Goldilocks: Learning Pattern-Based Task Assignment in Mobile Crowdsensing

Mobile crowdsensing (MCS) has heated up and has become a new paradigm of data collection. In the process of the task allocation of MCS, users are often required to provide their own location information with the server to conveniently dispatch some suitable tasks to them. However, it is possible for malicious servers to infer some sensitive information based on the user’s location such as the user’s home address or the user’s trajectory and so on which will cause serious privacy issues. Recently, differential privacy (DP) has become a promising privacy protection scheme. However, the existing DP schemes in location privacy protection for MCS do not pay attention to the accuracy of task allocation as equally as the effects of privacy protection, which often results in the task allocation to be inaccurate and inefficient. In order to overcome the shortcoming, we propose a novel MCS task allocation scheme integrating the mapping accuracy of task-worker with the privacy-preserving effect on the user’s location. Besides edge devices, its implementation system consists of a task allocation server and a third party, both of which are semitrusted, that is, both of them only know about the user’s rough location information but cannot obtain the user’s exact location. At first, to improve the response speed of MCS task requests and the location privacy protection of MCS users, our scheme can exploit more than one edge-computing node nearby a user to cooperatively participate in MCS task allocation by aggregating him/her and the near users into groups. Then, our scheme can further enhance the location privacy protection effect based on the Johnson–Lindenstrauss (JL) transformation, which can achieve accurate task allocation and hold the merits of DP. Finally, we verify the feasibility by some experiments based on two data sets. The performance is compared with that of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$t$ </tex-math></inline-formula> he typical DP. The results show that our scheme not only provides strict privacy guarantees but also has higher performance.

Worker selection is always one of the most fundamental problems in mobile crowdsensing (MCS), since the reliability of workers' sensing data is hugely significant to the service quality. In the worker selection process, it is inevitable for the workers to share some of their sensitive information. Consequently, numerous studies are conducted on the problem of privacy-preserving worker selection in MCS platforms. However, most of the existing methods focus on static and short-term situations. As a result, they are inapplicable to the highly dynamic environments where the MCS tasks are long term and the workers can continuously arrive at/leave the system. To solve these problems, in this article, we propose a privacy-preserving worker selection scheme based on the probabilistic skyline over sliding windows. Specifically, the proposed scheme can select reliable workers for each current sliding window in terms of working experience, expiry time, and trustability. Besides, we design an ElGamal encryption-based scheme for securely outsourcing and comparing workers' personal information without revealing their privacy. Detailed security analysis shows that the workers' sensitive information, e.g., working experience and trustability, are not revealed to any authorized parties during the process of MCS under our security model. Furthermore, extensive experiments on both real-world and simulated data sets demonstrate that our proposed scheme outperforms the baseline method in two application scenarios, i.e., 1) continuous worker arrival and 2) continuous worker departure.

Mobile crowdsensing (MCS) can be widely adopted for undertaking a variety of tasks for the smart city by recruiting distributed participants. With the rapid proliferation of MCS applications, tasks can be devised with different sensing scales. Hence, allocating the tasks to participants with variable spatiotemporal granularity would be urgent and challenging. To this end, we propose a behavior sequentializing and task mapping (BETA) framework, where MCS tasks can be flexibly mapped based on the moving behavior of a participant’s daily activities. A sequential behavior model is first proposed to describe the moving behavior of a participant’s daily activities. Specifically, the behavior of a daily activity is represented by the spatiotemporal trajectory distribution, and a sequential behavior graph is adopted to model the relationships among multiple activities. To extract the sequential behavior from historical trajectories, a projection-clustering (PC) algorithm and a behavior sequentializing approach are developed. Then, based on the extracted sequential behavior, a behavior-to-task mapping network is proposed to evaluate the fitness between tasks and a participant. Finally, extensive simulations are conducted to demonstrate the functionality and performance of BETA. The results show that our proposed method outperforms the state-of-the-art solutions in the mapping efficiency of tasks with different sensing scales.

