Next-generation Global Navigation Satellite Systems (GNSS) receivers should be capable of processing multi-frequency signals in order to provide better positioning accuracy and signal availability to end-user. Nevertheless, the realization of such receivers with conventional receiver architectures leads to power-hungry devices and hinders monolithic integration of the receiver. Multi–frequency receivers can be realized with lower power and cost by performing sub-Nyquist sampling (subsampling) at Radio Frequency (RF). In practice, the use of subsampling receivers has been limited due to their poor noise performance. This is particularly a concern in applications of Code Division Multiple Access (CDMA) such as GNSS since subsampling may saturate the Analog-to-Digital Converter (ADC) as the thermal noise floor is above the signals of interest. Continuous-Time (CT) Delta-Sigma (ΔΣ) modulation is an attractive candidate for subsampling Analog-to-Digital (A/D) conversion as it provides noise-shaping and inherent Anti-Alias (AA) filtering. However, the attenuation of the RF alias in the feedback path of the modulator and the reduction of the effective Quality (Q-factor) of the loop filter prevent conventional CT-ΔΣ modulators to be utilized in subsampling receivers. This thesis proposes a novel CT-ΔΣ modulator at both the system and the circuit level that is capable of compensating for the effects of subsampling. These are achieved by modifying the feedback path of the conventional modulator architecture to accommodate for the RF alias and by enhancing the Q-factor of the loop filter. The proposed CT-ΔΣ has significant improvements over previously published subsampling modulators as it provides jitter and alias suppression and excess loop delay compensation to improve the dynamic range, thus enabling the modulator to be utilized in subsampling receivers when a relatively low sampling rate is desired. Based on the novel CT-ΔΣ modulator, this thesis also proposes a subsampling receiver architecture for multi-constellation GNSS applications. Simulations results indicate that the proposed receiver architecture can successfully acquire and track the civilian radionavigation signals with a high performance.

This thesis addresses the power consumption problems resulting from the advent of multiple GNSS satellite systems which create the need for receivers supporting multi-frequency, multi-constellation GNSS systems. Such a multi-mode receiver requires a substantial amount of signal processing power which translates to increased hardware complexity and higher power dissipation which reduces the battery life of a mobile platform. During the course of the work undertaken, a power analysis tool was developed in order to be able to estimate the hardware utilisation as well as the power consumption of a digital system. By using the power estimation tool developed, it was established that most of the power was dissipated after the Analog to Digital Converter (ADC)by the filters associated with the decimation process. The power dissipation and the hardware complexity of the decimator can be reduced substantially by using a minimum-phase Infinite Impulse Response (IIR) filter. For Global Positioning System (GPS) civilian signals, the use of IIR filters does not deleteriously affect the positional accuracy. However, in the case where an IIR filter was deployed in a GLObalnaya NAvigatsionnaya Sputnikovaya Sistema (GLONASS) receiver, the pseudorange measurements of the receiver varied by up to 200 metres. The work undertaken proposes various methods that overcomes the pseudorange measurement variation and reports on the results that are on par with linear-phase Finite Impulse Response (FIR) filters. The work also proposes a modified tracking loop that is capable of tracking very low Doppler frequencies without decreasing the tracking performance.

