Real-time Precise Point Positioning (PPP) has been evolved as a cost-effective technique for highly precise maritime positioning. For a long period, maritime PPP technology has mainly relied on the Global Positioning System (GPS). With the revitalisation of GLONASS and the emerging BeiDou navigation satellite system (BDS), it is now feasible to investigate real-time navigation performance of multi-constellation maritime PPP with GPS, BDS and GLONASS. In this contribution, we focus on maritime PPP performance using real world maritime kinematic data and real-time satellite correction products. The results show that BDS has lower position accuracy and slower convergence time than GPS. The BDS and GPS combination has the best performance among the dual-constellation configurations. Meanwhile, the integration of BDS, GLONASS and GPS significantly improves the position accuracy and the convergence time. Some outliers in the single constellation configuration can be mitigated when multi-constellation observations are utilised.

In vector tracking loop (VTL), the relativity among received signals is exploited to deeply integrate the entire information within signal processing channels. However, the tracking error in one channel may corrupt other channels and lead to an increasing degradation in the tracking performance. An adaptive GPS vector tracking loop with the detection and isolation of contaminated channels is proposed to suppress the propagation of tracking error and to make the vector-based receiver less vulnerable in poor signal quality environments resulting from signal attenuation, interference and jamming. The vectorial transfer function models and the noise bandwidths of both the vector frequency lock loop and the vector delay lock loop are established to accurately calculate bandwidths and optimally design parameters for VTL. The autonomous fault detection algorithm based on uniformly most powerful test for VTL is designed to detect and isolate contaminated channels (with large tracking error due to the poor signal quality) by monitoring the bandwidth statistics of all tracking loops. The results of trials in harsh situations show that the proposed adaptive VTL is superior to baseline VTL, allowing the accuracy, the availability and the reliability of the vector-based receiver to be improved. In conclusion, the proposed adaptive VTL with the detection and isolation of contaminated channels is a powerful method for applications in GPS-challenged environments.

In order to enhance the tracking performance of global positioning system (GPS) receivers for weak signal applications under high-dynamic conditions, a high-sensitivity and high-dynamic carrier-tracking loop is designed. The high-dynamic performance is achieved by aiding from a strapdown inertial navigation system (SINS). In weak signal conditions, a dynamic-division fast Fourier transform (FFT)-based tracking algorithm is proposed to improve the sensitivity of GPS receivers. To achieve the best performance, the tracking loop is designed to run either in the conventional SINS-aided phase lock loop mode (time domain) or in the frequency-domain-tracking mode according to the carrier-to-noise spectral density ratio detected in real time. In the frequency-domain-tracking mode, the proposed dynamic-division FFT algorithm is utilized to estimate and correct the error of the SINS aiding. Furthermore, the optimal values of the dynamic-division step and the FFT size are selected to maximize the signal-to-noise ratio gain. Simulation results demonstrate that the designed loop can significantly improve the tracking sensitivity and robustness for weak GPS signals without compromising the dynamic performance.

Modeling and mitigating carrier-phase multipath errors continue to be a significant challenge for high-accuracy positioning using global navigation satellite systems. The multipath error is dependent on the operational environment and therefore cannot be mitigated by differencing techniques. The effect of multipath is accentuated when observables based on linear combinations of measurements from two or more frequencies are formulated. We develop a new carrier-phase multipath error observable that isolates the inter-frequency carrier-phase multipath error for linear combinations of observables, such as wide-lane (WL). Real-time kinematic positioning results from varying baseline lengths show that a significant reduction in between 14 and 47 % in the time to initial ambiguity resolution are achieved by correcting the WL observable using the new carrier-phase multipath error observable.