Internet of Things (IoT) generates large amounts of data at the network edge. Machine learning models are often built on these data, to enable the detection, classification, and prediction of the future events. Due to network bandwidth, storage, and especially privacy concerns, it is often impossible to send all the IoT data to the data center for centralized model training. To address these issues, federated learning has been proposed to let nodes use the local data to train models, which are then aggregated to synthesize a global model. Most of the existing work has focused on designing learning algorithms with provable convergence time, but other issues, such as incentive mechanism, are unexplored. Although incentive mechanisms have been extensively studied in network and computation resource allocation, yet they cannot be applied to federated learning directly due to the unique challenges of information unsharing and difficulties of contribution evaluation. In this article, we study the incentive mechanism for federated learning to motivate edge nodes to contribute model training. Specifically, a deep reinforcement learning-based (DRL) incentive mechanism has been designed to determine the optimal pricing strategy for the parameter server and the optimal training strategies for edge nodes. Finally, numerical experiments have been implemented to evaluate the efficiency of the proposed DRL-based incentive mechanism.

A Learning-Based Incentive Mechanism for Federated Learning

Mobile crowdsensing (MCS) has gained significant attention in recent years and has become an appealing paradigm for urban sensing. For data collection, MCS systems rely on contribution from mobile devices of a large number of participants or a crowd. Smartphones, tablets, and wearable devices are deployed widely and already equipped with a rich set of sensors, making them an excellent source of information. Mobility and intelligence of humans guarantee higher coverage and better context awareness if compared to traditional sensor networks. At the same time, individuals may be reluctant to share data for privacy concerns. For this reason, MCS frameworks are specifically designed to include incentive mechanisms and address privacy concerns. Despite the growing interest in the research community, MCS solutions need a deeper investigation and categorization on many aspects that span from sensing and communication to system management and data storage. In this paper, we take the research on MCS a step further by presenting a survey on existing works in the domain and propose a detailed taxonomy to shed light on the current landscape and classify applications, methodologies, and architectures. Our objective is not only to analyze and consolidate past research but also to outline potential future research directions and synergies with other research areas.

/pdf/a-survey-on-mobile-crowdsensing-systems-challenges-solutions-224oot2vez.pdf

A Survey on Mobile Crowdsensing Systems: Challenges, Solutions, and Opportunities

Recent years have witnessed the emergence of mobile crowd sensing (MCS) systems, which leverage the public crowd equipped with various mobile devices for large scale sensing tasks. In this paper, we study a critical problem in MCS systems, namely, incentivizing user participation. Different from existing work, we incorporate a crucial metric, called users' quality of information (QoI), into our incentive mechanisms for MCS systems. Due to various factors (e.g., sensor quality, noise, etc.) the quality of the sensory data contributed by individual users varies significantly. Obtaining high quality data with little expense is always the ideal of MCS platforms. Technically, we design incentive mechanisms based on reverse combinatorial auctions. We investigate both the single-minded and multi-minded combinatorial auction models. For the former, we design a truthful, individual rational and computationally efficient mechanism that approximately maximizes the social welfare with a guaranteed approximation ratio. For the latter, we design an iterative descending mechanism that achieves close-to-optimal social welfare while satisfying individual rationality and computational efficiency. Through extensive simulations, we validate our theoretical analysis about the close-to-optimal social welfare and fast running time of our mechanisms.

/pdf/quality-of-information-aware-incentive-mechanisms-for-mobile-2gmvug77wm.pdf

Quality of Information Aware Incentive Mechanisms for Mobile Crowd Sensing Systems

Big data is a term used for very large data sets that have more varied and complex structure. These characteristics usually correlate with additional difficulties in storing, analyzing and applying further procedures or extracting results. Big data analytics is the term used to describe the process of researching massive amounts of complex data in order to reveal hidden patterns or identify secret correlations. However, there is an obvious contradiction between the security and privacy of big data and the widespread use of big data. This paper focuses on privacy and security concerns in big data, differentiates between privacy and security and privacy requirements in big data. This paper covers uses of privacy by taking existing methods such as HybrEx, k-anonymity, T-closeness and L-diversity and its implementation in business. There have been a number of privacy-preserving mechanisms developed for privacy protection at different stages (for example, data generation, data storage, and data processing) of a big data life cycle. The goal of this paper is to provide a major review of the privacy preservation mechanisms in big data and present the challenges for existing mechanisms. This paper also presents recent techniques of privacy preserving in big data like hiding a needle in a haystack, identity based anonymization, differential privacy, privacy-preserving big data publishing and fast anonymization of big data streams. This paper refer privacy and security aspects healthcare in big data. Comparative study between various recent techniques of big data privacy is also done as well.

https://link.springer.com/content/pdf/10.1186%2Fs40537-016-0059-y.pdf

Big data privacy: a technological perspective and review

This paper provides a state-of-the-art literature review on economic analysis and pricing models for data collection and wireless communication in Internet of Things (IoT). Wireless sensor networks (WSNs) are the main components of IoT which collect data from the environment and transmit the data to the sink nodes. For long service time and low maintenance cost, WSNs require adaptive and robust designs to address many issues, e.g., data collection, topology formation, packet forwarding, resource and power optimization, coverage optimization, efficient task allocation, and security. For these issues, sensors have to make optimal decisions from current capabilities and available strategies to achieve desirable goals. This paper reviews numerous applications of the economic and pricing models, known as intelligent rational decision-making methods, to develop adaptive algorithms and protocols for WSNs. Besides, we survey a variety of pricing strategies in providing incentives for phone users in crowdsensing applications to contribute their sensing data. Furthermore, we consider the use of some pricing models in machine-to-machine (M2M) communication. Finally, we highlight some important open research issues as well as future research directions of applying economic and pricing models to IoT.

