Privacy of encrypted voice-over-IP

Abstract

In this paper, we present an early study on how timing-based traffic analysis attacks can be used to reconstruct the communication on end-to-end VOIP systems by taking advantage of the reduction or suppression of the generation of traffic whenever the sender detects a voice inactivity period. We describe a simple Bayesian classifier to identify simple voice signals from the pattern of packet timings. We then proceed to incorporate context awareness by using a Hidden Markov model. Experiments with very simple symbols show that the effectiveness to reconstruct the voice signal depends significantly on the quality of collected silence suppression information. We conclude by identifying a number of problems that need to be further studied in order to effectively assess the danger of silence suppression based attacks on VOIP systems.

On the Privacy of Encrypted Skype Communications

“…Furthermore, by using the DTW algorithm, the drawbacks of an approach involving HMMs, e.g., an extensive amount of training data, can be avoided. Also, the capacity of DTW to deal with time or speed variability along with the use of packet sizes as discriminant features renders the proposed approach stronger than the one described in [4]. Finally, because Skype is closed source and its audio codec is not widely published, the approach proposed in [5] and [6] could not have been applied.…”

“…The efficiency of the approach presented in [4] is strongly dependent on the quality of the network. Furthermore, sentences of similar lengths cannot be reliably identified.…”

“…In their paper Privacy of Encrypted Voice-over-IP that focuses on Google Talk, Lella and Bettati [4] assume that the packet sizes do not carry any information but that all the information is contained in the inter-arrival times. They use the silence phases, which are encoded in smaller packets and with bigger inter-arrival times, to isolate words.…”

Analysis of information leakage from encrypted Skype conversations

“…The efficiency of the approach presented in [10] is strongly dependent on the quality of the network. Furthermore, sentences of similar lengths cannot be reliably identified.…”

“…In their paper Privacy of Encrypted Voice over IP that focuses on Google Talk, Lella and Bettati [10] assume that the packet sizes do not carry any information but that all the information is contained in the inter-arrival times. They use the silence phases, which are encoded in smaller packets and with bigger inter-arrival times, to isolate words.…”