With the pervasiveness of mobile devices, satisfying spatial-temporal coverage requirements in interested regions while considering the quality of the sensing data and the budget constraint is a major research challenge in mobile crowdsensing (MCS). In this paper, we define a novel coverage metric, quality coverage, which considers both the fraction of subareas covered by sensor readings and the quality of sensing data in each covered subarea, for diversified location-based MCS tasks. In our system, as participants with high reliability level can contribute a lot to sensing tasks, we design an incentive model that provides higher bonus to more reliable participants. After knowing the reward for the per amount of sensing data, each worker can submit an interested task set that contains several unit tasks. The platform assigns MCS tasks to these mobile workers by solving the task allocation optimization problems, which select the optimal mobile worker set under the budget of data requesters to maximize diverse and spatial coverage level in the whole sensing area. As the task allocation problems are proven to be NP-hard, we propose a heuristic and a polynomialtime greedy approximation algorithm to solve optimization problems in two different application scenarios. Simulations using both real and synthetic data sets show that our algorithms outperform existing approaches.

Mobile crowd sensing (MCS) is a novel example of urban sensing, which uses smart phones, smart wearable devices, on-board sensors and other mobile sensor devices to collect data, and is widely used in intelligent transportation, medical care, environmental monitoring and other fields. A growing number of studies show that a good task allocation mechanism is an indispensable key factor in determining whether MCS applications can obtain good service quality. In this paper, we introduce the realistic background of MCS and combine some specific applications to exhaustively classify task allocation problems in MCS. Then, the typical algorithm for MCS task allocation is classified and illustrated. Finally, the current research challenges and future research directions are presented.

Opportunistic sensing has become an appealing mobile crowd sensing (MCS) paradigm due to the fact that it can reduce the energy consumption and cost of cellular network connections. However, its success rate and transmission speed depend on the social interaction and mobility patterns of nodes. In this paper, we provide a spatiotemporal opportunistic transmission method for MCS networks. Firstly, to characterize the mobility patterns and social attributes of nodes more precisely and combine their advantages, this method defines spatiotemporal encountering and visiting parameters related to specific space-time units for nodes in a MCS network. Further, to realize reliable opportunistic transmission across regions and time intervals, this method searches publishers or participants of sensing tasks in a space-time unit according to the spatiotemporal encountering parameters of nodes in the unit and tracks the publishers or participants across the space-time units according to the spatiotemporal visiting parameters of nodes. The simulation results verify that the proposed method can achieve higher success rate with less transmission delay than existing typical methods.

Mobile crowdsensing (MCS) has a huge potential to provide societal benefits by effectively utilizing the sensing, computing, and networking capability of mobile devices, which have become ubiquitous especially in the developed world. However, many challenges come along with MCS such as energy cost, privacy issues, and the data integrity of crowdsensed data. In this paper, we focus on analyzing the data integrity of mobile crowdsensed data in a user study of 60 people who participate in a crowdsensing campaign that collects barometric pressure data from various locations of our campus. Each set of 20 users runs one of the three different crowdsensing frameworks, one of which, called SENSE-AID, is an energy-efficient framework designed by us. We analyze the characteristics of the data with respect to their integrity. From our analysis, we find that for 90 MCS tasks there is a surprisingly high number of outlier percentage (close to 20%), only about 10% of the participants report data for the entire duration of 7 days of the MCS campaign, and using a reputation system to filter out spurious data values helps in getting within 0.8% of the ground truth barometric pressure, whereas not using the reputation system leads to a discrepancy of about 20%. We hope our analysis will provide a useful reference to MCS researchers and developers in their future work running MCS campaigns.

