Moving Network Protection from Software to Hardware: An Energy Efficiency Analysis

Software-based network security is constantly challenged by the increase in network speeds and number of attacks. At the same time, mobile network access underscores the need for energy efficiency. In this paper, we present a new way to improve the throughput and to reduce the energy consumption of an anomaly-based intrusion detection system for probing attacks. Our framework implements the same classifier algorithm in software (C++) and in hardware (synthesizable VHDL), and then compares the energy efficiency of the two approaches. Our results for a decision tree classifier show that the hardware version consumed only 0.03% of the energy used by the same algorithm in software, even though the hardware version operates with a throughput that is 15 times that of the software version.