Kentaroh Toyoda, Jun Zhao, Allan N. Zhang, P. Takis Mathiopoulos
Federated learning (FL) is a promising decentralized deep learning technique that allows users to collaboratively update models without sharing their own data. However, due to its decentralized nature, no one can monitor workers' behavior, and they may thus deviate protocols (e.g., participating without updating any models). To solve this problem, many researchers have proposed blockchain-enabled FL to reward workers (or users) with cryptocurrencies to encourage workers to follow the protocols. However, there is a lack of theoretical discussions concerning how such rewards impact workers' behavior and how much should be given to workers. In this article, we propose a mechanism-design-oriented FL protocol on a public blockchain network. Mechanism design (MD) is often used to make a rule intended to achieve a specific goal. With MD in mind, we introduce the concept of competition into blockchain-based FL so that only workers who have contributed well can obtain rewards, which naturally prevents workers from deviating from the protocol. We then mathematically answer the following questions with contest theory, a novel field of study in economics: i) What behavior will workers take?; ii) how much effort should workers exert to maximize their profits?; iii) how many workers should be rewarded?; and iv) what is the best proportion for reward distribution?
Maha Kadadha, Rabeb Mizouni, Shakti Singh, Hadi Otrok · 5 authors
In this paper, a fully distributed auction-blockchain-based crowdsourcing framework is proposed-ABCrowd. In a typical crowdsourcing framework, independent workers compete to be allocated requesters' tasks. These workers advertise their costs to the centralized platform, which then decides the final allocation of tasks. While performing the allocation, centralized platforms face two main challenges: 1) how to ensure trusted execution for the allocation of tasks, and 2) how to motivate workers to declare their truthful costs. To address these challenges, ABCrowd proposes to run the crowdsourcing platform entirely on Ethereum Blockchain while incorporating auctions. Blockchain and smart contracts guarantee trusted execution for the allocation through autonomous and transparent on-Chain execution. ABCrowd uses the Repeated-Single-Minded Bidder (R-SMB) auction mechanism, which motivates workers to bid truthfully before allocating them and calculating their payments. R-SMB is an approximation of the optimized off-Chain Vickrey-Clarke-Groves (VCG) mechanism in terms of maximized profit. It entails repeating the Single-Minded Bidder (SMB) auction mechanism to meet the allocation requirement of crowdsourcing applications. ABCrowd is implemented and evaluated using Solidity on a private Ethereum Blockchain, where a real publicly available dataset is used. The proposed on-Chain R-SMB auction mechanism is compared to the off-Chain VCG mechanism, where the results show that R-SMB provides similar performance to VCG in terms of the average number of allocated tasks. Furthermore, R-SMB outperforms VCG in workers' travelled distance and requesters' costs, at a low execution cost.
Streaming media has been largely used by millions of users every day. The number of customers and programs, e.g., TV series, movies, and various shows, are still growing fast. However, the demand for video transcoding for various personal terminal devices results in the shortage of computing resources and the prolongation of processing delay in centralized video transcoding systems. To solve this issue, we propose a blockchain, especially, smart contract based scheme that can achieve decentralized and on-demand crowdsourcing for video transcoding, which remarkably mitigates the transcoding overhead. Specifically, our scheme consists of four key components such as employers, workers, task allocation, and payment. An employer initializes the smart contract, releases the task, and initiates the smart contract. Workers bid for the task, and the successful bidder will obtain the task and execute the task. The task allocation mechanism and the payment mechanism can guarantee the profits of both and encourage both as well. Moreover, the smart contract consists of the bidding contract and the task execution contract. The extensive analysis of our proposed scheme justified the feasibility, security for defending against typical threats, applicability in realistic situations, and portability for most multimedia such as videos and audios.
