We conduct research in the exploitation of off-the-shelf FPGA technology e.g. from Xilinx and Altera, in digital signal processing applications, including image and video processing, digital communication, and data processing in general. A key objective in this research is to satisfy the dual requirement of efficient hardware and flexibility. Central to the satisfaction of this requirement is the concept of hardware skeletons which can be thought of as higher order functions which take not only scalars as parameters but also functions, possibly in the form of other skeletons. By focussing on particular application domains, a finite and often small number of basic skeletons can be found, from which a large set of applications can constructed.

Current research projects in this area include the design and development of efficient and adaptive FPGA-based MIMO detectors, generic hardware skeletons for classification algorithms e.g. k-nearest neighbour and SVM, and generic hardware skeletons for video processing applications.