English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

I N T R O D U C T I O N A trigger in I-.IEP experiment is a device which decides wheti~er an event is "interesting" and then must be stored or not (background) according to the configuration of the various pieces of data which make up the event. "r'he event selection is based upon few different conditions: one of these is to identify az~d count nuclear particle noised tracks which are parallel. This paper explores the possibility of using a hardware dedicated fuzzy processor designed to solve this pattern recognition problem. In most cases the event selection process requires a very short time: that is few microseconds, as in our case. The Fuzzy system Ill, we have developped, is much more powerful to detect and count parallel tracks in presence of positive or nega:ive noise in comparison with a u-aditional solution where a du'eshold system is used. To solve the problem we have simulated two Fuzzy systems: one. with a 6 input processor; one with two processors: the first has 4 inputs, the second 3 inputs and they work in parallel. ' t h e second solution is much more attractive both for the point of view of the necessary rules to be implemented and of the processing speed, because, having two processors working in parallel, the necessary time to process one event is much smaller. The number of possibile rules for the two processors also is much smaller than that o~e for the single processor. We hope that the hardware processor, here not illustrated, which we are designing at present with VLSI teclmiques, will work at 50 MLPS (Mega ln£er6nces Per Second). The trigger problem, in HEP, has been solved in the past year with dedicated HW. but to have a flexible solution DSP has been added to the system. Recently HW .neural networks [2], [3], [z,I have also been used. The problem of track finding, from a general point of view. has been faced with neural network {51, while, as far as we know. the problem reported in the title has non yet been faced. THE INPUT DATA The data to be processed come from a layer detector and can assume either high or low logic level. This detector is made of some planes (tea in this case), each of them is composed of a pixel linear array (in this case each array is made of 120 elements or pixels). When a particle crosses a pixel, it lights up. With x we indicate the position of the lit pixel along the plane while y indicates the plane itself. Y = 1 to l 0 ~ ~ , ,

A fuzzy system to detect and count parallel noised tracks