Front-end intelligence for triggering and local track measurement in gaseous pixel detectors

A number of applications in high-energy physics and medicine requires threedimensional reconstruction of the particle trajectories: for example, high momentum particles in accelerator-based experiments can be identified on the basis of the properties of their tracks, while in proton computed tomography accurate knowledge of the incoming and outgoing beam trajectory is crucial in reconstructing the most probable path of the proton traversing the patient. In this work we investigate the potential of Gaseous Pixel (GridPix) detectors for fast and efficient recognition of tracks and determination of their properties. This includes selection, without external trigger, of tracks with desired angles, for example tracks with small tilt angles corresponding to high momentum particles in a magnetic field. Being able to select these fast and without external input is of interest for the future upgrades of the LHC detectors. In this paper we present a track selection algorithm, and its physical implementation in 130 nm CMOS technology with estimates of power consumption, data rates, latency, and chip area. The Timepix3 chip, currently being designed for a wide range of applications, will also be suitable for readout of GridPix detectors. Both arrival time information (accuracy 1.6 ns) and charge deposit information will be delivered for each hit together with the coordinates of the active pixel. A short overview is presented of its architecture, which allows continuous self-triggered readout of sparsely distributed data with a rate up to 20 x 10 super(6) hits cm super(-2)sec super(-1). The addition of fast track pattern recognition logic to TimePix3 in a successor chip is currently being investigated.