TY - JOUR TI -

Analysis and forecast of undesirable cloud services traffic

T2 - IS - KW - forecasting KW - DDOS attack KW - cloud services KW - network traffic KW - modeling KW - additive time series model KW - autocorrelation function KW - error estimation AB -       These days one of the main problems that must be solved to ensure information security in cloud services for corporations as well as for individual clients is to correctly identify and predict hacking in the network traffic. This paper presents statistics on information security threats, provides classification of information security threats for cloud services, identifies hackers’ goals, and proposes countermeasures.      A vital task is to develop an effective method that could be used to protect cloud services from various network threats, as well as to analyze the network traffic. For these purposes, we chose a method based on an additive time series model, which allows us to predict the undesirable network traffic. To test this method, we obtained quantitative parameters for the undesirable traffic by simulating a network attack and collecting empirical data that describe this process. We used special software that simulates a network attack, and software that records and processes all the empirical data needed for the research.      Using the data obtained, we analyzed the efficiency of the method based on the additive time series model. We demonstrated that this method is also applicable for research into the general dynamics of the number of network attacks in cyberspace. This method also allows us to reveal how the dynamics of the number of hacker network attacks depends on season, date, or time. The results show that, based on data describing the network traffic, one can identify and predict the undesirable hacker threats. AU - Marina Tumbinskaya AU - Bulat Bayanov AU - Ruslan Rakhimov AU - Nikita Kormiltcev AU - Alexander Uvarov UR - https://bijournal.hse.ru/en/2019--1 Vol.13/269676885.html PY - 2019 SP - 71-81 VL -