INFOCOM 2018 Papers

The typical aim of User Identity Linkage (UIL) is to detect when users from across different social platforms are actually one and the same individual. Existing efforts to address this problem of practical relevance span from user-profile-based, through user-generated-content-based, user-behavior-based approaches to supervised or unsupervised learning frameworks, to subspace learning-based models. Most of them often require extraction of relevant features (e.g., profile, location, biography, networks, behavior, etc.) to model the user consistently across different social networks. However, these features are mainly derived based on prior knowledge and may vary for different platforms and applications. Inspired by the recent successes of deep learning in different tasks, especially in automatic feature extraction and representation, we propose a deep neural network based algorithm for UIL, called DeepLink. It is a novel end-to-end approach in a semi-supervised learning manner, without involving any hand-crafting features. Specifically, DeepLink samples the networks and learns to encode network nodes into vector representation to capture local and global network structures which, in turn, can be used to align anchor nodes through deep neural networks. A dual learning based paradigm is exploited to learn how to transfer knowledge and update the linkage using the policy gradient method. Experiments conducted on several public datasets show that DeepLink outperforms the state-of-the-art methods in terms of both linking precision and identity-match ranking.

Poor Wi-Fi quality can disrupt home users' internet experience, or the Quality of Experience (QoE). Detecting when Wi-Fi degrades QoE is extremely valuable for residential Internet Service Providers (ISPs) as home users often hold the ISP responsible whenever QoE degrades. Yet, ISPs have little visibility within the home to assist users. Our goal is to develop a system that runs on commodity access points (APs) to assist ISPs in detecting when Wi-Fi degrades QoE. Our first contribution is to develop a method to detect instances of poor QoE based on the passive observation of Wi-Fi quality metrics available in commodity APs (e.g., PHY rate). We use support vector regression to build predictors of QoE given Wi-Fi quality for popular internet applications. We then use K-means clustering to combine per-application predictors to identify regions of Wi-Fi quality where QoE is poor across applications. We call samples in these regions as poor QoE samples. Our second contribution is to apply our predictors to Wi-Fi metrics collected over one month from 3479 APs of customers of a large residential ISP. Our results show that QoE is good most of the time, still we find 11.6% of poor QoE samples. Worse, approximately 21% of stations have more than 25% poor QoE samples. In some cases, we estimate that Wi-Fi quality causes poor QoE for many hours, though in most cases poor QoE events are short.

By not requiring expensive licenses, unlicensed spectrum lowers the barriers for firms to offer wireless services. However, incumbent firms may still try to erect other entry barriers. For example, recent work has highlighted how customer contracts may be used as one such barrier by penalizing customers for switching to a new entrant. However, this work did not account for another potential benefit of unlicensed spectrum, having access to this open resource may incentivize entrants to invest in new and potentially better technology. This paper studies the interaction of contracts and the incentives of firms to invest in developing new technology. We use a game theoretic model to study this and characterize the effect of contracts on economic welfare. The role of subsidies or taxes by a social planner is also considered.

Conventional spinning inspection systems, equipped with separated sensors (e.g., accelerometer, laser, etc.) and communication modules, are either very expensive and/or suffering from occlusion and narrow field of view. The recently proposed RFID-based sensing solution draws much attention due to its intriguing features, such as being cost-effective, applicable to occluded objects and auto-identification, etc. However, this solution only works in quiet settings where the reader and spinning object remain absolutely stationary, as their shaking would ruin the periodicity and sparsity of the spinning signal, making it impossible to be recovered. This work introduces Tagtwins, a robust spinning sensing system that can work in noisy settings. It addresses the challenge by attaching dual RFID tags on the spinning surface and developing a new formulation of spinning signal that is shaking-resilient, even if the shaking involves unknown trajectories. Our main contribution lies in two newly developed techniques, relative spinning signal and dual compressive reading. We analytically demonstrate that our solution can work in various settings. We have implemented Tagtwins with COTS RFID devices and evaluated it extensively. Experimental results show that Tagtwins can inspect the rotation frequency with high accuracy and robustness.

