CBL - Campus del Baix Llobregat

Projecte llegit

Títol: Development of a Software-Defined Radio (SDR) Radio Telescope: SDR Front-End and Data Adquisition

Estudiants que han llegit aquest projecte:

Director/a: GELONCH BOSCH, ANTONI

Departament: TSC

Títol: Development of a Software-Defined Radio (SDR) Radio Telescope: SDR Front-End and Data Adquisition

Data inici oferta: 27-01-2026     Data finalització oferta: 27-09-2026


Estudis d'assignació del projecte:
    DG ENG AERO/SIS TEL
Tipus: Individual
 
Lloc de realització: EETAC
 
Segon director/a (UPC): GUTIÉRREZ CABELLO, JORDI
 
Paraules clau:
Radio Telescope, SDR, LNA
 
Descripció del contingut i pla d'activitats:
This thesis is devoted to the design, implementation, and characterization of a low-cost, high-flexibility radio telescope utilizing Software-Defined Radio (SDR) technology. Traditional radio astronomy instrumentation often relies on expensive, application-specific hardware that lacks the adaptability required for multi-purpose astronomical observations. By leveraging the versatility of SDR, this project aims to democratize access to radio astronomy for educational and small-scale research environments.
The system architecture consists of a parabolic reflector antenna and a high-sensitivity feedhorn optimized for observations in the range spanning from MHz to a few GHz. The signal chain incorporates a low-noise amplifier (LNA) and a bandpass filter to mitigate terrestrial interference before digitizing the signal via an RTL-SDR or similar high-bandwidth SDR platform.
The core of the project focuses on the development of a digital signal processing (DSP) pipeline. This software layer performs integration, and baseline subtraction. Calibration will be conducted using the "hot/cold" method, utilizing the ground and the empty sky as thermal references to determine the system's equivalent noise temperature.
A design requirement is that the entire system should be prepared to facilitate the incorporation of interferometry mechanisms.

Activities Plan:
1) SDR Front-end subsystems design
2) Subsystems development and integration
3) Performance system analysis
 
Overview (resum en anglès):
Radio astronomy allows astronomers studying celestial objects using radio waves such as the emission of the neutral hydrogen line, at 1420,4 MHz. Nevertheless, this activity is limited in most cases by the cost of conventional radio astronomy systems. This Bachelor's Thesis' objective is to approach this astronomy subfield to the general public, democratizing it. The project consists in the design, assembly and validation of both the Software-Defined Radio (SDR) Front-End and the Data Acquisition subsystems.

The hardware front-end development focuses on the realization of a two-stage Low Noise Amplifier (LNA) based on the QPL9065 integrated circuit. Using KiCad and Advanced Design System (ADS), a two-layer microstrip printed circuit board (PCB) was designed on a Rogers RO4003C high-frequency substrate and manually assembled utilizing surface-mount technology (SMT) laboratory infrastructure. Validation is carried out by RF characterization using a Vector Network Analyzer (VNA) and Spectrum Analyzer yielded.

The digital backend utilizes a LimeSDR Mini v2.4 transceiver integrated with a custom multithreaded DSP pipeline written in C using the SoapySDR hardware abstraction layer and FFTW3 libraries. To prevent real-time buffer overflows, the acquisition architecture separates signal capture, digital filtering, Fast Fourier Transform (FFT) calculation, non-coherent integration, and file streaming across dedicated threads linked via thread-safe circular buffers. The pipeline features a dynamically configured finite impulse response (FIR) anti-aliasing filter coupled with decimation, allowing for high spectral resolution and significant signal-to-noise ratio (SNR) enhancements.

Complementing the backend, a Human-Machine Interface (HMI) suite developed in Python provides real-time spectrum visualization, time-domain waterfall diagrams, and an automated event-logging system for astronomical detection. Empirical software validation confirms drastic variance reduction of additive white Gaussian noise through periodogram averaging, enabling the retrieval of weak stochastic signals buried well beneath the system noise floor. The complete system demonstrates a scalable, cost-effective alternative for educational and amateur radio observatories.

© CBLTIC Campus del Baix Llobregat - UPC