Real time data acquisition with transputers and PowerPCs using the wavelet transform for event detection
K. Hallatschek, Manfred Zilker · IEEE Transactions on Nuclear Science · 1998
In this paper a new data acquisition system for the soft-X-ray diagnostics in a large fusion plasma experiment is described. The main purpose of this system is to provide real time facilities for analyzing the acquired data, to generate necessary information for the dynamical adaptation of sample rates, and to deliver triggers when certain events in the plasma are detected. The hardware is built on a network of up to 72 transputers and up to four modules, each consisting of a PowerPC and a transputer coupled via dual-ported RAM. The transputers are used for collecting the incoming data from ADCs (each sampled with 500 KHz), and for communication and synchronization within the computer network. Four PowerPCs can analyze up to eight selected channels in real time and provide the results to the other transputers in the network. The algorithm running on the PowerPCs is performing a wavelet like time-frequency transform for the detection of plasma events. With this system it was possible to observe high-n/m mode cascades following impurity accumulation. A report on this effect has been published in Physical Review Letters.