Software-Defined Networking (SDN), which separates the control plane and data plane, is a promising new network architecture for the Future Internet. OpenFlow is the de facto standard which defines the communication protocol between the controller and switches. The most challenging issue in OpenFlow switches is the lookup of multiple OpenFlow tables. The lookup of OpenFlow tables is so complicated that the state-of-the-art research are still focusing on the design of lookup pipeline architecture, and there is no specific algorithm for the lookup of OpenFlow tables. In this paper, we revise the long-pipeline architecture of OpenFlow 1.4 to a 5-stage pipeline architecture to make a trade-off between flexibility and implementability, and decompose the lookup of OpenFlow tables into three kinds of lookup: longest prefix matching (IP lookup), multi-field matching (packet classification), and exact matching. Then we design new algorithms for packet classification, because the state-of-the-art solutions for them seldom support fast update which is highly demanding for OpenFlow. The other two kinds of lookups can be well handled by state-of-the-art. Experimental results show that our proposed algorithms work excellently, and outperform state-of-the-art solutions.

We consider a wireless distributed computing network, where all computing nodes (workers) are connected via wireless medium obeying the seminal protocol channel model. In particular, we focus on the MapReduce-type platform, where each worker is assigned to compute some arbitrary output functions from F input files, which are distributively cached in all workers. The overall computation is decomposed into computing a set of “Map” and “Reduce” functions across all workers. The goal is to characterize the minimum computing latency as a function of the computation load. Unlike other related works, which consider either wireline settings or restrict the communication among workers to single-hop, here we focus on the wireless scenario and do not constrain any communication schemes. We propose a data set cache strategy based on a deterministic assignment of Maximum Distance Separable (MDS)-coded date sets over all input files, and a coded multicast transmission strategy where the workers send linearly coded computing results to each other in order to collectively satisfy their assigned tasks. We show that our approach can achieve a scalable communication latency, outperform the state of the art schemes in the order sense, and achieve the information theoretic outer bound within a multiplicative constant factor in practical parameter regimes.

Wireless RF energy transfer for indoor sensors is an emerging paradigm that ensures continuous operation without battery limitations. However, high power radiation within the ISM band interferes with the packet reception for existing WiFi devices. The paper proposes the first effort in merging the RF energy transfer functions within a standards compliant 802.11 protocol to realize practical and WiFi-friendly Energy Delivery (WiFED). The WiFED architecture is composed of a centralized controller that coordinates the actions of multiple distributed energy transmitters (ETs), and a number of deployed sensors that periodically request energy from the ETs. The paper first describes the specific 802.11 supported protocol features that can be exploited by sensors to request energy and for the ETs to participate in the energy delivery process. Second, it devises a controller-driven bipartite matching-based algorithmic solution that assigns the appropriate number of ETs to energy requesting sensors for an efficient energy transfer process. The proposed in-band and protocol supported coexistence in WiFED is validated via simulations and partly in a software defined radio testbed, showing 15% improvement in network lifetime and 31% reduction in the charging delay compared to the classical nearest distance-based charging schemes that do not anticipate future energy needs of the sensors and are not designed to co-exist with WiFi systems.

Least-recently-used (LRU) caching systems have been widely used, and are increasingly deployed driven by emerging trends for big data. In a typical scenario, these systems are used to serve multiple flows of dependent data item requests that are also correlated over time. These flows compete for the limited cache space. Characterizing the miss ratios of these competing flows can facilitate the design and improve the system performance. The existing asymptotic analyses for correlated requests give explicit results for Zipf's distributions with the index greater than a critical value (one). Consequently, the asymptotic result is inaccurate around this critical point, which notably is also the typical parameter region reported by many empirical measurements. In contrast, we derive the asymptotic miss ratios of multiple flows for a large class of truncated heavy-tailed data item popularity distributions with time dependency. Importantly, it significantly improves the accuracy in numerical computations when the index of a Zipf's distribution is close to one. Moreover, the result generalizes beyond Zipf's distributions, e.g., to Weibull, for multiple flows of correlated data item requests. Our asymptotic result directly exploits the critical properties of the distribution and the truncated support region. As our versatile expression is explicit, it avoids the numerical computations required by the characteristic time approximation. Interestingly, it also validates the characteristic time approximation with new forms for multiple flows of competing requests that are correlated over time under certain conditions.

