Abstract
The Prism (precise, repeat integral signal monitor), is a recently developed signal processing object [1, 2]. It is based upon a Fourier-style double integration applied recursively via a sliding windows calculation. Either one or two outputs are generated. If there are two outputs, these form an orthogonal sine/cosine pair from which the properties of the input signal, such as amplitude, phase and frequency, can be calculated. Unlike conventional, convolution-based FIR filters, the Prism calculation is fully recursive, so it has a low computational cost, which is independent of window length. Other benefits include a linear phase response, and a low design cost, as the filter ‘coefficients’ are simply linearly spaced sine and cosine values. Although the Prism frequency response is relatively inflexible compared to conventional FIR and IIR filter designs, networks of Prisms can be designed to provide a wide range of responses.
This talk will provide an introduction to the basic mathematics of the Prism, and describe some elementary Prism signal processing techniques such as signal tracking, notch and bandpass filtering, and the tracking of multiple frequency components in a signal. As a worked example, a Prism-based ultra-narrowband filter (say 1 billion samples in length) can run in real time on a single FPGA, whereas the conventional convolution-based equivalent would require supercomputing resources to run at the same sample rate.
The talk will include a brief survey of current and future applications of Prism Signal Processing, and demonstrate how it matches the requirements for sensor development for the era of the Internet of Things.
References
[1] Henry, MP, Leach, F, Davy, M, Bushuev, O, Tombs, MS, Zhou, FB and Karout, S. “The Prism: Efficient Signal Processing for the Internet of Things”, IEEE Industrial Electronics Magazine, pp 2–10, December 2017. DOI: 10.1109/MIE.2017.2760108.
[2] Henry, MP. “The Prism: recursive FIR signal processing for instrumentation applications”. IEEE Transactions on Instrumentation and Measurement, May 2020. http://dx.doi.org/10.1109/TIM.2019.2916943
This talk will provide an introduction to the basic mathematics of the Prism, and describe some elementary Prism signal processing techniques such as signal tracking, notch and bandpass filtering, and the tracking of multiple frequency components in a signal. As a worked example, a Prism-based ultra-narrowband filter (say 1 billion samples in length) can run in real time on a single FPGA, whereas the conventional convolution-based equivalent would require supercomputing resources to run at the same sample rate.
The talk will include a brief survey of current and future applications of Prism Signal Processing, and demonstrate how it matches the requirements for sensor development for the era of the Internet of Things.
References
[1] Henry, MP, Leach, F, Davy, M, Bushuev, O, Tombs, MS, Zhou, FB and Karout, S. “The Prism: Efficient Signal Processing for the Internet of Things”, IEEE Industrial Electronics Magazine, pp 2–10, December 2017. DOI: 10.1109/MIE.2017.2760108.
[2] Henry, MP. “The Prism: recursive FIR signal processing for instrumentation applications”. IEEE Transactions on Instrumentation and Measurement, May 2020. http://dx.doi.org/10.1109/TIM.2019.2916943
| Original language | English |
|---|---|
| Publication status | Published - 30 Aug 2020 |
| Event | 2020 Academic Conference of China Instrument and Control Society - Shanghai, China Duration: 28 Aug 2020 → 30 Aug 2020 |
Conference
| Conference | 2020 Academic Conference of China Instrument and Control Society |
|---|---|
| Country/Territory | China |
| City | Shanghai |
| Period | 28/08/20 → 30/08/20 |