INFOCOM 2015 Papers

Widely deployed mHealth systems enable patients to efficiently collect, aggregate, and report their Personal Health Records (PHRs), and then lower the costs and shorten their response time. The increasing needs of PHR monitoring require the involvement of healthcare companies that provide monitoring programs for analyzing PHRs. Unfortunately, healthcare companies are lack of the computation, storage, and communication capability on supporting millions of patients. To tackle this problem, they seek for the help from the cloud. However, delegating monitoring programs to the cloud may incur serious security and privacy breaches because people have to provide their identity information and PHRs to the public domain. Even worse, the cloud may mistakenly return the incorrect computation results, which will put patients' life in jeopardy. In this paper, we propose a verifiable privacy-preserving monitoring scheme for cloud-assisted mHealth systems. Our scheme allows patients to verify the correctness of computation results from the cloud without revealing their PHRs and identity information. In addition, our advanced schemes offer efficient PHR updates and PHR computations on complex monitoring programs. By detailed performance evaluation, we have shown the security and efficiency of our proposed scheme.

Providing an adequate long-term user participation incentive is important for a participatory sensing system to maintain enough number of active users (sensors), so as to collect a sufficient number of data samples and support a desired level of service quality. In this work, we consider the sensor selection problem in a general time-dependent and location-aware participatory sensing system, taking the long-term user participation incentive into explicit consideration. We study the problem systematically under different information scenarios, regarding both future information and current information (realization). In particular, we propose a Lyapunov-based VCG auction policy for the on-line sensor selection, which converges asymptotically to the optimal off-line benchmark performance, even with no future information and under asymmetry of current information. Extensive numerical results show that our proposed policy outperforms the state-of-art policies in the literature, in terms of both user participation (e.g., reducing the user dropping probability by 25% ~ 90%) and social performance (e.g., increasing the social welfare by 15% ~ 80%).

With the growing awareness of data privacy, more and more cloud users choose to encrypt their sensitive data before outsourcing them to the cloud. Search over encrypted data is therefore a critical function facilitating efficient cloud data access given the high data volume that each user has to handle nowadays. Inverted index is one of the most efficient searchable index structures and has been widely adopted in plaintext search. However, securing an inverted index and its associated search schemes is not a trivial task. A major challenge exposed from the existing efforts is the difficulty to protect user's query privacy. The challenge roots on two facts: 1) the existing solutions use a deterministic trapdoor generation function for queries; and 2) once a keyword is searched, the encrypted inverted list for this keyword is revealed to the cloud server. We denote this second property in the existing solutions as one-time-only search limitation. Additionally, conjunctive multi-keyword search, which is the most common form of query nowadays, is not supported in those works. In this paper, we propose a public-key searchable encryption scheme based on the inverted index. Our scheme preserves the high search efficiency inherited from the inverted index while lifting the one-time-only search limitation of the previous solutions. Our scheme features a probabilistic trapdoor generation algorithm and protects the search pattern. In addition, our scheme supports conjunctive multi-keyword search. Compared with the existing public key based schemes that heavily rely on expensive pairing operations, our scheme is more efficient by using only multiplications and exponentiations. To meet stronger security requirements, we strengthen our scheme with an efficient oblivious transfer protocol that hides the access pattern from the cloud. The simulation results demonstrate that our scheme is suitable for practical usage with moderate overhead.

Deep Packet Inspection (DPI) plays a major role in contemporary networks. Specifically, in datacenters of content providers, the scanned data may be highly repetitive. Most DPI engines are based on identifying signatures in the packet payload. This pattern matching process is expensive both in memory and CPU resources, and thus, often becomes the bottleneck of the entire application. In this paper we show how DPI can be accelerated by leveraging repetitions in the inspected traffic. Our new mechanism makes use of these repetitions to allow the repeated data to be skipped rather than scanned again. The mechanism consists of a slow path, in which frequently repeated strings are identified and stored in a dictionary, along with some succinct information for accelerating the DPI process, and a data path, where the traffic is scanned byte by byte but strings from the dictionary, if encountered, are skipped. Upon skipping, the data path recovers to the state it would have been in had the scanning continued byte by byte. Our solution achieves a significant performance boost, especially when data is from the same content source (e.g., the same website). Our experiments show that for such cases, our solution achieves a throughput gain of 1.25–2.5 times the original throughput, when implemented in software.