Task allocation is a major challenge in Mobile Crowd Sensing (MCS). While previous task allocation approaches follow either the opportunistic or participatory mode, this paper proposes to integrate these two complementary modes in a two-phased hybrid framework called HyTasker. In the offline phase, a group of workers (called opportunistic workers) are selected, and they complete MCS tasks during their daily routines (i.e., opportunistic mode). In the online phase, we assign another set of workers (called participatory workers) and require them to move specifically to perform tasks that are not completed by the opportunistic workers (i.e., participatory mode). Instead of considering these two phases separately, HyTasker jointly optimizes them with a total incentive budget constraint. In particular, when selecting opportunistic workers in the offline phase of HyTasker, we propose a novel algorithm that simultaneously considers the predicted task assignment for the participatory workers, in which the density and mobility of participatory workers are taken into account. Experiments on two real-world mobility datasets demonstrate that HyTasker outperforms other methods with more completed tasks under the same budget constraint.

Mobile Crowdsensing (MCS) is a new paradigm that recruits users to cooperatively perform a sensing task. When recruiting users, existing works mainly focus on selecting a group of users with the best objective ability, e.g., the user's probability or frequency of covering the task locations. However, we argue that, for the cooperative MCS task, the completion effect depends not only on the user's objective ability, but also on their subjective collaboration likelihood with each other. In other words, in each single round, we prefer to recruit users with not only a strong objective ability but also good collaboration likelihood. Moreover, even though we could find a well-behaved group of users in a single round, in the multi-round scenario without enough prior knowledge, we still face the problem of recruiting previously well-behaved user groups (exploitation) or recruiting unknown user groups (exploration). To address these problems, in this paper, we first convert the single-round user recruitment problem into the min-cut problem and propose a graph theory based algorithm to find the optimal group of users. Furthermore, in the multi-round scenario, to balance the trade-off between exploration and exploitation, we propose the multi-round User Recruitment strategy based on the combinatorial Multi-armed Bandit model (URMB) and prove that it can achieve a tight regret bound. Finally, extensive experiments on three real-world datasets validate that the users recruited by URMB result in a better task completion effect than the state-of-the-art strategy.

Mobile crowdsensing (MCS) is a new and promising paradigm of data collection in large-scale sensing and computing. A large group of users with mobile devices are recruited in a specific area to accomplish sensing tasks. An essential aspect of an MCS application is task allocation, which aims to efficiently assign sensing tasks to the recruited workers. Due to various resource and quality constraints, the MCS task allocation problem is often an NP-hard optimization problem. Traditional greedy or heuristic approaches are usually subject to performance loss in a certain degree so as to maintain tractability or accommodate special requirements such as incentive constraints. In this paper, we attempt to employ a deep reinforcement learning method to search for a more efficient task allocation solution. Specifically, we use a double deep Q-network (DDQN) to solve the task allocation problem as a path-planning problem with time windows. Our formulated problem takes into account location-dependency and time-sensitivity of sensing tasks, as well as the resource limits of workers in terms of maximum travelling distances. Simulations are conducted to compare the DDQN-based solution with two standard baseline solutions. The results show that our proposed solution outperforms the baseline solutions in terms of the platform’s profit and the coverage of tasks.

Task assignment is a critical issue in mobile crowdsensing (MCS), an emerging sensing paradigm applied to realize various sensing applications for smart cities. Existing task assignment schemes mostly require the exact location information of tasks and workers for optimization, which inevitably brings the issue of location privacy leakage. Therefore, researchers have started investigating privacy-preserving task assignment schemes. However, most of these works either make inaccurate assignments or only concentrate on workers' privacy. In this article, we propose a novel bilateral privacy-preserving and accurate task assignment framework in fog-assisted MCS, called BRAKE. Specifically, we utilize the multisecret sharing scheme to preserve location privacy in the MCS task assignment, where tasks and workers only need to provide the secret shares of their real location information to fog nodes. Moreover, we consider distance-oriented and time-oriented tasks for assignment optimization and propose an adaptive top-k worker selection algorithm to accurately select the most suitable workers. The security analysis proves that BRAKE can resist collusion attacks, and the extensive evaluation results demonstrate the efficiency and accuracy of BRAKE.

