LSTM-CGAN: Towards Generating Low-Rate DDoS Adversarial Samples for Blockchain-Based Wireless Network Detection Models

Abstract

Low-rate Distributed DoS (LDDoS) attack is a complex large-scale attack behavior with strong time-domain characteristics in blockchain-based wireless network. Blockchain with Machine learning-based models, as promising ways, are taken to detect them and secure wireless network. However, researchers focused on how to improve models' detection performance and work out new blockchain-based protection technologies during the past decades. Due to lack of evolving data, these models and technologies may have poor stability in the face of confrontational samples. To cope with the problem, this paper proposes a novel LSTM-CGAN method to generate high-quality LDDoS adversarial samples for blockchain-based wireless network detection models. In this method, we give a brief feature analysis about LDDoS attack in blockchainbased wireless network and work out its corresponding time series model firstly. And then, we take use of Long Short-Term Memory Networks (LSTM) to learn relationships among sequenced network packages in the same flow. At last, we establish a Condition Generative Adversarial Networks (CGAN) model to use above relationships as specific conditions for generating mimicking behaviors of LDDoS attacks in blockchain-based wireless network. The experimental results show that these generated adversarial samples based on both public and private datasets can cheat the machine learning detection models, and have the similar attack characteristics with the real samples. Consequently, they can be used as blockchain-based wireless network dataset of machine learning classifiers for training to enhance models' stability.

Internet of Things Applications, Security Challenges, Attacks, Intrusion Detection, and Future Visions: A Systematic Review

“…The link-flooding DDoS attacks are difficult to mitigate; LSTM is utilized in this work to review the attack patterns periodically. For a similar low rate DDoS attack pattern in wireless systems, Liu and Yin [165] used a combination of LSTM and CGAN. This is because LSTM works well in identifying patterns in sequenced packages.…”

Approaches towards Blockchain Innovation: A Survey and Future Directions

“…DDoS attacks try to disconnect a network mining pool, bringing down a server. In 2017, Bitfinex suffered from a DDoS attack [43]. Transaction malleability attacks will try to trick the victim to pay twice for a transaction.…”

Collaborative Learning for Cyberattack Detection in Blockchain Networks

“…This solution showed an accuracy of attack detection up to 99%. In [20], the authors used a public dataset and a private dataset from their testbed. Then, they proposed to use a Long Short-Term Memory Network (LSTM) to learn the properties of normal samples in the datasets.…”

“…However, these datasets were not designed for blockchain networks, and thus they are not appropriate to use in intrusion detection systems in blockchain networks. Other works, e.g., [19]- [21], tried to build their own datasets for blockchain networks, e.g., by obtaining the normal samples from the Bitcoin network [19], creating simulation experiment to detect the LFA [21] and generating artificial attack samples by CGAN [20]. However, these methods have several issues.…”