HCDec 28, 2017

Modelling Noise-Resilient Single-Switch Scanning Systems

arXiv:1712.10073v12 citations
Originality Incremental advance
AI Analysis

This addresses communication challenges for nonspeaking individuals with motor disabilities, but appears incremental as it builds on existing scanning methods.

The paper tackled the problem of single-switch scanning systems failing to model noise sources in realistic use by individuals with motor disabilities, and developed a mathematical model with noise modelling and a fast-scan improvement, showing via simulation it can be more suitable for certain users.

Single-switch scanning systems allow nonspeaking individuals with motor disabilities to communicate by triggering a single switch (e.g., raising an eye brow). A problem with current single-switch scanning systems is that while they result in reasonable performance in noiseless conditions, for instance via simulation or tests with able-bodied users, they fail to accurately model the noise sources that are introduced when a non-speaking individual with motor disabilities is triggering the switch in a realistic use context. To help assist the development of more noise-resilient single-switch scanning systems we have developed a mathematical model of scanning systems which incorporates extensive noise modelling. Our model includes an improvement to the standard scanning method, which we call fast-scan, which we show via simulation can be more suitable for certain users of scanning systems.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes