Operator Alertness/Workload Assessment Using Stochastic Model-Based Analysis of Myoelectric Signals
Abstract
This interim report documents the work done to this point on Autoregressive Integrated Moving-Average (ARIMA) model-based analysis of myoelectric signals. The ARIMA modelling procedure and the hardware required for collecting myoelectric data are described in detail. Pattern analysis methods for characterizing the myoelectric signals under different levels of alertness/ workload are discussed. Additionally, the various tasks in the Experimental Control Package that subjects must perform while being monitored are described. Finally, an analysis of data obtained during experimental sessions is provided giving some indication of discriminability of the ARIMA signatures over different task difficulty levels and subjects. Results of this analysis indicate that the first AR parameter is the most useful feature in differentiating workload/altertness level. Additionally, this feature was shown to be reliable for each underlying level of alertness or load in a given task.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 30, 1984
- Accession Number
- ADA144535
Entities
People
- Azad M. Madni
- Denis D. Purcell
- Richard I. Scopp
- Yee-yeen Chu