Simon C. Pomeroy

Interference cancellation and alignment techniques for multiple-input and multiple-output cognitive relay networks

We consider a multiple-input and multiple-output (MIMO) cognitive radio (CR) network with a MIMO relay that opportunistically accesses the same frequency band as that of a MIMO primary network. In particular, both interference cancellation and interference alignment techniques have been investigated to enhance the achievable degrees of freedom (DoF) for the MIMO CR network. Based on the number of antennas at the primary network and the secondary network, the authors analytically quantify the maximum achievable DoF of the secondary network by using the proposed techniques. It is shown that the DoF obtained by the CR network in the presence of a MIMO relay is higher than that could be obtained without a relay. The analyses consider both sufficient and insufficient number of antennas at the relay in terms of the ability to separate and decode both the primary and secondary transmitted signals. The simulation results support the analytically quantified achievable DoF results.

Underwater noise assessment for energy extraction and production systems using unmanned arial vehicles (UAVs)

Traditional operations such as oil and gas exploration and production have long undergone sound field and environmental impact assessments of underwater acoustic noise. More recently emerging industries such as renewables (wind, wave, and tidal energy production) have also required scrutiny in terms of underwater noise sound fields. To make these assessments, sound fields are typically measured using hydrophones deployed from boats, drifting systems or moored acoustic data loggers. These measurements are often complex and expensive requiring complicated equipment deployments, boat operations and personnel in often-dangerous or hazardous environments. Unmanned Ariel Vehicles (UAV) or drone based technologies offer the opportunity for rapid deployment of smart hydrophone systems arrays over a large spatial area with significantly lower operator and boat interaction improving deployment flexibility, cost and minimising safety concerns for boat based deployments. Results presented are from tests of a prototyp...

Evaluation of UAVs as an underwater acoustics sensor deployment platform

ABSTRACT Marine surveys carried out by Passive Acoustic Monitors conventionally use towed hydrophone arrays, which requires dedicated surface observation boats. This is a costly and slow process, which could be made cheaper and quicker by using Unmanned Aerial Vehicles (UAVs). Presented in this paper are the initial findings from using UAVs to capture underwater acoustic signals from an acoustic test tank.

Computational electromagnetics for mobile devices on multiple platforms

Mobile computing devices are ubiquitous and can now carry out computationally-intensive tasks in a timely fashion. This paper presents multi-platform, computational electromagnetics educational software for modern mobile platforms. Performance of the Transmission Line Matrix (TLM) solver has been increased using a Single Instruction, Multiple Data (SIMD) processing strategy applicable to the majority of available platforms. Included is discussion of the SIMD strategy, performance benchmarking results and user uptake statistics.

SIMD acceleration of a TLM solver using ARM processors

In recent years, investigations into accelerating Transmission Line Matrix (TLM) solver programs have concentrated on the use of large-scale parallel computing systems. The focus for such systems rests on leveraging multi-core software architecture principles to achieve improved performances. Mirroring desktop CPU advances of the early 2000s, modern small-scale computing platforms are increasingly transitioning to dual or even quad-core CPU architectures to provide developers with opportunities for improving software performance. This paper discusses an alternative to traditional multi-core techniques for obtaining performance increases with such architectures. Using the NEON Single Instruction, Multiple Data (SIMD) instruction set present in the ARMv7 architecture processors often found within modern small-scale computing devices, we demonstrate performance increases nearly two times base figures without the normal pitfalls of multi-core software design.

Benchmarking of a TLM solver on mobile devices sharing a common hardware architecture

Modern mobile communication devices share a common set of hardware and software features. Within a brand, it is also common for similar hardware architectures to be used, and to be paired with the same operating system. Whilst mathematically trivial in nature, the Transmission Line Matrix (TLM) method is computationally-intensive, and therefore varies in performance dependant on available resources. This paper investigates how the performance of a mobile Computational Electromagnetics (mCEM) laboratory iPhone application is effected when executed on devices with different hardware configurations, but that are available from the same manufacturer. Hardware differences across the chosen platform are first discussed, with performance predictions made based on the device specifications. TLM is suitable for parallel computing implementations, and therefore potential multi-threading optimisation strategies are discussed for applicable devices. Results from the application performance benchmarking are presented, and demonstrated that the iPad 2 offered the best processing performance of the devices tested.

Raindrop modelling for high frequency propagation studies

At UHF and EHF wavelengths, air-born precipitation has a significant attenuating and depolarising effect on transmitted radio signals. Analysis of these effects in the time domain, using methods such as the Transmission Line Matrix (TLM) method requires an accurate reproduction of the raindrop geometry in order to obtain the most relevant results. This paper discusses the characteristics of different mathematical models for representing raindrops of various sizes. Justification for the selection of the Beard & Chuang Model as the preferred geometrical model is provided, along with details of implementing a raindrop model in CST MicroStripes and conducting simulations using a square cross-section waveguide. Discussion regarding use of the geometry data as part of a larger investigation is also included.

A Framework for teaching electromagnetic concepts using mobile devices

A novel pedagogical approach to the teaching of wave phenomena is presented. The growing popularity of smartphones is exploited to provide students with a personal, mobile modelling environment based on the Transmission Line matrix method (TLM). This supports material delivered by more traditional means. There are opportunities for features of the mobile device, together with the software delivery mechanism, to guide students through different levels of complexity in the teaching. This would allow the software on the device to give an accessible, hands-on experience to the students. No significant access to external computing resources is required, limiting the amount of data traffic generated by the use of the software. An initial proof of concept example is described with suggestions for development into a full learning framework.

A benchmark to quantify frequency domain errors from spatial quantization in time domain models

This paper proposes a benchmark to quantify frequency and amplitude variations caused by the spatial quantization of model geometry. A simple resonant cavity is modelled in 3 dimensions at a range of angles to the mesh axes. The quality measure resulting is compared with a classical analytical solution for such a cavity. Results from the Transmission Line Matrix method (TLM) are presented. The benchmark could be applied to most volume element based time domain techniques.

Low resolution radiation pattern sampling for a thin dipole

During antenna measurements the field pattern is typically sampled in discrete radiation directions and this inevitably affects the accuracy of the results. This paper considers the problem of sampling accuracy by taking an analytical solution for a thin dipole and numerically sampling using two different methods. The first assumes that the antenna is rotated in equal angle steps, the second uses a point set generated from a sphere partitioning algorithm. It is shown that the errors produced by the latter method are on average lower than the former. It is also inferred that the sphere-partitioning sampling method exhibits lower sensitivity to alignment of the antenna relative to the measurement system coordinate system.