Article

FBI: A Federated Learning-Based Blockchain-Embedded Data Accumulation Scheme Using Drones for Internet of Things

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

This letter presents a federated learning-based data-accumulation scheme that combines drones and blockchain for remote regions where Internet of Things devices face network scarcity and potential cyber threats. The scheme contains a two-phase authentication mechanism in which requests are first validated using a cuckoo filter, followed by a timestamp nonce. Secure accumulation is achieved by validating models using a Hampel filter and loss checks. To increase the privacy of the model, differential privacy is employed before sharing. Finally, the model is stored in the blockchain after consent is obtained from mining nodes. Experiments are performed in a proper environment, and the results confirm the feasibility of the proposed scheme.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Islam et al. [38] proposed a federated learning-based data accumulation solution. This latter combines drones and blockchain technology. ...
... AC Architecture AC Nature [27] Policy based Partially D [34] Policy based Partially D [35] Token based Partially D [36] Policy based Partially D [37] Policy based Partially D [38] Token and cryptography based Partially D [57] Policy and token based Fully D [40] Policy based Fully D [41] Policy based Fully D [42] Policy based Fully D [43] Token based Fully D [44] Cryptography based Fully D [46] Policy based Fully D [47] Cryptography and policy based Fully D [48] Policy and Permission delegation based Fully D 7. Cr7: Access control models A policy-based authorization and access control model is necessary to encapsulate security policies. Analyzing the policy-based authorization solutions presented recently (summarized in Table 7) showed that a large number of these solutions adopted the ABAC model [37,41,42,46,56,68,69]. ...
... Gupta et al. [73] proposed a game theory-based authentication framework with blockchain technology to resolve the Internet of Vehicles (IoV) cross trusted authority's authentication issues. To manage the network scarcity challenges, Islam et al. [38] proposed a lightweight scheme employing drones to assist IoT devices for secure data collection. Furthermore, dew computing was used to permit offline computations. ...
Article
Full-text available
The Internet of Things is gaining more importance in the present era of Internet technology. It is considered as one of the most important technologies of everyday life. Moreover, IoT systems are ceaselessly growing with more and more devices. They are scalable, dynamic, and distributed, hence the origin of the crucial security requirements in IoT. One of the most challenging issues that the IoT community must handle recently is how to ensure an access control approach that manages the security requirements of such a system. Traditional access control technologies are not suitable for a large-scale and distributed network structure. Most of them are based on a centralized approach, where the use of a trusted third party (TTP) is obligatory. Furthermore, the emergence of blockchain technology has allowed researchers to come up with a solution for these security issues. This technology is highly used to record access control data. Additionally, it has great potential for managing access control requests. This paper proposed a blockchain-based access control taxonomy according to the access control nature: partially decentralized and fully decentralized. Furthermore, it presents an overview of blockchain-based access control solutions proposed in different IoT applications. Finally, the article analyzes the proposed works according to certain criteria that the authors deem important.
... As the IoT evolves, there is a growing need for new approaches to addressing IoT security, privacy, and scalability challenges. The authors in [55] introduced a federated learning-based Blockchain-embedded data accumulation scheme for remote areas where IoT devices encounter network supply shortages and potential cyberattacks. The proposed model consisted of a two-authentication process that validates requests first with a cuckoo filter, then with a timestamp nonce. ...
... In [77], a Reduced Early Handover (REHO) technique was suggested to minimize both the ping-pongs and RLFs, achieving high energy efficiency while maintaining other performance parameters within appropriate limits. The finding in [55] also led to the outcomes of [56] where a fuzzy multiple criteria cell selection technique was used. This scheme considers the UE uplink conditions, resource block allocation, and selection criteria of the LTE's conventional cell selection approach. ...
Article
Full-text available
Drones have attracted extensive attention for their environmental, civil, and military applications. Because of their low cost and flexibility in deployment, drones with communication capabilities are expected to play key important roles in Fifth Generation (5G), Sixth Generation (6G) mobile networks, and beyond. 6G and 5G are intended to be a full-coverage network capable of providing ubiquitous connections for space, air, ground, and underwater applications. Drones can provide airborne communication in a variety of cases, including as Aerial Base Stations (ABSs) for ground users, relays to link isolated nodes, and mobile users in wireless networks. However, variables such as the drone’s free-space propagation behavior at high altitudes and its exposure to antenna sidelobes can contribute to radio environment alterations. These differences may render existing mobility models and techniques as inefficient for connected drone applications. Therefore, drone connections may experience significant issues due to limited power, packet loss, high network congestion, and/or high movement speeds. More issues, such as frequent handovers, may emerge due to erroneous transmissions from limited coverage areas in drone networks. Therefore, the deployments of drones in future mobile networks, including 5G and 6G networks, will face a critical technical issue related to mobility and handover processes due to the main differences in drones’ characterizations. Therefore, drone networks require more efficient mobility and handover techniques to continuously maintain stable and reliable connection. More advanced mobility techniques and system reconfiguration are essential, in addition to an alternative framework to handle data transmission. This paper reviews numerous studies on handover management for connected drones in mobile communication networks. The work contributes to providing a more focused review of drone networks, mobility management for drones, and related works in the literature. The main challenges facing the implementation of connected drones are highlighted, especially those related to mobility management, in more detail. The analysis and discussion of this study indicates that, by adopting intelligent handover schemes that utilizing machine learning, deep learning, and automatic robust processes, the handover problems and related issues can be reduced significantly as compared to traditional techniques.