The thesis provides a detailed overview of the work carried out by the author over the course of the research for the award of the degree of Doctor of Philosophy at the University of Westminster, and the performance results of the novel techniques introduced into the literature. The outcome of the work is collectively referred to as the Global Navigation System Scope (GNSScope) Toolbox, offering a complete, fully reconfigurable platform for the end-to-end modeling, simulation and analysis of satellite navigation signals and systems, covering the signal acquisition, tracking, and range processing operations that take place in a generic Global Navigation Satellite System (GNSS) receiver, accompanied by a Graphical User Interface (GUI) providing access to all the techniques available in the toolbox. Designed and implemented entirely in the MATLAB mathematical programming environment using Software Defined Radio (SDR) receiver techniques, the toolbox offers a novel new acquisition algorithm capable of handling all Phase-Shift Keying (PSK) type modulations used on all frequency bands in currently available satellite navigation signals, including all sub-classes of the Binary Offset Carrier (BOC) modulated signals. In order to be able to process all these signals identified by the acquisition search, a novel tracking algorithm was also designed and implemented into the toolbox to track and decode all acquired satellite signals, including those currently intended to be used in future navigation systems, such as the Galileo test signals transmitted by the GIOVE satellites orbiting the Earth. In addition to the developed receiver toolbox, three novel algorithms were also designed to handle weak signals, multipath, and multiple access interference in GNSScope. The Mirrored Channel Mitigation Technique, based on the successive and parallel interference cancellation techniques, reduces the hardware complexity of the interference mitigation process by utilizing the local code and carrier replicas generated in the tracking channels, resulting in a reduction in hardware resources proportional to the number of received strong signals. The Trigonometric Interference Cancellation Technique, used in cross-correlation interference mitigation, exploits the underlying mathematical expressions to simplify the interference removal process, resulting in reduced complexity and execution times by reducing the number of operations by 25% per tracking channel. The Split Chip Summation Technique, based on the binary valued signal modulation compression technique, enhances the amount of information captured from compressing the signal to reveal specific filtering effects on the positive and negative polarity chips of the spreading code. Simulation case studies generated entirely using the GNSScope toolbox will be used throughout the thesis to demonstrate the effectiveness of the novel techniques developed over the course of the research, and the results will be compared to those obtained from other techniques reported in the literature.

This thesis is concerned with the analysis of side-looking sonar images, and specif- ically with the identification of the types of seabed that are present in such images, and with the detection of man-made objects in such images. Side-looking sonar images are, broadly speaking, the result of the physical interaction between acous- tic waves and the bottom of the sea. Because of this interaction, the types of seabed appear as textured areas in side-looking sonar images. The texture descrip- tors commonly used in the field of sonar imagery fail at accurately identifying the types of seabed because the types of seabed, hence the textures, are extremely variable. In this thesis, we did not use the traditional texture descriptors to identify the types of seabed. We rather used scattering operators which recently appeared in the field of signal and image processing. We assessed how well the types of seabed are identified through two inference algorithms, one based on affine spaces, and the other based on the concept of similarity by composition. This thesis is also concerned with the detection of man-made objects in side-looking sonar im- ages. An object detector may be described as a method which, when applied to a certain number of sonar images, produces a set of detections. Some of these are true positives, and correspond to real objects. Others are false positives, and do not correspond to real objects. The present object detectors suffer from a high false positive rate in complex environments, that is to say, complex types of seabed. The hypothesis we will follow is that it is possible to reduce the number of false positives through a characterisation of the similarity between the detections and the seabed, the false positives being by nature part of the seabed. We will use scattering operators to represent the detections and the same two inference algorithms to quantify how similar the detections are to the seabed.

Mine Counter-Measure (MCM) missions are conducted to neutralise underwater explosives. Automatic Target Recognition (ATR) assists operators by increasing the speed and accuracy of data review. ATR embedded on vehicles enables adaptive missions which increase the speed of data acquisition. This thesis addresses three challenges; the speed of data processing, robustness of ATR to environmental conditions and the large quantities of data required to train an algorithm. The main contribution of this thesis is a novel ATR algorithm. The algorithm uses features derived from the projection of 3D boxes to produce a set of 2D templates. The template responses are independent of grazing angle, range and target orientation. Integer skewed integral images, are derived to accelerate the calculation of the template responses. The algorithm is compared to the Haar cascade algorithm. For a single model of sonar and cylindrical targets the algorithm reduces the Probability of False Alarm (PFA) by 80% at a Probability of Detection (PD) of 85%. The algorithm is trained on target data from another model of sonar. The PD is only 6% lower even though no representative target data was used for training. The second major contribution is an adaptive ATR algorithm that uses local sea-floor characteristics to address the problem of ATR robustness with respect to the local environment. A dual-tree wavelet decomposition of the sea-floor and an Markov Random Field (MRF) based graph-cut algorithm is used to segment the terrain. A Neural Network (NN) is then trained to filter ATR results based on the local sea-floor context. It is shown, for the Haar Cascade algorithm, that the PFA can be reduced by 70% at a PD of 85%. Speed of data processing is addressed using novel pre-processing techniques. The standard three class MRF, for sonar image segmentation, is formulated using graph-cuts. Consequently, a 1.2 million pixel image is segmented in 1.2 seconds. Additionally, local estimation of class models is introduced to remove range dependent segmentation quality. Finally, an A* graph search is developed to remove the surface return, a line of saturated pixels often detected as false alarms by ATR. The A* search identifies the surface return in 199 of 220 images tested with a runtime of 2.1 seconds. The algorithm is robust to the presence of ripples and rocks.