Due to the long train marshaling and complex line conditions, the operating modes in heavy haul rail systems frequently change when trains travel. Improper traction or braking operation made by drivers will increase the longitudinal impact force to trains and causes the train decoupling, severely affecting the safe operations of trains. It is quite desirable to replace the manual control with intelligent control in heavy haul rail systems. Traditional machine learning-based intelligent control methods suffer from insufficient data. Due to lacking effective incentives and trust, data from different rail lines or operators cannot be shared directly. In this paper, we propose an approach on blockchain-based federated learning to implement asynchronous collaborative machine learning between distributed agents that own data. This method performs distributed machine learning without a trusted central server. The blockchain smart contract is used to realize the management of the entire federated learning. Using the historical driving data collected from real heavy haul rail systems, the learning agent in the federated learning method adopts a support vector machine (SVM) based intelligent control model. To deal with the imbalanced traction and braking data, we optimize the classic SVM model via assigning different penalty factors to the majority and minority classes. The data set are mapped to a high dimension using kernel functions to make it linearly separable. We construct a mixing kernel function composed of polynomial and radial basis function (RBF) kernel functions, which uses a dynamic weight factor changing with train speeds to improve the model accuracy. The simulation results demonstrate the efficiency and accuracy of our proposed intelligent control method.
La technologie des registres distribués blockchain a perturbé et révolutionné le monde en introduisant un système transactionnel décentralisé traditionnel permettant l'échange de devises et de données de manière sécurisée. Le suivi de la chaîne d'approvisionnement est un domaine qui pourrait grandement bénéficier de cette avancée. Actuellement, les chaînes d'approvisionnement centralisées souffrent d'un manque de transparence et de contrôle des stocks pour vérifier la qualité des produits traités et éviter des problèmes tels que la contrefaçon. Le processus de gestion de la chaîne d'approvisionnement traditionnel insuffisant pour localiser un produit tout au long du processus de développement et de transport et conduit à la nécessité de réorganiser la manière de gérer la traçabilité des biens de consommation. Cette thèse tente d'évaluer la faisabilité de l'utilisation d'une chaîne d'approvisionnement avec un système de registres distributés IOTA pour des applications IoT. Pour valider le concept, une chaîne logistique pharmaceutique est mise en œuvre pour obtenir des données de capteurs de température, d'humidité et GNSS à partir d'un microprocesseur à faible puissance et à faible coût pour stocker et envoyer de manière fiable et sécurisée des données et les envoyer à une nouvelle structure de chaîne de blocs. Nous en faisons la démonstration avec un CC2650 SensorTag en tant que dispositif de systèmes intégrés, le X-NUCLEO-GNSS1A1 pour le module de positionnement et un Raspberry Pi 3, proxy IOTA, pour mesurer le temps, l’énergie et l’efficacité de la création de transactions et le calcul du \guillemotleft proof of work \guillemotright. Nous analysons également les résultats de l'expérience et discutons des avantages et des inconvénients de la mise en œuvre
Medication quality and safety are crucial to the health of the public. Responding to the urgent need for medication information provenance and anti-counterfeiting, this study proposes a blockchain based method for medication information storage, inquiry, and anti-counterfeiting along a medication supply chain. Leveraging the features of decentralization, tamper-proof, traceability, and participative node maintenance of blockchain technology, the proposed method can assure the transparency and openness of medication supply chains. An access control policy model based on smart contract is designed to prevent medication information from being altered or disclosed at nodes of the blockchain. In addition, a point-accumulation upgrade/downgrade mechanism is introduced to improve the consensus mechanism. The proposed solution eliminates the needs for centralized institutions and third-party organizations, and provides a full record of the medication circulation process. Our simulation results show that efficiency and security are enhanced by the improved consensus algorithm and access control mechanism. As a result, our method can render high level of security and privacy protection that is critical to the integrity of a medication information management system.
Participatory sensing is gaining popularity as a method for collecting and sharing information from distributed local environments using sensor-rich mobile devices. There are a number of participatory sensing applications currently in wide use, such as location-based service applications (e.g., Waze navigation). Usually, these participatory applications collect tremendous amounts of sensing data containing personal information, including user identity and current location. Due to the high sensitivity of this information, participatory sensing applications need a privacy-preserving mechanism, such as anonymity, to secure and protect personal user data. However, using anonymous identifiers for sensing sources proves difficult when evaluating sensing data trustworthiness. From this perspective, a successful participatory sensing application must be designed to consider two challenges: (1) user privacy and (2) data trustworthiness. To date, a number of privacy-preserving reputation techniques have been proposed to satisfy both of these issues, but the protocols contain several critical drawbacks or are impractical in terms of implementation. In particular, there is no work that can transparently manage user reputation values while also tracing anonymous identities. In this work, we present a blockchain-based privacy-preserving reputation framework called BPRF to transparently manage user reputation values and provide a transparent tracing process for anonymous identities. The performance evaluation and security analysis show that our solution is both practical and able to satisfy the two requirements for user privacy and data trustworthiness.
