Real-time AI application for digital signal processing
Document Type
Conference Proceeding
Publication Date
1-1-1992
Abstract
The goal to design a well-behaved real-time system is to guarantee that the execution of a task does not take longer than the specified amount of time and always output the optimal results. This work studies the application of artificial intelligence in a real-time system to meet this goal. A timeconstrained reasoning system is built into a real-time digital signal processing (DSP) task, this task evaluates the timing constraints during each execution cycle and selects an optimal path to provide the best possible results. A simulator is implemented to emulate a real-time DSP task. This simulator consists of input signal generator (sine wave, square wave, random pulses, and white noise), a digital filtering processor with a built-in knowledge-based reasoning system, a time loading/higher priority interrupt simulator, and an off-line knowledge base training/learning center. Simulation results show that within timing constraints, a real-time DSP task with AI has better performance than the one without AI.
Identifier
85067443896 (Scopus)
ISBN
[0780306325, 9780780306325]
Publication Title
Singapore International Conference on Intelligent Control and Instrumentation Proceedings
External Full Text Location
https://doi.org/10.1109/SICICI.1992.637634
First Page
772
Last Page
777
Volume
2
Recommended Citation
Wang, David T.; Pan, William W.; and Chao, Daniel Y., "Real-time AI application for digital signal processing" (1992). Faculty Publications. 17344.
https://digitalcommons.njit.edu/fac_pubs/17344
