A majority voter for intrusion tolerant software based on N-version programming techniques

One of the drawbacks of the existing majority voters, which are widely used in N-version programming (NVP) technique, is the high probability of agreement on incorrect results generated by variants. Therefore, to propose an intrusion-tolerant software architecture based on NVP for hostile environments and to consider possible attacks, a new voting scheme is required. In this paper, we propose a voting scheme to improve the correctness of the binary majority voters in the hostile environments to treat the situations that more than half of the variants may have been compromised. We have used stochastic activity networks (SANs) to model the scheme for a triple-version programming (3VP) system and measure the probability of detecting the correct outputs by the voter. The evaluation results showed that the proposed scheme can improve the correctness of the classic majority voting algorithms to detect the correct output, especially the intrusion detection mechanisms are used in the scheme.