Mining Parallel Patterns from Mobile Users

It is well-known that online services resort to various cookies to track users through users' online service identifiers (IDs) - in other words, when users access online services, various "fingerprints" are left behind in the cyberspace. As they roam around in the physical world while accessing online services via mobile devices, users also leave a series of "footprints" -- i.e., hints about their physical locations - in the physical world. This poses a potent new threat to user privacy: one can potentially correlate the "fingerprints" left by the users in the cyberspace with "footprints" left in the physical world to infer and reveal leakage of user physical world privacy, such as frequent user locations or mobility trajectories in the physical world - we refer to this problem as user physical world privacy leakage via user cyberspace privacy leakage. In this paper we address the following fundamental question: what kind - and how much - of user physical world privacy might be leaked if we could get hold of such diverse network datasets even without any physical location information. In order to conduct an in-depth investigation of these questions, we utilize the network data collected via a DPI system at the routers within one of the largest Internet operator in Shanghai, China over a duration of one month. We decompose the fundamental question into the three problems: i) linkage of various online user IDs belonging to the same person via mobility pattern mining; ii) physical location classification via aggregate user mobility patterns over time; and iii) tracking user physical mobility. By developing novel and effective methods for solving each of these problems, we demonstrate that the question of user physical world privacy leakage via user cyberspace privacy leakage is not hypothetical, but indeed poses a real potent threat to user privacy.

Finding prevalent mobile user patterns in large amount of data has been one of the major problems in the area of mobile data mining. Particularly, the algorithms of discovering frequent user's behavior patterns in the mobile agent system have been studied extensively in recent years. The key feature in most of these algorithms is that they use a location log dataset with userpsilas requested services. In this case, some problems occur because they do not consider that mobile user's behavior patterns are dynamically variable as time passes. In this paper, we propose a novel data mining method based on temporal mobile access patterns that can efficiently discover mobile user's behavior patterns. An advantage of our method compactly stores user's behavior patterns according to location log dataset and requested services in memory. Thus, even the information datasets require large shared memory when they store, our approach still provides faster access and consume less memory than existing techniques.

Understanding user behavior of mobile Internet is vital for network infrastructure planning and radio resource allocation. In this paper, we investigate the spatial-temporal characteristics of users' access behavior and mobility patterns based on a data set collected from a cellular network in a city of China. We first propose a method called Maximum Distance Criterion (MDC) to distinguish users' access stability in different areas and periods. The analysis shows that urban users' access behavior is less stable than that of rural users. Furthermore, urban users' access behavior is more stable in holidays than in workdays, but contrary for that of rural users. We then introduce the concept of centroid to explore users' mobility patterns at the collective level. We find that users in different areas all move toward a highway during holidays while come back when workdays approaching. These results give us a better understanding of user behavior in the mobile Internet from both spatial and temporal perspective, which is essential for network planning.

Mining user behavior patterns in mobile environments is an emerging topic in data mining fields with wide applications. By integrating moving paths with purchasing transactions, one can find the sequential purchasing patterns with the moving paths, which are called mobile sequential patterns of the mobile users. Mobile sequential patterns can be applied not only for planning mobile commerce environments but also for analyzing and managing online shopping websites. However, unit profits and purchased numbers of the items are not considered in traditional framework of mobile sequential pattern mining. Thus, the patterns with high utility (i.e., profit here) cannot be found. In view of this, we aim at integrating mobile data mining with utility mining for finding high-utility mobile sequential patterns in this study. Two types of algorithms, namely level-wise and tree-based methods, are proposed for mining high-utility mobile sequential patterns. A series of analyses and comparisons on the performance of the two different types of algorithms are conducted through experimental evaluations. The results show that the proposed algorithms outperform the state-of-the-art mobile sequential pattern algorithms and that the tree-based algorithms deliver better performance than the level-wise ones under various conditions.

Mobility modeling is the mathematical modeling of mobile users' (cars, cell phone users) movement patterns. The resulting model not only provides us with an understanding of past mobile user movements, but also enables us to predict how a mobile user might move in the future. It has been found useful in both infrastructure-based wireless network and ad hoc network for protocol evaluation, resource planning, etc. A nonparametric hierarchical Bayesian approach is proposed in this paper for extracting hierarchical mobility patterns from mobile user traces. Experiment results show that the proposed method is able to generate a hierarchical mobility model that better reflect the mobility pattern structure in many scenarios. It also has better future movement prediction compared to the hidden Markov model.

