The Internet of Things (IoT) is being hailed as the next wave revolutionizing our society, and smart homes, enterprises, and cities are increasingly being equipped with a plethora of IoT devices. Yet, operators of such smart environments may not even be fully aware of their IoT assets, let alone whether each IoT device is functioning properly safe from cyber-attacks. In this paper, we address this challenge by developing a robust framework for IoT device classification using traffic characteristics obtained at the network level. Our contributions are fourfold. First, we instrument a smart environment with 28 different IoT devices spanning cameras, lights, plugs, motion sensors, appliances, and health-monitors. We collect and synthesize traffic traces from this infrastructure for a period of six months, a subset of which we release as open data for the community to use. Second, we present insights into the underlying network traffic characteristics using statistical attributes such as activity cycles, port numbers, signalling patterns, and cipher suites. Third, we develop a multi-stage machine learning based classification algorithm and demonstrate its ability to identify specific IoT devices with over 99 percent accuracy based on their network activity. Finally, we discuss the trade-offs between cost, speed, and performance involved in deploying the classification framework in real-time. Our study paves the way for operators of smart environments to monitor their IoT assets for presence, functionality, and cyber-security without requiring any specialized devices or protocols.

Classifying IoT Devices in Smart Environments Using Network Traffic Characteristics

The exponential traffic growth, demand for high speed wireless data communications, as well as incessant deployment of innovative wireless technologies, services, and applications, have put considerable pressure on the mobile network operators (MNOs). Consequently, cellular access network performance in terms of capacity, quality of service, and network coverage needs further considerations. In order to address the challenges, MNOs, as well as equipment vendors, have given significant attention to the small-cell schemes based on cloud radio access network (C-RAN). This is due to its beneficial features in terms of performance optimization, cost-effectiveness, easier infrastructure deployment, and network management. Nevertheless, the C-RAN architecture imposes stringent requirements on the fronthaul link for seamless connectivity. Digital radio over fiber-based common public radio interface (CPRI) is the fundamental means of distributing baseband samples in the C-RAN fronthaul. However, optical links which are based on CPRI have bandwidth and flexibility limitations. Therefore, these limitations might constrain or make them impractical for the next generation mobile systems which are envisaged not only to support carrier aggregation and multi-band but also envisioned to integrate technologies like millimeter-wave (mm-wave) and massive multiple-input multiple-output antennas into the base stations. In this paper, we present comprehensive tutorial on technologies, requirements, architectures, challenges, and proffer potential solutions on means of achieving an efficient C-RAN optical fronthaul for the next-generation network such as the fifth generation network and beyond. A number of viable fronthauling technologies such as mm-wave and wireless fidelity are considered and this paper mainly focuses on optical technologies such as optical fiber and free-space optical. We also present feasible means of reducing the system complexity, cost, bandwidth requirement, and latency in the fronthaul. Furthermore, means of achieving the goal of green communication networks through reduction in the power consumption by the system are considered.

Toward an Efficient C-RAN Optical Fronthaul for the Future Networks: A Tutorial on Technologies, Requirements, Challenges, and Solutions

Programmable packet processing is increasingly implemented using kernel bypass techniques, where a userspace application takes complete control of the networking hardware to avoid expensive context switches between kernel and userspace. However, as the operating system is bypassed, so are its application isolation and security mechanisms; and well-tested configuration, deployment and management tools cease to function.To overcome this limitation, we present the design of a novel approach to programmable packet processing, called the eXpress Data Path (XDP). In XDP, the operating system kernel itself provides a safe execution environment for custom packet processing applications, executed in device driver context. XDP is part of the mainline Linux kernel and provides a fully integrated solution working in concert with the kernel's networking stack. Applications are written in higher level languages such as C and compiled into custom byte code which the kernel statically analyses for safety, and translates into native instructions.We show that XDP achieves single-core packet processing performance as high as 24 million packets per second, and illustrate the flexibility of the programming model through three example use cases: layer-3 routing, inline DDoS protection and layer-4 load balancing.

https://dl.acm.org/doi/pdf/10.1145/3281411.3281443

The eXpress data path: fast programmable packet processing in the operating system kernel

Mental health monitoring with multimodal sensing and machine learning: A survey

Human behavior is the focus of many studies in the social, health, and behavioral sciences. Yet, few studies use behavioral observation methods to collect objective measures of behavior as it occurs in daily life, out in the real world — presumably the context of ultimate interest. Here, we provide a review of recent studies focused on measuring human behavior using smartphones and their embedded mobile sensors. To draw attention to current advances in the field of smartphone sensing, we describe the daily behaviors captured using these methods, which include movement behaviors (physical activity, mobility patterns), social behaviors (face-to-face encounters, computer-mediated communications), and other daily activities (non-mediated and mediated activities). We conclude by pointing to promising areas of future research for studies using Smartphone Sensing Methods (SSMs) in the behavioral sciences.

https://www.repository.cam.ac.uk/bitstream/1810/274981/3/Attachment%20%281%29.pdf

Smartphone sensing methods for studying behavior in everyday life

Explosion of number of smartphone apps and their diversity has created a fertile ground to study behaviour of smartphone users. Patterns of app usage, specifically types of apps and their duration are influenced by the state of the user and this information can be correlated with the self-reported state of the users. The work in this paper is along the line of understanding patterns of app usage and investigating relationship of these patterns with the perceived stress level within the workplace context. Our results show that using a subject-centric behaviour model we can predict stress levels based on smartphone app usage. The results we have achieved, of average accuracy of 75% and precision of 85.7%, can be used as an indicator of overall stress levels in work environments and in turn inform stress-reduction organisational policies, especially when considering interrelation between stress and productivity of workers.