We study a new problem, refined localization, in this paper. Refined localization calculates the location of an object in high precision, given that the object is in a relatively small region such as the surface of a table. Refined localization is useful in many cyber-physical systems such as industrial autonomous robots. Existing vision-based approaches suffer from several disadvantages, including good lighting conditions, line of sight, pre-learning process, and high computation overhead. Also vision-based approaches cannot differentiate objects with similar colors and shapes. This paper presents a new refined localization system, called Trio, which uses passive Radio Frequency Identification (RFID) tags for low cost and easy deployment. Trio provides a new angle to utilize RF interference for tag localization by modeling the equivalent circuits of coupled tags. We implement our prototype using commercial off-the-shelf RFID reader and tags. Extensive experiment results demonstrate that Trio effectively achieves high accuracy of refined localization, i.e., <; 1 cm errors for several types of main stream tags.

Detecting anomalous traffic is a crucial task of managing networks. Many anomaly detection algorithms have been proposed recently. However, constrained by their matrix-based traffic data model, existing algorithms often suffer from low detection accuracy. To fully utilize the multi-dimensional information hidden in the traffic data, this paper takes an initiative to investigate the potential and methodologies of performing tensor factorization for more accurate Internet anomaly detection. Only considering the low-rank linearity features hidden in the data, current tensor factorization techniques would result in low anomaly detection accuracy. We propose a novel Graph-based Tensor Recovery model (Graph-TR) to well explore both low rank linearity features as well as the non-linear proximity information hidden in the traffic data for better anomaly detection. We encode the non-linear proximity information of the traffic data by constructing nearest neighbor graphs and incorporate this information into the tensor factorization using the graph Laplacian. Moreover, to facilitate the quick building of neighbor graph, we propose a nearest neighbor searching algorithm with the simple locality-sensitive hashing (LSH). We have conducted extensive experiments using Internet traffic trace data Abilene and GEANT. Compared with the state of art algorithms on matrix-based anomaly detection and tensor recovery approach, our Graph-Trcan achieve significantly lower False Positive Rate and higher True Positive Rate.

Vehicular ad hoc networks (VANETs) have a potential to promote vehicular telematics and infotainment applications, where a key and challenging issue is the design of robust and efficient vehicular content transmissions to combat the lossy inter-vehicle links. In this paper, we focus on the robust optimization of content transmissions over cooperative VANETs. We first derive a stochastic model for estimation of time-varying inter-vehicle distance, which is dependent of the vehicle real-time kinematics and the distribution of the initial space headway. With this model, we analytically formulate the transient inter-vehicle connectivity assuming Nakagami fading channels for the physical (PHY) layer. We also model the contention nature of the medium access control (MAC) layer, on which we are based to evaluate the throughput achieved by each vehicle equipped with dedicated short-range communication (DSRC). Combining these models, we derive a closed-formed expression for the upper bound of the probability of failure in intact-content transmissions. Based upon this theoretical bound, we develop a robust optimization model for assigning content data traffic among different cooperative transmission paths, where the objective is to minimize the maximum likelihood of unsuccessful content transmissions over the cooperative VANET. We mathematically transform the optimization model to another equivalent form, such that it can be practically deployed. Finally, we validate our theoretical development with extensive simulations. Numerical results are also provided to confirm the power of cooperation in boosting the VANET performance as well as demonstrate the advantage of the proposed robust optimization in terms of content data reception reliability.

