Software-Defined Intrusion Detection System for DDoS Attacks in IoT Edge Networks

Bolat-Akça, Burcu; Bozkaya, Elif

doi:10.1109/dasc/picom/cbdcom/cy59711.2023.10361494

Cited by 2 publications

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Blockchain-Based Secure Content Caching and Computation for Edge Computing

Bozkaya-Aras

2024

IEEE Access

View full text Add to dashboard Cite

The current explosion in user traffic necessitates the placement of edge servers in proximity to the Internet of Things (IoT) devices, allowing computation tasks to be offloaded to edge servers. This strategy aims to minimize the average delay of traffic requests by enabling user/IoT devices to locally execute timesensitive tasks or offload them to edge servers within the mobile edge computing (MEC) paradigm. This new paradigm will also allow to cache contents at edge servers, but considering such an increasing number of user requests and limited storage capability of edge servers, selection in edge caching decisions is challenging. In addition, while there is general consensus that this technology may provide a variety of benefits, there are serious questions about its security implications. This is because malicious users can manipulate the caching decisions of the edge servers by sending fake traffic requests, which reduces the caching efficiency of the resource-constrained edge servers. Driven by these issues, in this paper, we propose a blockchain-based content caching and computation strategy to validate the authenticity of cached content and thus prevent unauthorized requests from malicious users. Specifically, the Proof of Stake (PoS) consensus mechanism is presented to handle low computational work, validate the process of blocks, and manage the transactions between edge servers and legitimate users. Then, a Deep Q Network (DQN)-based solution is proposed to intelligently develop an effective content caching and computation strategy. According to performance evaluation, the proposed model significantly outperforms the conventional methods. It improves the cache hit rate by up to 8.2% on average and reduces the response delay by up to 7.45% on average.

show abstract