hht@sarnoff.sarnoff.com (Herbert H. Taylor x2733) (10/03/90)
This news group seems to focus on Real-time Operating System issues. We are interested in research projects in Real-time DSP, particularly "continuous" real-time image processing and "continuous" multimedia. Are any research groups working in this area? What are considered to be the issues? Are there any good references? The latest generation of massively parallel SIMD systems should be able to support continuous real-time signal processing. Our SIMD system (2048 16bit DSP's a.k.a "the Princeton Engine") can and CM2 and MasPar should be able to as well (perhaps at reduced frame rate). Does anyone know of any projects of this scope? Herb Taylor
isr@rodan.acs.syr.edu (Michael S. Schechter - ISR group account) (10/04/90)
Here at the I.S.R., while i don't know all the details of it, as i specialize in building lab systems, a couple of students wrote a real-time model of the Limulus (horseshoe crab) retina which runs on a CM2. each processor simulated a single retina cell or neuron and were interconnected the same way as the cells in the actual critter. Each cell was modelled to simulate it's behaivor when isolated, and amazingly enough, the connected computer model had behaivor close to that of the entire eye. -- Mike Schechter, Computer Engineer,Institute Sensory Research, Syracuse Univ. InterNet: Mike_Schechter@isr.syr.edu isr@rodan.syr.edu Bitnet: SENSORY@SUNRISE