A Novel Voting Algorithm Based on Weighted Average Voting with a Classification Technique

Abstract

Weighted average voting has been widely used in control and computing systems. A novel voting algorithm based on combining the weighted average voter (WAV) with the minimum distance classification (MDC) technique. This voting algorithm uses WAV to indicate the minimum distance that will be used in the classification technique to select one measurement as the voter output. It is applicable for handling the output of an array of skewed sensors in safety related applications. The performance of the proposed voter is evaluated through a series of fault injection experiments and the results are compared with those of the exact majority voter [1], the WAV [2] and the enhanced weighted average voter (EWAV) [3]. The experimental results showed that the novel voter gives better performance in terms of reliability and complexity than the three other voters.

Towards Data Reliability Based on Triple Redundancy and Online Outlier Detection

“…They are optimized to provide an output with the least error. The combination of algorithms and the use of classification to find the best output increases the reliability but also the computational complexity and the processing time (Kassab, Hashad, Taha, & Shedied, 2013). However, for real-time processing, computation time must be taken into account.…”