We propose a new self-localization method for a mobile robot equipped with a fixed-viewpoint active camera by observing multiple parallel lines patterns with the camera. The method utilizes the technique for estimating location and posture of the camera from one parallel lines pattern which was previously developed by our research group. The method switches from a certain parallel lines pattern used for self-localization to another parallel lines pattern while the robot is moving in the environment. After the new parallel lines pattern is selected, the method keeps estimating location and posture of the camera from the parallel lines pattern. Our method can estimate location and posture of the robot and the parallel lines patterns in the world coordinate system fixed to the initial parallel lines pattern by repeating the switch of parallel lines patterns and the self-localization based on the parallel line pattern. Through the experiment of the self-localization of the real mobile robot, we confirmed the validity of the proposed method.

In recent years, there has been a growing demand in 3G data services, leading to deteriorating 3G service quality. Noting that Wireless Local Area Networks (WLANs) as well as 3G cellular networks are widely available today, WLANs could be effectively utilized to relieve the overload in the 3G networks. On the other hand, use of IEEE 802.11 WLAN Access Points (APs) has proliferated tremendously, resulting in a communication device inside a mobile vehicle to access the Internet. However, using Internet through APs in moving vehicles is challenging since WLAN APs have a short range and are typically not deployed to cover all roads. Several studies have investigated the performance of using intermittently available WLAN connectivity from moving vehicles for data transfers and predictive offloading in WLAN/3G networks. However, these works have not addressed mobility pattern from the viewpoint that drivers' mobility is generally known to have a daily routine. Therefore, in this paper, we consider the user's historical mobility to decide to offload data to WLAN instead of switching to 3G network. The user's application usage pattern is also considered into predicting available WLANs. To evaluate the performance of our offloading algorithm, we analyze the prediction error and conduct simulations. The simulation results show that the proposed algorithm achieves shorter transmission time than the existing schemes that do not consider user's mobility pattern by delivering more data to the WLANs.

In this work, a cognitive relay network (CRN) with interference constraint from the primary user (PU) with a mobile end user is studied. The proposed system model employs a half-duplex transmission between a single PU and a single secondary user (SU). In addition, an amplify and forward (AF) relaying technique is employed between the SU source and SU destination. In this context, the mobile end user (SU destination) is assumed to move at high vehicular speeds, whereas other nodes (SU Source, SU relays and PU) are assumed to be stationary. The proposed scheme dynamically determines the best relay for transmission based on the highest signal-to-noise (SNR) ratio by deploying selection combiner at the SU destination, thereby achieving diversity. All channels connected with the stationary nodes are modelled using Rayleigh distribution, whereas all other links connected with the mobile end user are modelled using Nakagami-m fading distribution (m<1). The outage probabilities (OPs) are obtained considering several scenarios and Monte Carlo simulation is used to verify the numerical results. The obtained results show that a variety of factors, including the number of SU relays, the severity of the fading channels, the position of the PU, the fading model, and the mobile end user speed, might influence the CRN’s performance.

Location-based services (LBSs) typically crowdsource geo-tagged data from mobile users. Collecting more data will generally improve the utility for LBS providers; however, it also leads to more privacy exposure of users’ mobility patterns. Although the tension between data utility and user privacy has been recognized, there lacks a solution that determines how much data to collect—in both spatial and temporal domains—is the “best” for both mobile users and the service provider. This article proposes a strategy toward making an optimal tradeoff such that a user submits data only if her mobility privacy will not be compromised and the data utility of the LBS provider will be sufficiently improved. To this end, we first define and formulate a concept called privacy exposure , which incorporates both the spatial distribution and the temporal transition of a user’s activity points . Second, we define and quantify data utility in terms of spatial repetitions and temporal closeness among data based on an economic principle. Then, we propose a PRivacy-preserving and UTility-Enhancing Crowdsourcing (PRUTEC) algorithm to determine, on behalf of each mobile user, whether a newly sensed piece of data should be submitted to the LBS provider. Our simulation demonstrates that PRUTEC improves the data utility of the service provider with a much less amount of data to collect and reduces privacy exposure for mobile users while collecting useful data continuously.

