A Collusion-Resilient Hybrid P2P Framework for Massively Multiplayer Online Games

Massively Multiplayer Online Games (MMOG), on which many users play simultaneously in a large scale virtual space on the Internet, have long been popular services. Most of MMOGs work based on Client / Server (C/S) model. Although C/S model is easy to manage games, it lacks scalability since the capacity of storages and the performance of servers that needs to be maintained by administrators increases in proportion to the number of users. In order to solve the problem of scalability, it is expected that development of a P2P-based MMOGs framework takes the place of the conventional C/S model.P2P MMOGs has a disadvantage that cheats are easily performed compared to the C/S model. There are various studies to prevent cheats on P2P models, but research on frameworks considering the possibility of collusion cheats has been rarely done so far.In this paper, we propose a collusion-resilient hybrid P2P framework for MMORPG. We first analyze the features of RPG and we clarify the requirements. We investigate cheats which may be caused in a P2P framework and we categorize cheats into three. Then, we consider about the influences and detectability of them. In order to prevent the three types of cheats, it is necessary to calculate data concealed from users on the process of events. Therefore, we suppose that there are “scramble” that encrypt data in order to conceal what the data means and that we can calculate the scrambled data without decryption. We have devised a method using secret sharing based on Chinese Remainder Theorem that has almost all properties for scramble, and discuss its use in the hybrid P2P framework. We have compared the proposed framework to conventional frameworks and a framework with ideal “scramble,” with respect to the resistance against cheats.