Extensive game analysis and improvement strategy of DPOS consensus mechanism  

作　　者：Wang Lei Zhu Qinghua Li Baozhen 

机构地区：[1]School of Information Management,Nanjing University,Nanjing 210023,China [2]School of Information Engineering,Nanjing Audit University,Nanjing 211815,China

出　　处：《The Journal of China Universities of Posts and Telecommunications》2021年第5期27-35,101,共10页中国邮电高校学报（英文版）

基　　金：supported by the National Natural Science Foundation of China(71673122,72074117);the Fund of Social Sciences in Jiangsu Province(20WTB007);the Scientific Research Fund of Education Ministry-China Mobile(D203209000115)。

摘　　要：Delegated proof-of-stake(DPOS) consensus mechanism is widely adopted in blockchain platforms, but problems exist in its current applications. In order to explore the security risks in the voting attack of the DPOS consensus mechanism, an extensive game model between nodes was constructed, and it was concluded that the DPOS consensus mechanism relies too much on tokens, and the possibility of node attacks is very high. In order to solve the problems of frequent changes of DPOS consensus mechanism nodes, inactive node voting, excessive reliance on tokens, and malicious nodes, a dynamic, credible, and attack-evading DPOS consensus mechanism was proposed. In addition, the Python simulation results show that the improved Bayesian voting algorithm is effective in calculating node scores.

关 键 词：blockchain delegated proof-of-stake(DPOS)consensus mechanism extensive game Bayesian voting algorithm the objectivity of voting 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]