/pdf/smartphone-app-usage-as-a-predictor-of-perceived-stress-2yc79nbjfr.pdf

Smartphone app usage as a predictor of perceived stress levels at workplace

Document Clustering is a widely studied problem in Text Categorization. It is the process of partitioning or grouping a given set of documents into disjoint clusters where documents in the same cluster are similar. K-means, one of the simplest unsupervised learning algorithms, solves the well known clustering problem following a simple and easy way to classify a given data set through a certain number of clusters (assume k clusters) fixed a priori. The main idea is to define k centroids, one for each cluster. This clustering algorithm uses an iterative procedure which converges to one of numerous local minima. We have found that these iterative techniques are especially sensitive to initial starting conditions of the centroid of each cluster and the more the distance among the cluster centroid the better the clustering performance. In simple K-means algorithm the way to initialize the centroid is not specified and one popular way to start is to randomly choose k points of the samples as k centroids but this process does not guarantee to choose the maximum dissimilar documents as the centroid point for k-cluster. In this paper we proposed a modified k-means algorithm which uses Jaccard distance measure for computing the most dissimilar k documents as centroids for k clusters. Our experimental results demonstrate that our proposed K-means algorithm with Jaccard distance measure for computing the centroid improves the clustering performance of the simple K-means algorithm.

An efficient k-means algorithm integrated with Jaccard distance measure for document clustering

Despite the considerable attention that the ICN paradigm received so far, its deployment has been hindered by the scale of upgrades required to the existing infrastructure. Software programmable networking frameworks would constitute a remarkable opportunity for ICN as they enable fast deployment of novel technologies on commodity hardware. However, a software ICN router implementation for commodity platforms guaranteeing adequate packet processing performance is not available yet. This paper introduces Augustus, a software architecture for ICN routers, and detail two implementations, stand-alone and modular, released as open-source code. We deployed both implementations on a state-of-the-art hardware platform and analyzed their performance under different configurations. Our analysis shows that with both implementations it is possible to achieve a throughput of approximately 10 Mpps, saturating 10 Gbit/s links with packet as small as 100 bytes. However, to achieve such performance, routers must be carefully configured to fully exploit the capabilities of the hardware platforms they run on.

Augustus: a CCN router for programmable networks

The integration of both fronthaul and backhaul into a single transport network (namely, 5G-Crosshaul) is envisioned for the future 5G transport networks. This requires a fully integrated and unified management of the fronthaul and backhaul resources in a cost-efficient, scalable and flexible way through the deployment of an SDN/NFV control framework. This paper presents the designed 5G-Crosshaul architecture, two selected SDN/NFV applications targeting for cost-efficient resource and energy usage: the Resource Management Application (RMA) and the Energy Management and Monitoring Application (EMMA). The former manages 5G-Crosshaul resources (network, computing and storage resources). The latter is a special version of RMA with the focus on the objectives of optimizing the energy consumption and minimizing the energy footprint of the 5G-Crosshaul infrastructure. Besides, EMMA is applied to the mmWave mesh network and the high speed train scenarios. In particular, we present the key application design with their main components and the interactions with each other and with the control plane, and then we present the proposed application optimization algorithms along with initial results. The first results demonstrate that the proposed RMA is able to cost-efficiently utilize the Crosshaul resources of heterogeneous technologies, while EMMA can achieve significant energy savings through energy-efficient routing of traffic flows. For experiments in real system, we also set up Proof of Concepts (PoCs) for both applications in order to perform real trials in the field.

/pdf/novel-resource-and-energy-management-for-5g-integrated-52qw3a28da.pdf

Novel Resource and Energy Management for 5G integrated backhaul/fronthaul (5G-Crosshaul)

Voice and multimedia communications are rapidly migrating from traditional networks to TCP/IP networks (Internet), where services are provisioned by SIP (Session Initiation Protocol). This paper proposes an on-line filter that examines the stream of incoming SIP messages and classifies them as good or bad. The classification is carried out in two stages: first a lexical analysis is performed to weed out those messages that do not belong to the language generated by the grammar defined by the SIP standard. After this first stage, a second filtering occurs which identifies messages that somehow differ - in structure or contents - from messages that were previously classified as good. While the first filter stage is straightforward, as the classification is crisp (either a messages belongs to the language or it does not), the second stage requires a more delicate handling, as it is not a sharp decision whether a message is semantically meaningful or not. The approach we followed for this step is based on using past experience on previously classified messages, i.e. a "learn-by-example" approach, which led to a classifier based on Support-Vector-Machines (SVM) to perform the required analysis of each incoming SIP message. The paper describes the overall architecture of the two-stage filter and then explores several points of the configuration-space for the SVM to determine a good configuration setting that will perform well when used to classify a large sample of SIP messages obtained from real traffic collected on a VoIP installation at our institution. Finally, the performance of the classification on additional messages collected from the same source is presented.

Raihana Ferdous

Papers

Smartphone app usage as a predictor of perceived stress levels at workplace

An efficient k-means algorithm integrated with Jaccard distance measure for document clustering

Augustus: a CCN router for programmable networks

Novel Resource and Energy Management for 5G integrated backhaul/fronthaul (5G-Crosshaul)

On the Use of SVMs to Detect Anomalies in a Stream of SIP Messages