Sleep monitoring has drawn increasingly attention as the quality and quantity of the sleep are important to maintain a person's health and well-being. For example, inadequate and irregular sleep are usually associated with serious health problems such as fatigue, depression and cardiovascular disease. Traditional sleep monitoring systems, such as PSG, involve wearable sensors with professional installations, and thus are limited to clinical usage. Recent work in using smartphone sensors for sleep monitoring can detect several events related to sleep, such as body movement, cough and snore. Such coarse-grained sleep monitoring however is unable to detect the breathing rate which is an important vital sign and health indicator. This work presents a fine-grained sleep monitoring system which is capable of detecting the breathing rate by leveraging smartphones. Our system exploits the readily available smartphone earphone placed close to the user to reliably capture the human breathing sound. Given the captured acoustic sound, our system performs noise reduction to remove environmental noise and then identifies the breathing rate based on the signal envelope detection. Our system can further detect detailed sleep events including snore, cough, turn over and get up based on the acoustic features extracted from the acoustic sound. Our experimental evaluation of six subjects over six months time period demonstrates that the breathing rate monitoring and sleep events detection are highly accurate and robust under various environments. By combining breathing rate and sleep events, our system can provide continuous and noninvasive fine-grained sleep monitoring for healthcare related applications, such as sleep apnea monitoring as evidenced by our experimental study.

Network configuration and policy updates occur frequently, and must be performed in a way that minimizes transient effects caused by intermediate states of the network. It has been shown that accurate time can be used for coordinating network-wide updates, thereby reducing temporary inconsistencies. However, this approach presents a great challenge; even if network devices have perfectly synchronized clocks, how can we guarantee that updates are performed at the exact time for which they were scheduled? In this paper we present a practical method for implementing accurate time-based updates, using TIMEFLIPs. A TimeFlip is a time-based update that is implemented using a timestamp field in a Ternary Content Addressable Memory (TCAM) entry. TIMEFLIPs can be used to implement Atomic Bundle updates, and to coordinate network updates with high accuracy. We analyze the amount of TCAM resources required to encode a TimeFlip, and show that if there is enough flexibility in determining the scheduled time, a TimeFlip can be encoded by a single TCAM entry, using a single bit to represent the timestamp, and allowing the update to be performed with an accuracy on the order of 1 microsecond.

To cope with the unprecedented growth of mobile data traffic, we investigate the performance gains obtained from unifying coverage-centric 4G mobile networks and capacity-centric fiber-wireless (FiWi) broadband access networks based on data-centric Ethernet technologies with resulting fiber backhaul sharing and WiFi offloading capabilities. Despite recent progress on backhaul-aware 4G studies with capacity-limited backhaul links, the performance-limiting impact of backhaul latency and reliability has not been examined in sufficient detail previously. In this paper, we evaluate the maximum aggregate throughput, offloading efficiency, and in particular the delay performance of FiWi enhanced LTE-A heterogeneous networks (HetNets), including the beneficial impact of various localized fiber-lean backhaul redundancy and wireless protection techniques, by means of probabilistic analysis and verifying simulation, paying close attention to fiber backhaul reliability issues and WiFi offloading limitations due to WiFi mesh node failures as well as temporal and spatial WiFi coverage constraints.

Physical human mobility has played an important role in the design and operation of mobile networks. Physical mobility, however, differs from user identity (name) mobility in both traditional mobility management protocols such as Mobile-IP and in new architectures, such as XIA and MobilityFirst, that support identity mobility and location independence as first class objects. A multi-homed stationary user or a stationary user shifting among multiple devices attached to different networks will persistently keep his/her identity but will change access networks and the IP address to which his/her identity is associated. We perform a measurement study of such user transitioning among networks from a network-level point of view, characterizing the sequence of networks to which a user is attached and discuss insights and implications drawn from these measurements. We characterize network transitioning in terms of network residency time, degree of multi-homing, transition rates and more. We find that users typically spend time attached to a small number of access networks, and that a surprisingly large number of users access two networks contemporaneously. We develop and validate a parsimonious Markov chain model of canonical user transitioning among networks that can be used to provision network services and to analyze mobility protocols.

The outputs of network monitoring such as traffic matrix and elephant flow identification are essential inputs to many network operations and system designs in DCNs, but most solutions for network monitoring adopt direct measurements or inference alone, which may suffer from either high network overhead or low precision. Different from those approaches, we combine the direct measurements offered by software defined network (SDN) and inference techniques based on network tomography to derive a hybrid network monitoring scheme in this paper; it can strike a balance between measurement overhead and accuracy. Essentially, we use SDN to make the severely low determined network tomography (TM estimation) problem in DCNs to be a more determined one. Thus many classic network tomography algorithms in ISP networks become feasible for DCNs. By combining SDN with network tomography, we can also identify the elephant flows with high precision while occupying very little network resource. According to our experiment results, the accuracy of estimating the TM is far higher than those inferred by SNMP link counters only and the performance of identifying elephant flows is also very promising.