... Other characteristics, such as security or mobility, are specific properties for the underlaying systems. the edge, so protecting the point of decision as other Federated Learning [15] implementations are trying to achieve. The embedded device/module from the glove was improved for the motion commands by a small neuronal network applied to the values processed by the glove. ...
... When discussing IoT and performing on edge simulation, the problem of data acquisition is a stringent matter from a security perspective, as shown in [15]. ...
Article
Full-text available
This paper presents edge machine learning (ML) technology and the challenges of its implementation into various proof-of-concept solutions developed by the authors. Paper presents the concept of Edge ML from a variety of perspectives, describing different implementations such as: a tech-glove smart device (IoT embedded device) for controlling teleoperated robots or an UAVs (unmanned aerial vehicles/drones) that is processing data locally (at the device level) using machine learning techniques and artificial intelligence neural networks (deep learning algorithms), to make decisions without interrogating the cloud platforms. Implementation challenges used in Edge ML are described and analyzed in comparisons with other solutions. An IoT embedded device integrated into a tech glove, which controls a teleoperated robot, is used to run the AI neural network inference. The neural network was trained in an ML cloud for better control. Implementation developments, behind the UAV device capable of visual computation using machine learning, are presented.
... Implementation Evaluated metrics [6] Java Processing time [47] NS-3 Authentication time, block size [48] Cooja, NS-3 Consensus processing time, time overhead, energy consumption, packet overhead [49] Cooja & NS-3 Processing time, energy consumption, packet overhead [50] N/A Computational time, communication cost [51] Multichain Authentication service execution time, block transmission rate, block validation delay [52] Multichain Authentication requests execution time [53] Java Block generation time, energy consumption [54] C/C++ Time per transaction, private keys distribution time, network supervision time [55] Matlab Routing latency, traffic of swarm of UAS networking [57] NS-3 Processing time, data transferred [58] N/A Processing time, transactions per second, package overhead [56] Hyperledger Throughput, transaction latency, communication time, block generation time [59] Python Consensus delay, blockchain size [60] ZeroCaloSimu Bandwidth consumption, transactions per second [61] BlockLite Block generation time, mining time per block, blocks per second [62] iOS Swift ...
Article
Full-text available
The Internet of Things (IoT) has become an essential part of our society. IoT devices are used in our houses, hospitals, cars, industry, etc., making our lives easier. Nonetheless, there are a number of serious concerns about security, privacy and performance issues in IoT. It has been proven that the aforementioned issues are strictly related to the high degree of centralisation of current IoT architecture. Thus, there is an increasing interest in adopting blockchain in IoT. However, blockchain adoption is not straightforward due to the power, storage and computational limitations of IoT. Consequently, the concept of lightweight blockchain is getting more and more attention from researchers and engineers. In this paper, we conduct a systematic literature review on the lightweight blockchain concept for IoT following the PRISMA methodology. We systematically analyse "lightweight blockchain for IoT" proposals in order to better understand the limitations of blockchain for IoT, the characteristics of the current work on this topic and further research opportunities. Specifically, we analyse the definition of lightweight blockchain that other authors give, the characteristics of the reviewed proposals, their "lightweight" aspects and their evaluation. Finally, we discuss the results of the review along with further research opportunities. Consequently, this work is mostly focused on understanding the technical and performance-related aspects of blockchain for IoT as a prelude to more specific analysis such as security (i.e., attacks, vulnerabilities, etc.).
... The author's implemented novelty relies on monitoring pandemic outbreaks, with an additional two-phase lightweight security mechanism being adapted for authorization purposes. Another most latest work, in [110], mentions federated learningbased blockchain intervention for authentication purposes with strong differential-based privacy presence. Finally, au-thors in [109] invest more in the Privacy concerns of the Blockchain with the suggestion of a privacy-enhancing content erasure mechanism that aims to increase the anonymization concept of the ledger. ...
