication link module.



We chose the USRP (Universal Software Radio Peripheral) because of its competitive specification and ease of programming and combining data from multiple devices. Using the USRP-2922, we can receive frequency-modulated continuous-wave (FMCW) or pulse signals transmitted from the BRT with the frequency of four periods. In most tests, the FMCW signals were transmitted due to the possibility of measuring a greater amount of parameters simultaneously. The FMCW signal consists of two parts with constant frequency and two rising sawtooth FMCW parts. The period of the FMCW signal is 1.2 µs. Transmitted signal bandwidth is 3 MHz and power does not cross 2 W effective isotropic radiated power, which means we can work in free ISM frequency band (Figure 3). Using the two USRP transceivers with the proposed signal allows measurement of distance, angle, and radial velocity between a GRS and the BRT. Creating dedicated hardware would require implementing a highly adaptable reconfigurable RF front end, which would consume a lot of time and tests. Also it would be problematic to synchronize data and times between separated devices and create a fast storage for a large amount of data. We can easily solve all of those problems with proper use of LabVIEW and drivers for USRP.


Experimental Testbeds

We conducted several tests using test rockets in different locations. Test locations included the old runway in Czaplinek and fields near the village of Mrzezino. We connected pairs of NI USRP devices and a computer and enclosed them in one case to facilitate the deployment process. We placed the GRS a great distance between them and the launching positions in all of the tests.


Measurement Results

We recorded large amounts of data during each test. In the first tests, we controlled each GRS individually. In subsequent tests, we controlled the GRS from the central unit after adding Radiolines to the system. We created plots showing the estimated trajectory from different methods based on the recorded data (Figure 4 and Figure 5). Distance data from Czaplinek shows a strong resemblance between the real positions of launching and landing of the rocket and the measured data. From the tests near Mrzezino, each method gave better results in different flight phases. Distance measurement was the most accurate, but in descending phases there were problems with low amplitude of signal. The GPS positions from accelerating phases were too rarely sampled. Data from the IMU and angle measurement shows incorrect behaviour probably due to problems with calibration and antenna configuration.


Using the USRP-2922 transceivers and available libraries in LabVIEW, we created a system that can analyse trajectory in real time and record data for offline processing. In the postprocessing case, LabVIEW is used to analyse data from different modules and merge them.


In the future, we plan to adjust data merging algorithms. Additionally, we can verify if the USRP RIO boosts recording parameters and if it can process all of the data in real time using the open FPGA feature.


Author Information:

Damian Duraj
Gdańsk University of Technology
Gabriela Narutowicza
Gdańsk 11/12 80-233

Figure 1.
Figure 2.
Figure 3.
Figure 4.
Figure 5.