With the proliferation of mobile devices and various sensors (e.g., GPS, magnetometer, accelerometers, gyroscopes) equipped, richer services, e.g. location based services, are provided to users. A series of methods have been proposed to protect the users' privacy, especially the trajectory privacy. Hardware fingerprinting has been demonstrated to be a surprising and effective source for identifying/authenticating devices. In this work, we show that a few data samples collected from the motion sensors are enough to uniquely identify the source mobile device, i.e., the raw motion sensor data serves as a fingerprint of the mobile device. Specifically, we first analytically understand the fingerprinting capacity using features extracted from hardware data. To capture the essential device feature automatically, we design a multi-LSTM neural network to fingerprint mobile device sensor in real-life uses, instead of using handcrafted features by existing work. Using data collected over 6 months, for arbitrary user movements, our fingerprinting model achieves 93 % F -score given one second data, while the state-of-the-art work achieves 79% F-score. Given ten seconds randomly sampled data, our model can achieve 98.8% accuracy. We also propose a novel generative model to modify the original sensor data and yield anonymized data with little fingerprint information while retain good data utility.

LTE in Unlicensed (LTE-U) constitutes a new source of interference in the 5 GHz ISM band with a potentially strong impact on WiFi performance. Cross-technology interference and radio resource management are the best ways to assure efficient coexistence but require proper signaling channels. We present LtFi, a system which enables to set-up a cross-technology communication between nodes of co-located LTE-U and WiFi networks. LtFi follows a two-step approach: using an innovative side channel on their air-interface LTE-U BSs are broadcasting connection and identification information to adjacent WiFi nodes, which is used in a subsequent step to create a bi-directional control channel over the wired backhaul. The simple LtFi is fully compliant with LTE-U and works with COTS WiFi hardware. The achievable data rate on the air-interface based broadcast side channel (up to 665 bps) is sufficient for this and multiple other purposes. Experimental evaluation of a fully operational prototype has demonstrated reliable data transmission even in crowded wireless environments for LTE-U receive power levels down to -92 dBm. Moreover, system-level simulations demonstrate accurate recognition of the complete set of interfering LTE-U BSs in a typical LTE-U multi-cell environment.

Modern mobile devices are equipped with multicore-based processors, which introduce new challenges on computation offloading. With the big.LITTLE architecture, instead of only deciding locally or remotely running a task in the traditional architecture, we have to consider how to exploit the new architecture to minimize energy while satisfying application completion time constraints. In this paper, we address the problem of energy-efficient computation offloading on multicore-based mobile devices running multiple applications. We first formalize the problem as a mixed-integer nonlinear programming problem that is NP-hard, and then propose a novel heuristic algorithm to jointly solve the offloading decision and task scheduling problems. The basic idea is to prioritize tasks from different applications to make sure that both application time constraints and task-dependency requirements are satisfied. To find a better schedule while reducing the schedule searching overhead, we propose a critical path based solution which recursively checks the tasks and moves tasks to the right CPU cores to save energy. Simulation and experimental results show that our offloading algorithm can significantly reduce the energy consumption of mobile devices while satisfying the application completion time constraints.

Better understanding of mobile applications' behaviors would lead to better malware detection/classification and better app recommendation for users. In this work, we design a framework AppDNA to automatically generate a compact representation for each app to comprehensively profile its behaviors. The behavior difference between two apps can be measured by the distance between their representations. As a result, the versatile representation can be generated once for each app, and then be used for a wide variety of objectives, including malware detection, app categorizing, plagiarism detection, etc. Based on a systematic and deep understanding of an app's behavior, we propose to perform a function-call-graph-based app profiling. We carefully design a graph-encoding method to convert a typically extremely large call-graph to a 64-dimension fix-size vector to achieve robust app profiling. Our extensive evaluations based on 86,332 benign and malicious apps demonstrate that our system performs app profiling (thus malware detection, classification, and app recommendation) to a high accuracy with extremely low computation cost: it classifies 4024 (benign/malware) apps using around 5.06 second with accuracy about 93.07%; it classifies 570 malware's family (total 21 families) using around 0.83 second with accuracy 82.3%; it classifies 9,730 apps' functionality with accuracy 33.3% for a total of 7 categories and accuracy of 88.1 % for 2 categories.

