Gabriel Maciá-Fernández

Network anomaly detection is a very relevant research area nowadays, especially due to its multiple applications in the field of network security. The boost of new models based on variational autoencoders and generative adversarial networks has motivated a reevaluation of traditional techniques for anomaly detection. It is, however, essential to be able to understand these new models from the perspective of the experience attained from years of evaluating network security data for anomaly detection. In this paper, we revisit anomaly detection techniques based on PCA from a probabilistic generative model point of view, and contribute a mathematical model that relates them. Specifically, we start with the probabilistic PCA model and explain its connection to the Multivariate Statistical Network Monitoring (MSNM) framework. MSNM was recently successfully proposed as a means of incorporating industrial process anomaly detection experience into the field of networking. We have evaluated the mathematical model using two different datasets. The first, a synthetic dataset created to better understand the analysis proposed, and the second, UGR'16, is a specifically designed real-traffic dataset for network security anomaly detection. We have drawn conclusions that we consider to be useful when applying generative models to network security detection.

Network anomaly detection is a very relevant research area nowadays, especially due to its multiple applications in the field of network security. The boost of new models based on variational autoencoders and generative adversarial networks has motivated a reevaluation of traditional techniques for anomaly detection. It is, however, essential to be able to understand these new models from the perspective of the experience attained from years of evaluating network security data for anomaly detection. In this paper, we revisit anomaly detection techniques based on PCA from a probabilistic generative model point of view, and contribute a mathematical model that relates them. Specifically, we start with the probabilistic PCA model and explain its connection to the Multivariate Statistical Network Monitoring (MSNM) framework. MSNM was recently successfully proposed as a means of incorporating industrial process anomaly detection experience into the field of networking. We have evaluated the mathematical model using two different datasets. The first, a synthetic dataset created to better understand the analysis proposed, and the second, UGR’16, is a specifically designed real-traffic dataset for network security anomaly detection. We have drawn conclusions that we consider to be useful when applying generative models to network security detection.

The IoT-as-a-Service (IoTaaS) business model has already been identified by some people from both industry and academia, but has not been formally defined. IoTaaS offers IoT devices on demand, with considerable cost savings and resource optimization. In addition, it enables different applications to reuse the existing devices. However, this business model is associated with different technological challenges that need to be addressed, one of which is the identity problem. Focusing on this, self-sovereign identity (SSI) schemes have proven to provide better privacy and scalability than traditional identity paradigms, which is especially important in the IoT owing to its characteristics. In this paper, we formally analyze an IoTaaS business model, identifying and detailing its main technological challenges. In addition, we tackle the identity problem of this business model and propose an SSI-based identity management system, which is compliant with the existing standards from the W3C, and include a performance evaluation.

In this paper, we introduce a new methodology for network access control for Android devices based on app risk assessment. Named ARANAC (which stands for Application Risk Assessment based Network Access Control), this methodology is specially tailored for scenarios using the Bring-Your-Own-Device (BYOD) policy, where the adoption of some solutions can lead to problems in security and privacy for both the employees and the business organization. ARANAC mainly relies on the analysis of an aggregate of permissions declared in the manifests of installed applications on users' devices. The access control scheme combines three operational modules: i) a device monitoring tool, ii) a novel permission-based risk model, and iii) an anomaly-based detection machine learning module based on a methodology (called MSNM, from Multivariate Statistical Network Monitoring) that provides both detection and diagnostic capabilities. ARANAC's novelty is in the combination of four features. Firstly, it is privacy-aware, and thus, it does not require detailed information about installed applications but only an aggregate of permissions. Secondly, it builds a normality model by combining expert knowledge with data, capturing the behavior of a complete population of mobile devices. Thirdly, it is dynamic, as permissions are updated in real time, allowing the network to re-assess access control on a continuous basis. Finally, its diagnostic capabilities allow for giving recommendations to final users so that they are capable of mitigating their risks when accessing networks. We evaluated the approach with more than 80 Android devices at a university campus network and obtained interesting results regarding security risks in the usual deployment of device apps.

Network Security Monitoring (NSM) is a popular term to refer to the detection of security incidents by monitoring the network events. An NSM system is central for the security of current networks, given the escalation in sophistication of cyberwarfare. In this paper, we review the state-of-the-art in NSM, and derive a new taxonomy of the functionalities and modules in an NSM system. This taxonomy is useful to assess current NSM deployments and tools for both researchers and practitioners. We organize a list of popular tools according to this new taxonomy, and identify challenges in the application of NSM in modern network deployments, like Software Defined Network (SDN) and Internet of Things (IoT).