This paper presents the first, to the best of our knowledge, detailed experimental study of 802.11n/ac throughput and power consumption in modern smartphones. We experiment with a variety of smartphones, supporting different subsets of 802.11n/ac features. We investigate the power consumption in various states of the wireless interface (sleep, idle, active), the impact of various features of 802.11n/ac (PHY bitrate, frame aggregation, channel bonding, MIMO) on both throughput and power consumption, and the tradeoffs between these two metrics. Some of our findings are significantly different from the findings of previous studies using 802.11n/ac wireless cards for laptop/desktop computers. We believe that these findings will help in understanding various performance and power consumption issues in today's smartphones and will guide the design of power optimization algorithms for the next generation of mobile devices.

A variety of dynamic bandwidth allocation (DBA) algorithms have been proposed to foster the performance of Ethernet passive optical networks (EPONs). These DBA algorithms use packet delay as an important quality of service (QoS) metric. This has led to a significant interest in developing mathematical models for analyzing the delay. These delay models often provide valuable qualitative results and worthwhile insights in understanding the mechanism of the delay and the manner in which it depends upon algorithm characteristics. Up to now, the delay models have been developed under some approximations, e.g., fixed packet sizes, negligible distances between a server and its users, a gated bandwidth assignment method, and Poisson traffic. In this paper, we develop the delay models for more realistic scenarios than the current state-of-the-art, including gated and limited bandwidth assignment methods, Poisson and Pareto traffic, and long-reach PONs in which the distance between the server and the users is significant and hence not negligible. We model different DBA paradigms, such as REPORT after data, REPORT before data, and multi-thread polling. The results from simulation experiments confirm the accuracy of the delay models.

Bluetooth Low Energy is a time-slotted wireless protocol aimed towards low power communication for battery-driven devices. As a power-management capability, whenever there is less data to send, the slave is allowed to remain in a low power mode during a given number of time-slots in a row. However, since the master does not know the exact sleep behavior of the slave, it has to wake-up at every time-slot and repeat its packets until the slave is awake. As a result, applications with variable throughput lead to many energy-consuming idle-slots at the master. In such applications, usually the connection parameters are chosen considering the worst case at design time and remain constant during operation. In this paper, we propose a novel power-management framework for BLE. Rather than skipping slots at the slave side, the proposed system updates the interval between two consecutive time-slots during runtime by applying online algorithms. To avoid data-loss or high delays, the framework guarantees that constraints on latency are met and buffers never overflow. Energy measurements of three different test-cases show that up to 42 percent of the energy consumption of a BLE master can be saved with our power management system.

The widely used RFID tags impose serious privacy concerns as a tag responds to queries from readers no matter they are authorized or not. The common solution is to use a commercially available blocker tag which behaves as if a set of tags with known blocking IDs are present. The use of blocker tags makes RFID estimation much more challenging as some genuine tag IDs are covered by the blocker tag and some are not. In this paper, we propose REB, the first RFID estimation scheme with the presence of blocker tags. REB uses the framed slotted Aloha protocol specified in the C1G2 standard. For each round of the Aloha protocol, REB first executes the protocol on the genuine tags and the blocker tag, and then virtually executes the protocol on the known blocking IDs using the same Aloha protocol parameters. The basic idea of REB is to conduct statistically inference from the two sets of responses and estimate the number of genuine tags. We conduct extensive simulations to evaluate the performance of REB, in terms of time-efficiency and estimation reliability. The experimental results reveal that our REB scheme runs tens of times faster than the fastest identification protocol with the same accuracy requirement.

Border surveillance for intrusion detection is an important application of wireless sensor networks. Given a set of mobile sensors and their initial positions, how to move these sensors to a region border to achieve barrier coverage energy-efficiently is challenging. In this paper, we study the 2-D MinMax barrier coverage problem of moving n sensors in a two-dimensional plane to form a barrier coverage of a specified line segment in the plane while minimizing the maximum sensor movement for the sake of balancing battery power consumption. Previously, this problem was shown to be NP-hard for the general case. It was an open problem whether the problem is polynomial-time solvable for the case when sensors have a fixed number of sensing ranges. We study a special case of great practical significance that the sensors have the same sensing range and present an O(n3log n) time algorithm. Our algorithm computes a permutation of the left and right endpoints of the moving ranges of all the sensors forming a barrier coverage and minimizes the maximum sensor movement distance by characterizing permutation switches that are critical. To the best of our knowledge, this is the first result for solving the 2-D MinMax barrier coverage problem for the case that all sensors have a uniform sensing range.

