The energy cost of network security: A hardware vs. software comparison

  title={The energy cost of network security: A hardware vs. software comparison},
  author={Andre Luiz Pereira de Franca and Ricardo P. Jasinski and Paulo Cemin and Volnei A. Pedroni and Altair Olivo Santin},
  journal={2015 IEEE International Symposium on Circuits and Systems (ISCAS)},
The increasing network speeds, number of attacks, and need for energy efficiency are pushing software-based network security to the limit. A common kind of threat is probing attacks, in which an attacker tries to find vulnerabilities by sending many probe packets to a target machine. In this paper, we evaluate three machine learning classifiers (Decision Tree, Naive Bayes, and k-Nearest Neighbors), implemented in hardware and software, for the detection of probing attacks. We present detailed… Expand
Towards an Energy-Efficient Anomaly-Based Intrusion Detection Engine for Embedded Systems
It is demonstrated that a hardware (HW) implementation of network security algorithms can significantly reduce their energy consumption compared to an equivalent software (SW) version. Expand
A reliable and energy-efficient classifier combination scheme for intrusion detection in embedded systems
This paper presents an anomaly-based method for network intrusion detection in embedded systems that maintains the classifier reliability even when network traffic contents changes and is energy-efficient and well suited for hardware implementation. Expand
Network Intrusion Detection Using Neural Networks on FPGA SoCs
  • L. Ioannou, Suhaib A. Fahmy
  • Computer Science
  • 2019 29th International Conference on Field Programmable Logic and Applications (FPL)
  • 2019
This work presents an approach for network intrusion detection using neural networks, implemented on FPGA SoC devices that can achieve the required performance on embedded systems. Expand
IoT Security through the Lens of Energy Efficiency: Energy as a First-Order Security Consideration
  • David E. Ott
  • Engineering
  • 2016 Cybersecurity Symposium (CYBERSEC)
  • 2016
IoT devices are expected to proliferate significantly in the coming years. While computational requirements have, in general, become more demanding, capacity gains in device battery technology haveExpand
Tuning Stacked Auto-encoders for Energy Consumption Prediction: A Case Study
This work investigates the effects of tuning stacked auto-encoder in computer processor with regard to the energy consumption, and proposes a grid search based training method and a data preprocessing algorithm to prepare data to prediction. Expand


Moving Network Protection from Software to Hardware: An Energy Efficiency Analysis
This paper presents a new way to improve the throughput and to reduce the energy consumption of an anomaly-based intrusion detection system for probing attacks, and implements the same classifier algorithm in software (C++) and in hardware (synthesizable VHDL). Expand
Investigating the utility of S-transform for detecting Denial-of-Service and probe attacks
A novel signal and image processing-based method for detecting network probe and DoS attacks in which prior knowledge of attacks is not required and the results clearly indicated that the method is effective for detecting the probe and doS attack streams which were generated to real-world Internet. Expand
Intrusion detection by machine learning: A review
This chapter reviews 55 related studies in the period between 2000 and 2007 focusing on developing single, hybrid, and ensemble classifiers and discusses current achievements and limitations in developing intrusion detection systems by machine learning. Expand
An FPGA-Based Network Intrusion Detection Architecture
This work designs an FPGA-based architecture for anomaly detection in network transmissions and demonstrates the use of principal component analysis as an outlier detection method for NIDSs. Expand
Power profiling and optimization for heterogeneous multi-core systems
The approach shows that, for N-body computation, the fastest design which involves 2 CPU cores, 10 FPGA cores and 40960 GPU threads, is 2 times faster than a design with only FPGAs while achieving better overall energy efficiency. Expand
Machine learning techniques for data mining: A survey
This survey provides a comparative Analysis of various classification algorithms and presents the various classification techniques including decision tree, Support vector Machine, Nearest Neighbor etc. Expand
Bandwidth, Packets Per Second, and Other Network Performance Metrics
  • 2009