Combating Informational Denial-of-Service (IDoS) Attacks: Modeling and Mitigation of Attentional Human Vulnerability

This work proposes a new class of proactive attacks called the Informational Denial-of-Service (IDoS) that exploit attentional human vulnerability. By generating large volume feints, IDoS deplete cognitive resources operators to prevent humans from identifying real hidden among feints. aims formally define attacks, quantify their consequences, and develop human-assistive security technologies mitigate severity level risks attacks. To this end, we use semi-Markov process model sequential arrivals feints with category labels attached in associated alerts. The assistive technology strategically manages attention by highlighting selective alerts periodically distraction other A data-driven approach is applied evaluate performance under different Attention Management (AM) strategies. Under representative special case, establish computational equivalency between two dynamic programming representations reduce computation complexity enable online learning samples reduced size zero delays. case study corroborates effectiveness framework. numerical results illustrate how AM strategies can alleviate risk Furthermore, show minimum achieved proper intentional inattention alerts, which refer as law rational risk-reduction inattention.