مشخصات مقاله:
عنوان فارسی مقاله:
توزیع SlowDoS در زمان واقعی و تشخیص حملات بر روی ترافیک رمزگذاری شده با استفاده از هوش مصنوعی
عنوان انگلیسی مقاله:
Distributed real-time SlowDoS attacks detection over encrypted traffic using Artificial Intelligence
سال انتشار مقاله:
2021
کلمات کلیدی مقاله:
امنیت سایبری، هوش مصنوعی، حملات سایبری، یادگیری ماشین
مناسب برای رشته های دانشگاهی زیر:
کامپیوتر
مناسب برای گرایش های دانشگاهی زیر:
هوش مصنوعی، مهندسی نرم افزار، امنیت اطلاعات
وضعیت مقاله انگلیسی و ترجمه:
مقاله انگلیسی را میتوانید به صورت رایگان با فرمت PDF با کلیک بر روی دکمه آبی، دانلود نمایید. برای ثبت سفارش ترجمه نیز روی دکلمه قرمز رنگ کلیک نمایید. سفارش ترجمه نیازمند زمان بوده و ترجمه این مقاله آماده نمیباشد و پس از اتمام ترجمه، فایل ورد تایپ شده قابل دانلود خواهد بود.
فهرست مطالب:
Abstract
Keywords
1. Introduction
2. Related work
3. Background
3.1. Intrusion Detection systems (IDS)
3.2. Slow rate denial-of-service against web applications
4. AI-based cyber-attacks detection framework for application-level encrypted traffic
4.1. Real-time network monitoring
4.2. Conversation processing
4.3. Applying artificial intelligence techniques
4.4. Publisher/subscriber system
4.5. Streaming processing
5. Artificial intelligence for SlowDoS attacks detection
5.1. Model training
5.1.1. Phase 1. preprocessing
5.1.2. Phase 2. clustering
5.1.3. Phase 3. histogram matrix
5.1.4. Phase 4. deep learning training
5.2. Attack detection
6. Implementation and evaluation
6.1. System implementation and testbed
6.2. Generated dataset description
6.3. Quality results
6.4. System performance results
7. Conclusions and future work
Credit author statement
Declaration of competing interest
Acknowledgements
References
قسمتی از مقاله انگلیسی:
1. Introduction
Denial-of-Service-(DoS) based attacks are continuously evolving increasing its complexity and range, thereby making more and more difficult to perform a timely and accuracy detection. Traditional DoS attacks, that aims to incapacitate a resource from serving its genuine clients, have been intensively studied in the literature through different schemes intended to protect network infrastructures [1]. DoS attacks, concretely application-layer DoS attacks are recently getting research attraction [2], since they are able to compromise a web-server through other means beyond traditional ones such as network flooding or exhausting server’s resources such as sockets, memory, CPU, and I/O bandwidth. In particular, SlowDoS attacks [3] are a type of application-layer DoS using low-rate packet transmission [4]. Namely, most of the SlowDoS attacks, such as Slowris or SlowPost, exploits HTTP protocol, widely adopted in application-layer services, by sending incomplete http requests, or keeping the connection with the server busy through sending the HTTP posts using slow ratio and without reaching content-length values.
Furthermore, the prominence of the Internet of Things [5] is introducing millions of interconnected network devices which suggest new scenarios and connectivity models that are increasing the attack surface. Since neither highperformance systems nor immense network bandwidth are needed to perform a SlowDoS attack, simple constrained IoT devices are suitable to carry out a denial. Indeed, the biggest advantage of SlowDoS attacks relies on the scarce bandwidth needed, meaning few IoT attackers –i.e bots– using a low-rate packet transmission can overwhelm their victim