In Long-Term Evolution (LTE), the security threat posed by malicious users can be reduced by using the traffic detection function (TDF) in the policy and charging control (PCC) system to inspect the packets passing through the Packet Data Network Gateway. However, as the volume of global cellular data traffic continues to grow, it becomes impossible for the TDF to inspect all of the packets. Consequently, more efficient Intrusion Detection Systems (IDSs) are required. Accordingly, the present study proposes a random packet inspection scheme in which the inspection rate is dynamically adjusted based on the perceived intrusion period of the session. An analytical model is proposed to evaluate the effect of the inspection rate on the intrusion detection rate, inspection cost, and detection latency. The validity of the proposed model is demonstrated through numerical simulations. It is shown that the model provides an effective means of setting the inspection rate in such a way as to maximize the intrusion detection rate, while simultaneously achieving a satisfactory tradeoff between the inspection cost and the detection latency.
All Science Journal Classification (ASJC) codes