Advanced signal processing techniques for interference removal in Satellite Navigation Systems

This thesis investigates the use of innovative interference detection and mitigation techniques for GNSS based applications. The main purpose of this thesis is the development of advanced signal processing techniques outperforming current interference mitigation algorithms already implemented in off-the-shelf GNSS receivers. State-of-the-art interference countermeasures already investigated in literature, which process the signal at the ADC output, provide interference components suppression in the time domain or in the frequency domain, thus leading to a significant signal degradation in harmful interference scenarios where the GNSS signals spectra at the receiver antenna is completely jammed by external intentional or unintentional RFI sources. The proposed advanced interference countermeasures overcome such a limit, since they are based on particular signal processing techniques which manipulate the received samples at the ADC output, providing a representation in new domains where interference component can be better detected and separated from the rest of the signal, minimizing the useful signal distortion even in presence of multiple interference sources. At the cost of an increased computational complexity, such techniques can be optimized for increasing the sensitivity and the robustness of GNSS receiver merged in harmful environments. The work of this thesis addresses the design of such techniques by means of theoretical analyses, their performance assessment by means of simulation and their validation by means of synthetic and real GNSS data. Furthermore performance comparison with more traditional interference countermeasures is also presented considering a variety of harmful interference scenarios. In addition to the investigation of such new interference countermeasures, part of the thesis deals with the limit of current interference suppression technique, such as the pulse blanking, and its impact on the data demodulation performance. A very general investigation of the pulse blanking impact on the data demodulation performance for un-coded BPSK DSSS is provided. Then, the analysis focuses on the assessment of the navigation data demodulation performance for the current SBAS, then providing a proposal for system improvements, in terms of robustness and data rate increase, in future SBAS generation. Among the different interference scenarios considered, the thesis focuses on the potential interference environment expected in aviation context, since the Galileo E5 and GPS L5 bands, where the future GNSS based aviation services will be broadcast, are shared with other ARNS broadcasting strong pulsed interfering signals, which may seriously threat the on-board GNSS receiver operations . For such scenarios, simulation and analytic models are discussed and used as benchmark cases for assessing the mitigation techniques, in terms of SNR gain and data demodulation capability. The presence of interference (mitigated or not) causes a loss in the carrier to noise density ratio CN0 value for the received signal. For this reason, in order to reliably deal with such signals, the GNSS receiver must be able to feature high-sensitivity algorithms at the acquisition and tracking stages. For this reason the last part of the thesis investigates HS acquisition schemes for very weak GNSS signal detection. In particular, the purpose of this part of the work is to present a theoretical methodology for the design of an acquisition scheme capable of detecting signal down to 5 dB-Hz. The analysis carried out assuming the presence of assistance information which allows the receiver employing long coherent integration time (order of seconds). The particular scenario of the GNSS space environment is taken into consideration and the analysis is also focused on the definition of the requirements on the accuracy for potential Doppler aiding sources at the receiver level. The theoretical analysis is also supported by fully software simulation.