Adaptive Update Distribution Estimation under Probability Byzantine Attack  

作　　者：Gang Long Zhaoxin Zhang 

机构地区：[1]Faculty of Computing,Harbin Institute of Technology,Harbin,150000,China

出　　处：《Computers, Materials & Continua》2024年第10期1667-1685,共19页计算机、材料和连续体（英文）

摘　　要：The secure and normal operation of distributed networks is crucial for accurate parameter estimation.However,distributed networks are frequently susceptible to Byzantine attacks.Considering real-life scenarios,this paper investigates a probability Byzantine(PB)attack,utilizing a Bernoulli distribution to simulate the attack probability.Historically,additional detection mechanisms are used to mitigate such attacks,leading to increased energy consumption and burdens on distributed nodes,consequently diminishing operational efficiency.Differing from these approaches,an adaptive updating distributed estimation algorithm is proposed to mitigate the impact of PB attacks.In the proposed algorithm,a penalty strategy is initially incorporated during data updates to weaken the influence of the attack.Subsequently,an adaptive fusion weight is employed during data fusion to merge the estimations.Additionally,the reason why this penalty term weakens the attack has been analyzed,and the performance of the proposed algorithm is validated through simulation experiments.

关 键 词：Distribution estimation network security least-mean-square binomial distribution probability Byzantine attack 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]