Modern online multiplayer games are complex heterogeneous distributed systems comprised of servers and untrusted clients, which are often engineered under considerable commercial pressures. Under these conditions, security breaches allowing clients to employ illegal behaviours have become common; current commercial approaches have limited capabilities for reacting rapidly to such threats. This paper presents an approach to the detection of a cheating player, and describes a proof-of-concept system designed to detect cheating play (specifically wall-hacking) through the analysis of player behaviour. This approach differs from current methods in that it does not rely on knowledge about specific vulnerabilities and their method of exploitation in order to protect the system, but instead monitors player behaviour for indications of cheating play. Statistical evidence is presented which shows that the proof-of-concept correctly distinguishes between most cheating and non-cheating players.
|Publication status||Published - 2007|