Communication among neurons is the highly evolved and efficient nanoscale communication paradigm, hence the most promising technique for biocompatible nanonetworks. This necessitates the understanding of neuro-spike communication from information theoretical perspective to reach a reference model for nanonetworks. This would also contribute towards developing ICT-based diagnostics techniques for neuro-degenerative diseases. Thus, in this paper, we focus on the fundamental building block of neuro-spike communication, i.e., signal transmission over a synapse, to evaluate its information transfer rate. We aim to analyze a realistic synaptic communication model, which for the first time, encompasses the variation in vesicle release probability with time, synaptic geometry and the re-uptake of neurotransmitters by pre-synaptic terminal. To achieve this objective, we formulate the mutual information between input and output of the synapse. Then, since this communication paradigm has memory, we evaluate the average mutual information over multiple transmissions to find its overall capacity. We derive a closed-form expression for the capacity of the synaptic communication as well as calculate the capacity-achieving input probability distribution. Finally, we find the effects of variation in different synaptic parameters on the information capacity and prove that the diffusion process does not decrease the information a neural response carries about the stimulus in real scenario.

A switching/forwarding fabric with high-bandwidth one-to-one and low-bandwidth many-to-many data forwarding can be achieved by combing electronic packet switches (EPS) and optical circuit switches (OCS). This hybrid solution scales well but it is not suitable for cloud computing/datacenter applications, which typically rely on one-to-many and many-to-one communications. Recently, composite-path switching (cp-switching), which adds paths between EPS and ocs is introduced to deal with this skewed traffic pattern. The state-of-the-art scheduling algorithm for CP-switches reduces, by heuristics, a CP-switching problem with one composite path to a hybrid switching problem without the composite path, and leverages existing scheduling techniques to tackle the latter problem. Unfortunately, the approach provides neither performance guarantee nor the support for multiple composite paths. In this paper, we systematically study the shortest time CP-switch scheduling problem with multiple composite paths and show that the problem can be expressed as an optimization problem with sparsity constraints. An LP-based algorithm is derived accordingly which supports multiple composite paths and more importantly, it provides performance guarantee. Simulations demonstrate that adding more composite paths can help shorten the schedule, and our approach outperforms existing methods by 30% to 70%.

Scalable, fine-grained access control for Internet-of-Things is needed in enterprise environments, where thousands of subjects need to access possibly one to two orders of magnitude more objects. Existing solutions offer all-or-nothing access, or require all access to go through a cloud backend, greatly impeding access granularity, robustness and scale. In this paper, we propose Heracles, an IoT access control system that achieves robust, fine-grained access control at enterprise scale. Heracles adopts a capability-based approach using secure, unforgeable tokens that describe the authorizations of subjects, to either individual or collections of objects in single or bulk operations. It has a 3-tier architecture to provide centralized policy and distributed execution desired in enterprise environments, and delegated operations for responsiveness of resource-constrained objects. Extensive security analysis and performance evaluation on a testbed prove that Heracles achieves robust, responsive, fine-Qrained access control in large scale enterprise environments.

Network function virtualization (NFV) decouples network functions from the dedicated hardware and enables them running on commodity servers, facilitating widespread deployment of virtualized network functions (VNFs). Network operators tend to deploy VNFs in virtual machines (VMs) due to VM's ease of duplication and migration, which enables flexible VNF placement and scheduling. Efforts have been paid to provide efficient VNF placement approaches, aiming at minimizing the resource cost of VNF deployment and reducing the latency of service chain. However, existing placement approaches may result in hardware resource competition of co-located VNFs, leading to performance degradation. In this paper, we present a measurement study on the performance interference among different types of co-located VNFs and analyze how VNFs' competitive hardware resources and the characteristics of packet affect the performance interference. We disclose that the performance interference between co-located VNFs is ubiquitous, which causes the performance degradation, in terms of VNFs' throughput, ranging from 12.36% to 50.3%, and the competition of network I/O bandwidth plays a key role in the performance interference. Based on our measurement results, we give some advices on how to design more efficient VNF placement approaches.