This thesis is concerned with the problem of automating the interpretation of data representing the underwater environment retrieved from sensors. This is an important task which potentially allows underwater robots to become completely autonomous, keeping humans out of harm’s way and reducing the operational time and cost of many underwater applications. Typical applications include unexploded ordnance clearance, ship/plane wreck hunting (e.g. Malaysia Airlines flight MH370), and oilfield inspection (e.g. Deepwater Horizon disaster). Two attributes of the processing are crucial if automated interpretation is to be successful. First, computational efficiency is required to allow real-time analysis to be performed on-board robots with limited resources. Second, detection accuracy comparable to human experts is required in order to replace them. Approaches in the open literature do not appear capable of achieving these requirements and this therefore has become the objective of this thesis. This thesis proposes a novel approach capable of recognizing targets in sonar data extremely rapidly with a low number of false alarms. The approach was originally developed for face detection in video, and it is applied to sonar data here for the first time. Aside from the application, the main contribution of this thesis, therefore, is in the way this approach is extended to reduce its training time and improve its detection accuracy. Results obtained on large sets of real sonar data on a variety of challenging terrains are presented to show the discriminative power of the proposed approach. In real field trials, the proposed approach was capable of processing sonar data real-time on-board underwater robots. In direct comparison with human experts, the proposed approach offers 40% reduction in the number of false alarms.

Global Navigation Satellite Systems (GNSS) are the main systems that provide positioning, navigation and timing at a global level. They are being used in numerous applications in different sectors including transport, military, oil & gas, agriculture as well as location based services. A significant number of these applications require centimetre-level positioning accuracy, a challenging feat due to the many error sources that affect GNSS measurements. These include errors at the satellite, propagation medium, and receiver levels. Most of these errors can be mitigated by modeling, or by exploiting their spatial and temporal correlation characteristics. However, multipath errors, which result from the combination of the direct signal with reflected signals in the vicinity of the receiver antenna, are difficult to model and therefore, difficult to mitigate. Furthermore, high accuracy positioning applications typically rely on linear combinations of measurements at different frequencies (e.g. L1 and L2 in the case of the Global Positioning System) to mitigate frequency-dependent errors such as ionospheric errors (i.e. ionosphere free combination) or otherwise facilitate position calculation (e.g. Wide Lane observable). The multipath errors associated with such combinations are significantly larger than those of individual signals. The dependency of the multipath error on the environment and its low level in single frequency measurements (i.e. up to quarter of wavelength) makes modelling and mitigating it very difficult. Current techniques attempt to mitigate multipath errors for measurements at each individual frequency, independently of the error at other frequencies, even when linear combinations of measurements are used. The literature review carried out in this thesis has drawn three main conclusions regarding carrier multipath mitigation. Firstly, existing carrier multipath mitigation techniques are inaccurate, impractical or not effective. Secondly most of the practical techniques attempt to mitigate the error by de-weighting the measurements which are most prone to the multipath error (i.e measurement at low elevation). Thirdly, existing weighting techniques are oversimplified and do not reflect the error level accurately. In this research and for the first time, carrier multipath errors have been studied directly at the linear combination level. This is by exploiting the dispersive nature of multipath errors in order to model and correct them. New carrier multipath mitigation techniques applicable to linear combinations of measurements have been developed in this thesis on the basis of a new error model and a new observable referred to as the IFM (Inter-Frequency carrier Multipath). The IFM is computed from carrier phase measurements at two different frequencies, and corresponds to the combined multipath errors of those signals. In addition to multipath mitigation, this observable has various other applications. The well-defined relationship between the IFM and carrier multipath errors is used in this thesis to develop multipath mitigation techniques based on two approaches: multipath correction and measurement weighting. The new mitigation techniques are applicable to linear combinations of observations such as Wide Lane (WL) and Ionosphere Free (IF) carrier phase measurements in double differenced mode. The new multipath mitigation techniques have been validated using real data and the results compared with those obtained using the elevation weighting technique. The results show that the new methods developed in this thesis improve the mean error of horizontal position by up to 33% when using the IF combination. The results also show improvements of up to 78% in the time it takes to resolve ambiguities when using the WL combination.

