Pulse Processing Techniques for Ultra Wide Band Electromagnetic Tomography

点击翻译按钮获取中文摘要

The demand for improved products and processes has driven research in using tomography techniques for industrial and environmental applications. As a result, a range of sensing modalities have been researched and developed, embracing electrical, electromagnetic (EM), acoustic, optical and nucleonic phenomena, with each modality offering particular advantages for their target application areas.Electrical and electromagnetic tomography (EMT) systems are fast, low cost, non-intrusive and relatively safe. However, their image quality is relatively poor compared to other modalities such as X-ray, MR, US, etc. This is because electrical systems have a limited number (say 23 to 25) of fixed, single frequency, excitation / detection elements. Consequently, the number of independent measurements upon which an image frame can be reconstructed is limited (typically 102 to 103), and therefore high-resolution image reconstruction is difficult without the use of a priori knowledge. Another fundamental is limit is that the object material affects the direction of the interrogating EM fields. This is often known as the soft field effect and presents a major challenge for image reconstruction. Consequently the range of applications for electrical and EMT systems has been limited to those where only modest resolution is acceptable. Clearly a high resolution EM technique would represent a major breakthrough for the discipline.UWB electromagnetic tomography (EMT) is based on the propagation of EM pulses through the target and offer the potential for high-resolution imaging, especially if timing jitters of <10ps can be achieved (equivalent to a sub mm distance for a pulse travelling in water) and pulse dispersion mechanisms can be adequately modelled. High resolutions are possible, compared to standard electrical tomography techniques, because precision timing information is available along with much greater bandwidth. This project considers pulse processing techniques which may help to improve image quality by providing more accurate estimations of the arrival times of the pulse components.