Summary Smart cities aim to provide smart governance with the emphasis on gaining high transparency and trust in public services and enabling citizen participation in decision making processes. This means on the one hand data generated from urban transactions need to be open and trustworthy. On the other hand, security and privacy of public data needs to be handled at different administrative and geographical levels. In this paper, we investigate the pivotal role of blockchain in providing privacy, self‐verification, authentication, and authorization of participatory transactions in open governance. We also investigate up to what extent edge computing can contribute toward management of permissioned sharing at specific administrative levels and enhance privacy and provide an economic approach for resource utilization in a distributed environment. We introduce a novel architecture that is based on distributed hybrid ledger and edge computing model. The architecture provides refined and secure management of data generated and processed in different geographical and administrative units of a city. We implemented a proof of concept of the architecture and applied it on a carefully designed use case, ie, citizen participation in administrative decisions through consensus. This use case highlights the need to keep and process citizen participation data at local level by deploying district chaincodes and only share consensus results through permissioned chaincodes. The results reveal that proposed architecture is scalable and provide secure and privacy protected environment for citizen participatory applications. Our performance test results are promising and show that under control conditions, the average registration time for a citizen transaction is about 42 ms, whilst the validation and result compilation of 100 concurrent citizens' transactions took about 2.4 seconds.
Recently, crowd sensing has been intensively researched, due to the rapid growth of sensor-integrated mobile devices. Crowd sensing is a participatory sensing service where a server gathers and analyzes sensing data submitted from mobile devices of lots of users. In crowd sensing, the user's anonymity is desired, since the server gathers sensitive data from the participants including their GPS locations and moving path. However, the anonymous data submission may compromise the trust of the sensing data, because anonymous users may submit inappropriate sensing data, but they cannot be traced. Therefore, as the system to achieve both anonymity and trust in crowd sensing, ARTSense has been proposed. In the system, the trust of the sensing data is assessed on the sensed environment, other users' sensing, and the reputation of the user, and furthermore the reputation of the user is anonymously managed on the feedback from the trust assessment for the data. However, the anonymous reputation system of ARTSense has the efficiency problem, i.e., the user needs to wait a random time after the data submission phase before requesting the reputation update, which causes the communication delay. In this paper, we propose an efficient anonymous reputation system for crowd sensing, which can be integrated to the trust assessment in ARTSense. In the proposed system, during the data submission, the reputation update is anonymously completed. This is because the server does not manage the reputation of each user, but each user manages his/her reputation in the user side, where the the validity of the reputation is ensured by a certificate and anonymously checked by zero-knowledge proofs. Therefore, the proposed system achieves the better efficiency with no delay.
With the rapid development of Internet of Things (IoT), mobile crowdsensing (MCS), i.e., outsourcing sensing tasks to mobile devices or vehicles, has been proposed to address the problem of data collection in the scenarios such as smart city. Despite its benefits for a wide range of applications, MCS lacks an efficient incentive mechanism, restricting the development of IoT applications, especially for Internet of Vehicles (IoV)-a typical example of IoT applications; this is because vehicles are usually reluctant to participate these sensing tasks. Moreover, in practice, some sensing tasks may arrive suddenly (called an emergent task) in the IoV environment, but the resources of a single vehicle may be insufficient to handle, and thus multivehicles collaboration is required. In this case, the incentive mechanisms for the participation of multiple vehicles and the task scheduling for their collaborations are collectively needed. To address this important problem, we first propose a new model for the scenario of two vehicles collaboration, considering the situation of the emergent appearance of a task. In this model, for a general sensing task, we propose a bidding mechanism to better encourage vehicles to contribute their resources, and the tasks for those vehicles are scheduled accordingly. Second, for an emergent task, a novel time-window-based method is devised to manage the tasks among vehicles and to incent the vehicles to participate. Finally, we develop a blockchain framework to achieve the secured information exchange through smart contract for the proposed models in IoV.
