Enhanced receiver architectures for processing multi GNSS signals in a single chain : based on partial differential equations mathematical model

Al-Aboodi, Maher (2016) Enhanced receiver architectures for processing multi GNSS signals in a single chain : based on partial differential equations mathematical model. Doctoral thesis, University of Buckingham.

Maher ALAboodi PhD ThesisFV.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (7MB) | Preview


The focus of our research is on designing a new architecture (RF front-end and digital) for processing multi GNSS signals in a single receiver chain. The motivation is to save in overhead cost (size, processing time and power consumption) of implementing multiple signal receivers side-by-side on-board Smartphones. This thesis documents the new multi-signal receiver architecture that we have designed. Based on this architecture, we have achieved/published eight novel contributions. Six of these implementations focus on multi GNSS signal receivers, and the last two are for multiplexing Bluetooth and GPS received signals in a single processing chain. We believe our work in terms of the new innovative and novel techniques achieved is a major contribution to the commercial world especially that of Smartphones. Savings in both silicon size and processing time will be highly beneficial to reduction of costs but more importantly for conserving the energy of the battery. We are proud that we have made this significant contribution to both industry and the scientific research and development arena. The first part of the work focus on the Two GNSS signal detection front-end approaches that were designed to explore the availability of the L1 band of GPS, Galileo and GLONASS at an early stage. This is so that the receiver devotes appropriate resources to acquire them. The first approach was based on folding the carrier frequency of all the three GNSS signals with their harmonics to the First Nyquist Zone (FNZ), as depicted by the BandPass Sampling Receiver technique (BPSR). Consequently, there is a unique power distribution of these folded signals based on the actual present signals that can be detected to alert the digital processing parts to acquire it. Volterra Series model is used to estimate the existing power in the FNZ by extracting the kernels of these folded GNSS signals, if available. The second approach filters out the right-side lobe of the GLONASS signal and the left-side lobe of the Galileo signal, prior to the folding process in our BPSR implementation. This filtering is important to enable none overlapped folding of these two signals with the GPS signal in the FNZ. The simulation results show that adopting these two approaches can save much valuable acquisition processing time. Our Orthogonal BandPass Sampling Receiver and Orthogonal Complex BandPass Sampling Receiver are two methods designed to capture any two wireless signals simultaneously and use a single channel in the digital domain to process them, including tracking and decoding, concurrently. The novelty of the two receivers is centred on the Orthogonal Integrated Function (OIF) that continuously harmonies the two received signals to form a single orthogonal signal allowing the “tracking and decoding” to be carried out by a single digital channel. These receivers employ a Hilbert Transform for shifting one of the input signals by 90-degrees. Then, the BPSR technique is used to fold back the two received signals to the same reference frequency in the FNZ. Results show that these designed methods also reduce the sampling frequency to a rate proportional to the maximum bandwidth, instead of the summation of bandwidths, of the input signals. Two combined GPS L1CA and L2C signal acquisition channels are designed based on applying the idea of the OIF to enhance the power consumption and the implementation complexity in the existing combination methods and also to enhance the acquisition sensitivity. This is achieved by removing the Doppler frequency of the two signals; our methods add the in-phase component of the L2C signal together with the in-phase component of the L1CA signal, which is then shifted by 90-degree before adding it to the remaining components of these two signals, resulting in an orthogonal form of the combined signals. This orthogonal signal is then fed to our developed version of the parallel-code-phase-search engine. Our simulation results illustrate that the acquisition sensitivity of these signals is improved successfully by 5.0 dB, which is necessary for acquiring weak signals in harsh environments. The last part of this work focuses on the tracking stage when specifically multiplexing Bluetooth and L1CA GPS signals in a single channel based on using the concept of the OIF, where the tracking channel can be shared between the two signals without losing the lock or degrading its performance. Two approaches are designed for integrating the two signals based on the mathematical analysis of the main function of the tracking channel, which the Phase-Locked Loop (PLL). A mathematical model of a set of differential equations has been developed to evaluate the PLL when it used to track and demodulated two signals simultaneously. The simulation results proved that the implementation of our approaches has reduced by almost half the size and processing time.

Item Type: Thesis (Doctoral)
Uncontrolled Keywords: GNSS signals; smartphones
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: School of Computing
Depositing User: Users 4 not found.
Date Deposited: 30 Jun 2016 14:19
Last Modified: 12 Dec 2019 14:53
URI: http://bear.buckingham.ac.uk/id/eprint/136

Actions (login required)

View Item View Item