/pdf/data-collection-and-wireless-communication-in-internet-of-k2ds6e1lip.pdf

Data Collection and Wireless Communication in Internet of Things (IoT) Using Economic Analysis and Pricing Models: A Survey

In this paper, we tackle the problem of stimulating smartphone users to join mobile crowdsourcing applications with smartphones. Different from existing work of mechanism design, we uniquely take into consideration thecrucial dimension of location informationwhen assigning sensing tasks to smartphones. However, the location awareness largely increases the theoretical and computational complexity. In this paper, we introduce a reverse auction frameworkto model the interactions between the platform and the smartphones. We rigorously prove that optimally determining the winning bids isNP hard. In this paper we design a mechanism called TRACwhich consists of two main components. The first component is a near-optimal approximate algorithm for determining the winning bids with polynomial-time computation complexity, which approximates the optimal solution within a factor of 1+l n(n), wheren is the maximum number of sensing tasks that a smartphone can accommodate. The second component is a critical payment scheme which, despite the approximation of determining winning bids, guarantees that submitted bids of smartphones reflect their real costs of performing sensing tasks. Through both rigid theoretical analysis and extensive simulations, we demonstrate that the proposed mechanism achieves truthfulness, individual rationality and high computation efficiency.

TRAC: Truthful Auction for Location-Aware Collaborative Sensing in Mobile Crowdsourcing

Rapid advance of location acquisition technologies boosts the generation of trajectory data, which track the traces of moving objects. A trajectory is typically represented by a sequence of timestamped geographical locations. A wide spectrum of applications can benefit from the trajectory data mining. Bringing unprecedented opportunities, large-scale trajectory data also pose great challenges. In this paper, we survey various applications of trajectory data mining, e.g., path discovery, location prediction, movement behavior analysis, and so on. Furthermore, this paper reviews an extensive collection of existing trajectory data mining techniques and discusses them in a framework of trajectory data mining. This framework and the survey can be used as a guideline for designing future trajectory data mining solutions.

A Survey on Trajectory Data Mining: Techniques and Applications

The invention discloses a method and a system for preventing fraudulent auction in a group intelligent perception system. The method comprises the following steps that each intelligent mobile phone user registers in an auction part platform, and the platform sends all perception task lists and auction modes to the user; each intelligent mobile phone user makes a bid document and submits the bid document to the auction part platform; the platform selects a group of intelligent mobile phone users according to the bid documents submitted by all the intelligent mobile phone users, so that the group of user can finish all the perception tasks, the sum of quoted price is the smallest, and meanwhile, the platform calculates the payment which should be paid to each selected intelligent mobile phone user and informs the selected users of starting executing corresponding perception tasks; each selected intelligent mobile phone user executes the corresponding perception tasks and feeds a result to the platform; the platform pays corresponding payment to each participant user according to the calculated payment. The auction method which can efficiently finish the perception tasks and can prevent cheat is realized.

Method and system for preventing fraudulent auction in group intelligent perception system

Mobile crowdsourcing with smartphones advocates the cooperative effort of mobile smartphones to perform a joint distributed sensing task, which has gained growing importance for its potential to support a wide spectrum of large-scale sensing applications. Smartphone users in the real world are strategic and rational. Thus, one crucial problem in mobile crowdsourcing with smartphones is to stimulate cooperation from smartphone users. Several major challenges should be addressed. First, the actual cost incurred for a sensing task is private information and unknown to other users and the mobile crowdsourcing platform. Second, smartphone users are strategic, which suggest a user may deliberately misreport its cost (different from the real cost) in order to maximize its own utility. In this paper, we propose a strategy-proof incentive mechanism called iMac based on the Vickrey-Clarke-Groves (VCG) mechanism. The main idea of iMac is to stimulate smartphone users to truthfully disclose their real costs in spite of strategic behavior of the users. iMac introduces two main components. The first component determines the allocation of a sensing task to smartphone users given the user costs. And the second component decides the payment to each user. We prove that iMac can successfully produce a unique Nash equilibrium at which each user truthfully discloses the cost. Meanwhile, the minimization of the social cost is achieved. Simulation results demonstrate iMac achieves the desired design objectives and the overpayment is modest.

Zhenni Feng

Papers

TRAC: Truthful Auction for Location-Aware Collaborative Sensing in Mobile Crowdsourcing

A Survey on Trajectory Data Mining: Techniques and Applications

Method and system for preventing fraudulent auction in group intelligent perception system

IMac: Strategy-proof incentive mechanism for mobile crowdsourcing

Truthful incentive mechanisms for mobile crowd sensing with dynamic smartphones