Dimitrios G. Kogias, Helen C. Leligou, Michael G. Xevgenis, Maria Polychronaki · 8 authors
Crowdsourcing has been pursued as a way to leverage the power of the crowd for many different purposes in diverse sectors from collecting information, aggregating funds, and gathering employees to perform tasks of different sizes among other targets. Data integrity and nonrepudiation are of utmost importance in these systems and are currently not guaranteed. Blockchain technology has been proven to improve on these aspects. In this article, we investigate the benefits that the adoption of Blockchain technology can bring in crowdsourcing systems. To this end, we provide examples of real-life crowdsourcing use cases and explore the benefits of using Blockchain, mainly as a database.
Naman Goel, Cyril van Schreven, Aris Filos-Ratsikas, Boi Faltings
Blockchain based systems allow various kinds of financial transactions to be executed in a decentralized manner. However, these systems often rely on a trusted third party (oracle) to get correct information about the real-world events, which trigger the financial transactions. In this paper, we identify two biggest challenges in building decentralized, trustless and transparent oracles. The first challenge is acquiring correct information about the real-world events without relying on a trusted information provider. We show how a peer-consistency incentive mechanism can be used to acquire truthful information from an untrusted and self-interested crowd, even when the crowd has outside incentives to provide wrong informations. The second is a system design and implementation challenge. For the first time, we show how to implement a trustless and transparent oracle in Ethereum. We discuss various non-trivial issues that arise in implementing peer-consistency mechanisms in Ethereum, suggest several optimizations to reduce gas cost and provide empirical analysis.
Dinh C. Nguyen, Pubudu N. Pathirana, Ming Ding, Aruna Seneviratne
\n\t\t\t\t\tBlockchain technology with its secure, transparent and decentralized nature has been recently employed in many mobile applications. However, the process of executing extensive tasks such as computation-intensive data applications and blockchain mining requires high computational and storage capability of mobile devices, which would hinder blockchain applications in mobile systems. To meet this challenge, we propose a mobile edge computing (MEC) based blockchain network where multi-mobile users (MUs) act as miners to offload their data processing tasks and mining tasks to a nearby MEC server via wireless channels. Specially, we formulate task offloading, user privacy preservation and mining profit as a joint optimization problem which is modelled as a Markov decision process, where our objective is to minimize the long-term system offloading utility and maximize the privacy levels for all blockchain users. We first propose a reinforcement learning (RL)-based offloading scheme which enables MUs to make optimal offloading decisions based on blockchain transaction states, wireless channel qualities between MUs and MEC server and user’s power hash states. To further improve the offloading performances for larger-scale blockchain scenarios, we then develop a deep RL algorithm by using deep Q-network which can efficiently solve large state space without any prior knowledge of the system dynamics. Experiment and simulation results show that the proposed RL-based offloading schemes significantly enhance user privacy, and reduce the energy consumption as well as computation latency with minimum offloading costs in comparison with the benchmark offloading schemes.\n\t\t\t\t
Jernej Mihelj, Yuan Zhang, Andrej Kos, Urban Sedlar
Real-time data about various traffic events and conditions-offences, accidents, dangerous driving, or dangerous road conditions-is crucial for safe and efficient transportation. Unlike roadside infrastructure data which are often limited in scope and quantity, crowdsensing approaches promise much broader and comprehensive coverage of traffic events. However, to ensure safe and efficient traffic operation, assessing trustworthiness of crowdsourced data is of crucial importance; this also includes detection of intentional or unintentional manipulation, deception, and spamming. In this paper, we design and demonstrate a road traffic event detection and source reputation assessment system for unreliable data sources. Special care is taken to adapt the system for operation in decentralized mode, using smart contracts on a Turing-complete blockchain platform, eliminating single authority over such systems and increasing resilience to institutional data manipulation. The proposed solution was evaluated using both a synthetic traffic event dataset and a dataset gathered from real users, using a traffic event reporting mobile application in a professional driving simulator used for driver training. The results show the proposed system can accurately detect a range of manipulative and misreporting behaviors, and quickly converges to the final trust score even in a resource-constrained environment of a blockchain platform virtual machine.