Many people increasingly rely on mobile communication services to carry out daily activities. Due to the limitation of the PCS network architecture, a constantly relocating user may encounter significant delay when requesting data or value-added services. Previous research showed that this inefficiency can be effectively reduced by predicting the user's mobile patterns. However, most research merely focused on the user's moving nodes without considering the traffic times and requested services which could dramatically affect the user's behavior. The research also did not consider how likely the user is going to relocate. Thus, in this work, we extend the hidden Markov model for modeling the behavior of the mobile users with regard to the following important factors: 1) moving node, 2) requested service, 3) user state, and 4) traffic time. Our novel approach requires only one scan of the target dataset. Moreover, the needed memory space and processing time can be independent of the transaction size. A user model can be built to predict the user's mobile patterns at different granularity levels, as well as for decision support and service improvement. Moreover, the built model can be easily adjusted later to reflect the latest user behavior without re-scanning the original dataset. Our approach can also be readily used to mine streaming data.

We formulate the optimum personal paging area problem as an interval estimation problem, with an aim to reduce the paging signaling cost by minimizing the size of the paging area. A paging algorithm is developed to set-up the optimal one-step paging area that contains the locations of a mobile user with a high probability and thus to gain a coverage that is matched to the user's time-varying mobility pattern. With a particular mobility model, the effects of the user's mobility characteristics and the system parameters on the performance of the paging procedure are investigated.

A number of developments in areas related to Big Data analytics have led to the realization of our PDM framework reported in this monograph. Mobile data mining, data stream mining, and parallel and distributed data mining and mobile software agents are four areas that collectively contributed to the possibility of developing PDM. Mobile software agents as a flexible software development framework allowed PDM to have on the fly deployment of data mining techniques, if needed. Based on these areas, the four pillars for Pocket Data Mining can be depicted as in Figure 2.1. These areas will be reviewed in this chapter.KeywordsData MiningMobile DeviceCloud ComputingData Mining TechniqueData Mining AlgorithmThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Fast Fourier Transform(FFT) represents a common communication pattern shared by a large class of scientific and engineering problems and wavelength assignment is a key issue to increase efficiency and reduce cost in Wavelength Division Multiplexing (WDM) optical networks. In this paper, we propose a new scheme for the wavelength assignment of parallel FFT communication pattern on WDM linear arrays. By lattice embedding, the number of wavelengths required to realize parallel FFT communication pattern on WDM linear arrays significantly improves the known result. Our proposed embedding method also provides a new approach to the hypercube layout problem considering connections dimension by dimension rather than all connections as in the traditional approach.

Increasing number of mobile users accesses large multimedia files (such as high definition audio, video, images, etc.) over the Internet. Therefore, the bandwidth demand for mobile Internet access is getting exponentially larger. To answer users' demand, Carrier Aggregation is proposed in LTE-A. In Carrier Aggregation, multi bands are used and the bands have supported different coverage. Therefore, mobile users can simultaneously connect only one or multi bands. Because of mobility of users, traffic types and assigned channel errors, the best available Carrier Components of each band should be assigned to each user in order to increase quality of services. Several works have been proposed in the literature to address Carrier Components assignment to mobile users by using Channel Quality Indicator, quality of service and traffic types in LTE systems. However, continuously increasing data requests of users forces the operators to manage traffic more intelligently. Therefore, we have proposed a novel Carrier Component assignment method which considers user profiles and traffic types to increase quality of services and experiences getting by mobile users. Results show that the proposed method uses system resources efficiently and can provide improved throughput rate in LTE and LTE-A systems. Our method will help service providers build efficient Carrier Component assignment services through considering user profile and traffic types.

In this paper, the problem of proactive caching is studied for cloud radio access networks (CRANs). In the studied model, the baseband units (BBUs) can predict the content request distribution and mobility pattern of each user and determine which content to cache at remote radio heads and the BBUs. This problem is formulated as an optimization problem, which jointly incorporates backhaul and fronthaul loads and content caching. To solve this problem, an algorithm that combines the machine learning framework of echo state networks (ESNs) with sublinear algorithms is proposed. Using ESNs, the BBUs can predict each user's content request distribution and mobility pattern while having only limited information on the network's and user's state. In order to predict each user's periodic mobility pattern with minimal complexity, the memory capacity of the corresponding ESN is derived for a periodic input. This memory capacity is shown to capture the maximum amount of user information needed for the proposed ESN model. Then, a sublinear algorithm is proposed to determine which content to cache while using limited content request distribution samples. Simulation results using real data from Youku and the Beijing University of Posts and Telecommunications show that the proposed approach yields significant gains, in terms of sum effective capacity, that reach up to 27.8% and 30.7%, respectively, compared with two baseline algorithms: random caching with clustering and random caching without clustering.