Layered representations for learning and inferring office activity from multiple sensory channels
Computer Vision and Image Understanding - Special issue on event detection in video
Hierarchical hidden Markov models with general state hierarchy
AAAI'04 Proceedings of the 19th national conference on Artifical intelligence
Hi-index | 0.00 |
Generative models are created to be used in the design and performance assessment of high layer wireless communication protocols and some error control strategies. Generative models can replace real digital wireless channels to significantly reduce the time and complexity of system simulation. The errors occurring in digital wireless channels are not independent but form clusters or bursts. Generative models have to produce error sequences having similar burst error statistics to those of original error sequences obtained from real digital systems. In this paper, we propose a generative hidden Markov model (HMM) with three layers. It is shown that the proposed three layered HMM can generate error sequences that have statistics compatible with those of original error sequences derived from an enhanced general packet radio service (EGPRS) transmission system.