Andrew Ensor

Implementation of a Fast Fourier transform algorithm onto a manycore processor

The Fourier transform is the main processing step applied to data collected from the Square Kilometre Array (SKA) receivers. The requirement is to compute a Fourier transform of 219 real byte samples in real-time, while minimizing the power consumption. We address this challenge by optimizing a FFT implementation for execution on the Kalray MPPA manycore processor. Although this processor delivers high floating-point performances, we use fixed-point number representations in order to reduce the memory consumption and the I/O bandwidth. The result is an execution time of 1,07ms per FFT, including data transfers. This enables to use only two first-generation MPPA chips per flow of data coming from the receivers, for a total power consumption of 17.4W.

Colored-Edge Graph Approach for the Modeling of Multimodal Transportation Systems

Many networked systems involve multiple modes of transport. Such systems are called multimodal, and examples include logistic networks, biomedical phenomena and telecommunication networks. Existing techniques for determining minimal paths in multimodal networks have either required heuristics or else application-specific constraints to obtain tractable problems, removing the multimodal traits of the network during analysis. In this paper weighted colored-edge graphs are introduced for modeling multimodal networks, where colors represent the modes of transportation. Minimal paths are selected using a partial order that compares the weights in each color, resulting in a Pareto set of minimal paths. Although the computation of minimal paths is theoretically intractable and 𝒩𝒫-complete, the approach is shown to be tractable through experimental analyses without the need to apply heuristics or constraints.

ColourFAST: GPU-based feature point detection and tracking on mobile devices

A realtime feature point detection algorithm called ColourFAST is introduced. ColourFAST extracts vector-based feature strength and direction measures from the colour channels of any pixel in an image. The extracted information is applied to create an effective feature point tracker. These feature point and tracker algorithms have a pipeline design optimized for GPU processors. Results are provided for an implementation on mobile devices developed using programmable shaders. Its performance demonstrates several improvements over conventional FAST feature point detection and Lucas-Kanade tracking.

On the Tractability of Shortest Path Problems in Weighted Edge-Coloured Graphs

A weighted edge-coloured graph is a graph for which each edge is assigned both a positive weight and a discrete colour, and can be used to model transportation and computer networks in which there are multiple transportation modes. In such a graph paths are compared by their total weight in each colour, resulting in a Pareto set of minimal paths from one vertex to another. This paper will give a tight upper bound on the cardinality of a minimal set of paths for any weighted edge-coloured graph. Additionally, a bound is presented on the expected number of minimal paths in weighted edge–bicoloured graphs. These bounds indicate that despite weighted edge-coloured graphs are theoretically intractable, amenability to computation is typically found in practice.

Precision analysis of the imaging pipeline in the square kilometre array

In this paper we present our end-to-end model of the imaging pipeline in the Square Kilometre Array. Our Sky Generator models the signals that are received by the Central Signal Processor (CSP), our CSP Correlator model then processes those signals to generate visibilities to pass to the Science Data Processor (SDP). Our SDP Imaging model then grids the visibilities and inverse Fourier transforms them to produce a dirty image of the sky. Our modelling allows us to investigate the error that is introduced due to reduced numerical precision, and we then propose techniques to mitigate this error, and thus reduce the required amount of computational hardware.

GPU-accelerated feature point matching using extended ColourFAST descriptors

A real-time feature point matching algorithm is introduced. It extracts vector-based ColourFAST feature strength and direction measures from the colour channels of the pixels in an image. This information is combined with the relative locations of the feature points to provide frame-by-frame scale and rotation invariant matching. The resulting algorithm is specifically designed for high throughput and optimised for GPU pipelining on embedded devices. Results are given showing high-framerate matching is achieved on 720p resolution images.