Humans vary greatly in their motor learning abilities, yet little is known about the neural mechanisms that underlie this variability. Recent neuroimaging and electrophysiological studies demonstrate that large-scale neural dynamics inhabit a low-dimensional subspace or manifold, and that learning is constrained by this intrinsic manifold architecture. Here, we asked, using functional MRI, whether subject-level differences in neural excursion from manifold structure can explain differences in learning across participants. We had subjects perform a sensorimotor adaptation task in the MRI scanner on 2 consecutive days, allowing us to assess their learning performance across days, as well as continuously measure brain activity. We find that the overall neural excursion from manifold activity in both cognitive and sensorimotor brain networks is associated with differences in subjects’ patterns of learning and relearning across days. These findings suggest that off-manifold activity provides an index of the relative engagement of different neural systems during learning, and that subject differences in patterns of learning and relearning are related to reconfiguration processes occurring in cognitive and sensorimotor networks.

It is evident that user training significantly affects performance of pattern-recognition-based myoelectric prosthetic device control. Despite plausible classification accuracy on offline datasets, online accuracy usually suffers from the changes in physiological conditions and electrode displacement. The user ability in generating consistent electromyographic (EMG) patterns can be enhanced via proper user training strategies in order to improve online performance. This study proposes a clustering-feedback strategy that provides real-time feedback to users by means of a visualized online EMG signal input as well as the centroids of the training samples, whose dimensionality is reduced to minimal number by dimension reduction. Clustering feedback provides a criterion that guides users to adjust motion gestures and muscle contraction forces intentionally. The experiment results have demonstrated that hand motion recognition accuracy increases steadily along the progress of the clustering-feedback-based user training, while conventional classifier-feedback methods, i.e., label feedback, hardly achieve any improvement. The result concludes that the use of proper classifier feedback can accelerate the process of user training, and implies prosperous future for the amputees with limited or no experience in pattern-recognition-based prosthetic device manipulation.It is evident that user training significantly affects performance of pattern-recognition-based myoelectric prosthetic device control. Despite plausible classification accuracy on offline datasets, online accuracy usually suffers from the changes in physiological conditions and electrode displacement. The user ability in generating consistent electromyographic (EMG) patterns can be enhanced via proper user training strategies in order to improve online performance. This study proposes a clustering-feedback strategy that provides real-time feedback to users by means of a visualized online EMG signal input as well as the centroids of the training samples, whose dimensionality is reduced to minimal number by dimension reduction. Clustering feedback provides a criterion that guides users to adjust motion gestures and muscle contraction forces intentionally. The experiment results have demonstrated that hand motion recognition accuracy increases steadily along the progress of the clustering-feedback-based user training, while conventional classifier-feedback methods, i.e., label feedback, hardly achieve any improvement. The result concludes that the use of proper classifier feedback can accelerate the process of user training, and implies prosperous future for the amputees with limited or no experience in pattern-recognition-based prosthetic device manipulation.

First European Experience of Shape-Sensing Robotic-Assisted Bronchoscopy: Learning Curve Analysis

Web based learning has been promoting alternative way of learning for decades. The difficulty of web based learning is to provide the appropriate support for the learners so that the learners will not get lost and their learning achievements can be ensured. This paper thus proposes the method for discovering learning patterns of the learners on web based learning particularly for ensuring the learning achievement. The learning pattern is discovered by analyzing the interactions among the learners and the learning objects with social network analysis. Then, the achievement learning pattern is finally determined by analyzing the sets of obtained social network measurements. The interaction data is gathered from online course named introduction to Information Technology in the 2013 academic year, particularly for spreadsheet content module having 10 learning objects. The interaction patterns only of two groups of students including scientific and nonscientific background knowledge who pass the spreadsheet examination are analyzed. Finally, learning patterns ensuring learning achievement for spreadsheet content module of those students having different background knowledge is revealed.