This thesis investigates the potential for estimating tropospheric delay from Global Navigation Satellite Systems (GNSS) stations on moving platforms experiencing a change in altitude. The ability to accurately estimate tropospheric delay in kinematic GNSS positioning has implications for improved height accuracy due to the mitigation of a major GNSS error source, and for the collection of atmospheric water vapour data for meteorology and climate studies. The potential for extending current kinematic GNSS positioning estimates of tropospheric delay from sea level based studies to airborne experiments, and the achievable height accuracy from a range of tropospheric mitigation strategies used in airborne GNSS positioning, are explored. An experiment was established at the Snowdon Mountain Railway (SMR), utilising the railway to collect a repeatable kinematic dataset, profiling 950 m of the lower atmosphere over a 50 day period. GNSS stations on stable platforms and meteorological sensors were installed at the extremities of the trajectory, allowing reference tropospheric delays and coordinates to be established. The retrieval of zenith wet delay (ZWD) from kinematic GNSS solutions using tropospheric estimation strategies is validated against an interpolated reference ZWD between GNSS stations on stable platforms, together with profiles from 100 m resolution runs of the UK Met Office Unified Model. Agreement between reference ZWD values and a combined GPS+GLONASS precise point positioning (PPP) solution is demonstrated with an accuracy of 11.6 mm (RMS), similar to a relative positioning solution and previous shipborne studies. The impact on the height accuracy from estimating tropospheric delay in kinematic GNSS positioning is examined by comparing absolute and relative GNSS positioning solutions to a reference trajectory generated from a relative GNSS positioning solution ii processed with reference to the GNSS stations on stable platforms situated at the extremities of the SMR. A height accuracy with a standard deviation of 72 mm was demonstrated for the GPS+GLONASS PPP solution, similar to a GPS-only relative solution, and providing an improvement over the GPS-only PPP solution.

No description available.

GPS has become an almost indispensable part of our infrastructure and modern life. Yet because its accuracy, reliability, and integrity depend on the number and geometric distribution of the visible satellites, it is not reliable enough for the safety of life, environmental or economically critical applications. Traditionally, this has been addressed by augmentation from dedicated support systems, or integration with other sensors. However, from an engineering perspective only expensive inertial systems or pseudolites offer the accuracy required. In the case of pseudolites, the equivalent of ground based satellites, geometry constraints, fading multipath, imprecise clocks, the near-far effect, tropospheric delay and legislative obstructions make them difficult to implement. This thesis takes a step forward, by proposing a loosely coupled integration with Locata, a novel, terrestrial positioning technology, based on the pseudolite concept. It avoids the above pitfalls by utilising frequency and spatially separated antennas and a license-free frequency band, though this comes at the cost of in-bound interference. Its ability to provide stand-alone position and network synchronisation at nanosecond level is used commercially in open-cast mining and in military aviation. Discussion of Locata and GPS technology has identified their shortcomings and main limiting factors as well as the advantages of the proposed integration. During the course of this research, tropospheric delay, planar solution and known point initialisation ambiguity resolution methods have been identified as the main limiting factors for Locata. These are analysed in various static and kinematic scenarios. Discussion also includes ambiguity resolution, noise and interference detection and system performance in indoor and outdoor scenarios. The proposed navigation engine uses a closely coupled integration at the measurement level and LAMBDA as the ambiguity resolution method for Locata and GPS. A combined solution is demonstrated to offer a geometrical improvement, especially in the respect of height determination, with centimetre to decimetre accuracy and a minimum requirement of two signals from any component. This study identifies that proper separation and de-correlation of Locata and GPS ambiguities and better tropospheric models are essential to reach centimetre level accuracy. The thesis concludes with examples of system implementation including: seamless navigation, city-wide network deployment, urban canyons, a long term-monitoring scenario and indoor positioning. This demonstrates how the proposed navigation engine can be an advantage in areas such as: civil engineering, GIS, mobile mapping, deformation, machine navigation and control.