Yury Zhauniarovich, Yazan Boshmaf, Husam Al Jawaheri, Mashael Al Sabah
Web-based hosting services for version control, such as GitHub, have made it easier for people to develop, share, and donate money to software repositories. In this paper, we study the use of Bitcoin to make donations to open source repositories on GitHub. In particular, we analyze the amount and volume of donations over time, in addition to its relationship to the age and popularity of a repository. We scanned over three million repositories looking for donation addresses. We then extracted and analyzed their transactions from Bitcoin's public blockchain. Overall, we found a limited adoption of Bitcoin as a payment method for receiving donations, with nearly 44 thousand deposits adding up to only 8.3 million dollars in the last 10 years. We also found weak positive correlation between the amount of donations in dollars and the popularity of a repository, with highest correlation (r=0.013) associated with number of forks.
The crowdsourcing schemes which utilize the social network to solve complex tasks are an important part of open cooperation over the Internet. Although blockchain-based crowdsourcing schemes have considerable advantages in decentralization and data sharing, there is still a challenge to gurantee the security of crowdsourced-sensitive information and the fairness of crowdsourcing on the blockchain. To this end, this article investigates a crowdsourcing scheme based on blockchain. First, we define the basic requirements of blockchain-based crowdsourcing schemes including fairness, confidentiality, and integrity. And then, using secure hash, commitment, and homomorphic encryption, we propose a blockchain-based secure and fair crowdsourcing scheme, that is, BFC. The analysis results show that our scheme can satisfy the above requirements. Finally, the experimental results show that the computational overhead of the BFC scheme is acceptable to both the requester and the workers. In a word, our proposed crowdsourcing scheme has good expansibility in reality.
Machine learning has recently enabled large advances in artificial intelligence, but these tend to be highly centralized. The large datasets required are generally proprietary; predictions are often sold on a per-query basis; and published models can quickly become out of date without effort to acquire more data and re-train them. We propose a framework for participants to collaboratively build a dataset and use smart contracts to host a continuously updated model. This model will be shared publicly on a blockchain where it can be free to use for inference. Ideal learning problems include scenarios where a model is used many times for similar input such as personal assistants, playing games, recommender systems, etc. In order to maintain the model's accuracy with respect to some test set we propose both financial and non-financial (gamified) incentive structures for providing good data. A free and open source implementation for the Ethereum blockchain is provided at https://github.com/microsoft/0xDeCA10B.
Yang Zhao, Jun Zhao, Linshan Jiang, Rui Tan · 5 authors
Internet-of-Things (IoT) companies strive to get feedback from users to improve their products and services. However, traditional surveys cannot reflect the actual conditions of customers' due to the limited questions. Besides, survey results are affected by various subjective factors. In contrast, the recorded usages of IoT devices reflect customers' behaviours more comprehensively and accurately. We design an intelligent system to help IoT device manufacturers to take advantage of customers' data and build a machine learning model to predict customers' requirements and possible consumption behaviours with federated learning (FL) technology. The FL consists of two stages: in the first stage, customers train the initial model using the phone and the edge computing server collaboratively. The mobile edge computing server's high computation power can assist customers' training locally. Customers first collect data from various IoT devices using phones, and then download and train the initial model with their data. During the training, customers first extract features using their mobiles, and then add the Laplacian noise to the extracted features based on differential privacy, a formal and popular notion to quantify privacy. After achieving the local model, customers sign on their models respectively and send them to the blockchain. We use the blockchain to replace the centralized aggregator which belongs to the third party in FL. In the second stage, miners calculate the averaged model using the collected models sent from customers. By the end of the crowdsourcing job, one of the miners, who is selected as the temporary leader, uploads the model to the blockchain. Besides, to attract more customers to participate in the crowdsourcing FL, we design an incentive mechanism, which awards participants with coins that can be used to purchase other services provided by the company.