Sensor Network Calculus (SensorNC) provides a framework for worst-case analysis of wireless sensor networks. The analysis proceeds in two steps: For a given flow, (1) the network is reduced to a tandem of nodes by computing the arrival bounds of cross-traffic; (2) the flow is separated from the cross-traffic by subtracting cross-flows and concatenating nodes on its path. While the second step has seen much treatment, the first step has not at all. This is in sharp contrast to the fact that arrival bounding takes roughly 80% of the total analysis time and is equally crucial for the tightness of the bounds. Therefore, we turn our attention to this first SensorNC analysis step with the goal to boost the performance and applicability of the overall framework. The main technical contribution is a generalized version of the concatenation theorem within the SensorNC setting. This generalization is instrumental in simplifying and streamlining the cross-traffic arrival bound computations such that run times can be reduced by more than a factor of 5. Even more important, it enables a localization of the information necessary to execute the calculations at the node level, thus enabling a distribution of the SensorNC analysis within a self-modeling WSN.

In this paper we develop an analytical framework, based on Che's approximation [2], for the analysis of Least Recently Used (LRU) caches operating under the Shot Noise requests Model (SNM). The SNM was recently proposed in [10] to better capture the main characteristics of today Video on Demand (Vod) traffic. In this context, Che's approximation is derived as the application of a mean field principle to the cache eviction time. We investigate the validity of this approximation through an asymptotic analysis of the cache eviction time. Particularly, we provide a large deviation principle and a central limit theorem for the cache eviction time, as the cache size grows large. Furthermore, we obtain a non-asymptotic analytical upper bound on the error entailed by Che's approximation of the hit probability.

Smart Grid addresses the problem of existing power grid's increasing complexity, growing demand and requirement for greater reliability, through two-way communication and automated residential load control among others. These features also makes the Smart Grid a target for a number of cyber attacks. In the paper, we study the problem of rate alteration attack (RAA) through fabrication of price messages which induces changes in load profiles of individual users and eventually causes major alteration in the load profile of the entire network. Combining with cascading failure, it ends up with a highly damaging attack. We prove that the problem is NP-Complete and provide its inapproximability. We devise two approaches for the problem, former deals with maximizing failure of lines with the given resource and then extending the effect with cascading failure while the later takes cascading potential into account while choosing the lines to fail. To get more insight into the impact of RAA, we also extend our algorithms to maximize number of node failures. Empirical results on both IEEE Bus data and real network help us evaluate our approaches under various settings of grid parameters.

In this paper, we propose a Social-based Mobile Sybil Detection (SMSD) scheme to detect Sybil attackers from their abnormal contacts and pseudonym changing behaviors. Specifically, we first define four levels of Sybil attackers in mobile environments according to their attacking capabilities. We then exploit mobile users' contacts and their pseudonym changing behaviors to distinguish Sybil attackers from normal users. To alleviate the storage and computation burden of mobile users, the cloud server is introduced to store mobile user's contact information and to perform the Sybil detection. Furthermore, we utilize a ring structure associated with mobile user's contact signatures to resist the contact forgery by mobile users and cloud servers. In addition, investigating mobile user's contact distribution and social proximity, we propose a semi-supervised learning with Hidden Markov Model to detect the colluded mobile users. Security analysis demonstrates that the SMSD can resist the Sybil attackers from the defined four levels, and the extensive trace-driven simulation shows that the SMSD can detect these Sybil attackers with high accuracy.

Today's worldwide mobile data traffic is roughly 18× larger than the full internet traffic in 2000, and continued large growth is expected. High mobile data usage has implications both for users and providers. For individual users, relying on cellular data connectivity incurs high cellular data fees. For cellular network providers, high mobile data usage requires expensive, ongoing infrastructure upgrades. Cellular data usage can be reduced by offloading to WiFi when available. If not available, prior work has considered delaying transmissions to wait for WiFi availability. While exploiting such application delay tolerance offers significant energy and performance leverage for data offloading and other techniques, a key question is: how long to wait? Prior work does not discuss how to estimate application delay tolerance without explicit help from programmers, nor how to adjust the estimate dynamically. This work proposes, implements, and evaluates four schemes to dynamically and adaptively deduce an application's delay tolerance. These schemes (Adaptive, Decision Tree-Based, Hybrid and Lazy) are low-overhead and effective. In our experiments, they cut cellular usage by 2× or more compared to non-delay-tolerant approaches. Furthermore, our dynamically adaptive decision schemes achieve up to 15% further cellular data reduction compared to fixed static delay tolerance values.