On the Privacy Risks of Publishing Anonymized IP Network Traces

Abstract

Abstract. Networking researchers and engineers rely on network packet traces for understanding network behavior, developing models, and evaluating network performance. Although the bulk of published packet traces implement a form of address anonymization to hide sensitive information, it has been unclear if such anonymization techniques are sufficient to address the privacy concerns of users and organizations.In this paper we attempt to quantify the risks of publishing anonymized packet traces. In particular, we examine whether statistical identification techniques can be used to uncover the identities of users and their surfing activities from anonymized packet traces. Our results show that such techniques can be used by any Web server that is itself present in the packet trace and has sufficient resources to map out and keep track of the content of popular Web sites to obtain information on the network-wide browsing behavior of its clients. Furthermore, we discuss how scan sequences identified in the trace can easily reveal the mapping from anonymized to real IP addresses.

Browser Fingerprinting from Coarse Traffic Summaries: Techniques and Implications

“…Several works have examined the susceptibility of anonymized traffic traces to deanonymization, e.g., [16,17,18]. Similar to our work, these approaches re-identify hosts or websites on the basis of their behaviors as exhibited in the anonymized traffic traces.…”

“…Our choice of 10 seconds is based on empirical evaluations. The use of the flow payload for parsing web retrievals can be replaced, for example, by checking for a certain amount of idle time before a burst of web traffic [16], though we do not explore this alternative here. Incomplete retrievals, or those with less than three flows, do not carry enough information about the browser implementation in order for the classifier to make a well-grounded decision, and so we only consider retrievals with more than three flows in our analysis.…”

A taxonomy and adversarial model for attacks against network log anonymization

“…While very useful, these tools alone do not solve the entire problem. One must know how to create anonymization policies that provide the necessary level of assurance-especially in lieu of many recently developed attacks against anonymization [8,4,5,18]. These tools alone do nothing if one does not know how to use them intelligently.…”

Privacy-preserving anomaly detection across multi-domain networks

“…Each ISP network has its own management and measurement infrastructure and collect traffic trace by itself. They seldom share traffic information to their partners since most of information are extremely confidential and sensitive [9] [10]. Revealing such information can make an ISP subject to directed attacks or reveal shortcomings of their own deployments to competitors [11].…”