Yang Zhao, Jun Zhao, Linshan Jiang, Rui Tan · 8 authors
Home appliance manufacturers strive to obtain feedback from users to improve their products and services to build a smart home system. To help manufacturers develop a smart home system, we design a federated learning (FL) system leveraging the reputation mechanism to assist home appliance manufacturers to train a machine learning model based on customers' data. Then, manufacturers can predict customers' requirements and consumption behaviors in the future. The working flow of the system includes two stages: in the first stage, customers train the initial model provided by the manufacturer using both the mobile phone and the mobile edge computing (MEC) server. Customers collect data from various home appliances using phones, and then they download and train the initial model with their local data. After deriving local models, customers sign on their models and send them to the blockchain. In case customers or manufacturers are malicious, we use the blockchain to replace the centralized aggregator in the traditional FL system. Since records on the blockchain are untampered, malicious customers or manufacturers' activities are traceable. In the second stage, manufacturers select customers or organizations as miners for calculating the averaged model using received models from customers. By the end of the crowdsourcing task, one of the miners, who is selected as the temporary leader, uploads the model to the blockchain. To protect customers' privacy and improve the test accuracy, we enforce differential privacy on the extracted features and propose a new normalization technique. We experimentally demonstrate that our normalization technique outperforms batch normalization when features are under differential privacy protection. In addition, to attract more customers to participate in the crowdsourcing FL task, we design an incentive mechanism to award participants.
Abstract In this paper, the authors propose to increase the efficiency of blockchain mining by using a population-based approach. Blockchain relies on solving difficult mathematical problems as proof-of-work within a network before blocks are added to the chain. Brute force approach, advocated by some as the fastest algorithm for solving partial hash collisions and implemented in Bitcoin blockchain, implies exhaustive, sequential search. It involves incrementing the nonce (number) of the header by one, then taking a double SHA-256 hash at each instance and comparing it with a target value to ascertain if lower than that target. It excessively consumes both time and power. In this paper, the authors, therefore, suggest using an inner for-loop for the population-based approach. Comparison shows that it’s a slightly faster approach than brute force, with an average speed advantage of about 1.67% or 3,420 iterations per second and 73% of the time performing better. Also, we observed that the more the total particles deployed, the better the performance until a pivotal point. Furthermore, a recommendation on taming the excessive use of power by networks, like Bitcoin’s, by using penalty by consensus is suggested.
Distributed Ledger Technologies (DLT), of which Blockchain is a popular example, are increasingly becoming an integral feature of many modern systems.While cryptocurrencies are a common motivating example, and they drive systems such as Bitcoin, DLT has many other uses across industries including Health Care, Supply Chain, IoT, and Finance among others.One of the key concepts that make DLT appealing is the ability for large-scale systems that do not trust each other to reach consensus and share a commonly verifi able ledger, in order to track resources, changes to system-wide data, and other artifacts.Cryptocurrencies are one type of resource that can be tracked, but DLT has been used for energy, pharmaceuticals, and many other domains.
Blockchains - with their inherent properties of transaction transparency, distributed consensus, immutability and cryptographic verifiability - are increasingly seen as a means to underpin innovative products and services in a range of sectors from finance through to energy and healthcare. Discussions, too often, make assertions that the trustless nature of blockchain technologies enables and actively promotes their suitability - there being no need to trust third parties or centralised control. Yet humans need to be able to trust systems, and others with whom the system enables transactions. In this paper, we highlight that understanding this need for trust is critical for the development of blockchain-based systems. Through an online study with 125 users of the most well-known of blockchain based systems - the cryptocurrency Bitcoin - we uncover that human and institutional aspects of trust are pervasive. Our analysis highlights that, when designing future blockchain-based technologies, we ought to not only consider computational trust but also the wider eco-system, how trust plays a part in users engaging/disengaging with such eco-systems and where design choices impact upon trust. From this, we distill a set of guidelines for software engineers developing blockchain-based systems for societal applications.
Blockchain technology has found a great number of applications, from banking to the Internet of Things (IoT). However, it has not yet been envisioned whether and which problems in Software Engineering (SE) Blockchain technology could solve. In this paper, we coin this field "Blockchain-based Software Engineering" and exemplify how Blockchain technology could solve two core SE problems: Continuous Integration (CI) Services such as Travis CI and Package Managers such as apt-get. We believe that Blockchain technology could help (1) democratize and professionalize Software Engineering infrastructure that currently relies on free work done by few volunteers, (2) improve the quality of artifacts and services, and (3) increase trust in ubiquitously used systems like GitHub or Travis CI.