Article
Full-text available
With the faster maturity and stability of digitization, connectivity and edge technologies, the number of the Internet of Things (IoT) devices and sensors is flourishing fast in important junctions such as homes, hotels, hospitals, retail stores, manufacturing floors, railway stations, airports, oil wells, warehouses, etc. However, in this extremely connected world, the security implications for IoT devices are getting worse with the constant rise in malicious cyberattacks. The challenge is how to secure IoT sensors, services and data. The blockchain technology, a prominent distributed ledger technology (DLT), is being pronounced as the way forward for safeguarding IoT devices and data. The Directed Acyclic Graph (DAG)-based DLT has the inherent potential to realize the benefits of blockchain with better performance. IOTA is a DAG-based blockchain implementation for the IoT era. The Tangle, the IOTA’s network immutably records the exchange of data and value. It ensures that the information is trustworthy and cannot be tampered with nor destroyed. In this work, we depict a thorough analysis of the existing security studies for IOTA. Then, we identify the gaps and the limitations of these security solution schemes, and finally, propose future security research recommendations that can potentially fill these gaps to secure DLT-enabled IoT devices.
... The Internet of Things (IoT) network architecture is evolving rapidly to cover various fields and applications [3]. UAVs have been deployed as air-ground equipment to address processing and storage requirements at the IoT networks [4]. However, there are substantial disadvantages to using UAVs, such as their inability to fly in inclement weather and the controller's requirement for visual line of sight (LOS) [5]. ...
Article
Full-text available
Unmanned aerial vehicles (UAVs) extend the traditional ground-based Internet of Things (IoT) into the air. UAV mobile edge computing (MEC) architectures have been proposed by integrating UAVs into MEC networks during the current novel coronavirus disease (COVID-19) era. UAV mobile edge computing (MEC) shares personal data with external parties (such as edge servers) during intelligent medical analytics. However, this technique raises privacy concerns about patients’ health data. More recently, the concept of federal learning (FL) has been set up to protect mobile user data privacy. Compared to traditional machine learning, federated learning requires a decentralized distribution system to enhance trust for UAVs. Blockchain technology provides a secure and reliable solution for FL settings between multiple untrusted parties with anonymous, immutable, and distributed features. Therefore, blockchain-enabled FL provides both theories and techniques to improve the performance of intelligent UAV edge computing networks from various perspectives. This survey begins by discussing the current state of research on blockchain and FL. Then, compare the leading technologies and limitations. Second, we will discuss how to integrate blockchain and FL into UAV edge computing networks and the associated challenges and solutions. Finally, we discuss the fundamental research challenges and future directions.
... Since the battery capacity of the UAV is limited, it is critical to research how to reduce the UAV's energy consumption so that the wireless network's connection time may be extended. In some of the research works, the authors assumed that the UAV has sufficient energy for completing the given task for secure data collection [15,16] whereas in [17] reduce the energy usage of the communication by finding an optimized trajectory path using successive convex approximation technique for a mobile relay network [18]. In [19] the placement of the UAV optimization and performance analysis was studied to understand various characteristics of UAV enabled base station. ...
Article
Full-text available
The water distribution system has deployed several low-power IoT devices on an uneven surface where battery power is a major concern. Therefore, this paper focuses on using a UAV-enabled wireless powered communication network capable of directing energy to a target location and using it for communication, thereby reducing battery issues. In this paper, a static optimization was applied to find the initial height values using 3D clustering and beamforming method and dynamic optimization using extremum seeking method was applied to find the optimized height. The optimized height values were calculated and Travelling Salesman Problem (TSP) was applied to create the trajectory of the UAV. The overall energy consumption of the UAV was minimized by integrating dynamic optimization and dome packing method, which can find an optimal position and trajectory where the UAV will be hovering to direct energy and collect data. Moreover, we also minimized the total flight time of the UAV.
... Blockchain as a Smart City Solution [5] Authentication and trust management [6,7] Healthcare [8] Waste management [9] IoT sensors [10][11][12][13][14] Drone applications for Smart City surveillance [14,15] Smart contracts [16] Online insurance [17] Legal and technical adoption issues for Smart City [18,19] Blockchain oracles problem [20][21][22][23] Blockchain consensus type overview [24][25][26][27][28][29][30][31] Security of data sent by oracles [32][33][34][35] Interplanetary File System for data storing [36][37][38] Smart City interoperability [39][40][41][42] Data consolidation and GDPR Thus, the contribution of this paper is a presentation of the whole topic of the digital transformation of a city and the peculiarities of possible application of blockchain in each instance. In doing so, it was beneficial to collaborate with the City of Osijek, with a goal to define the digitalization strategy document as the basis for later implementation. ...
Article
Full-text available
With advances in Information and Communication Technologies (ICT) in convergence with blockchain technology, cities have been given the opportunity to improve their services, efficiently use resources, and, thus, become Smart Cities. The main properties of blockchain technology like decentralization, immutability, transparency, consensus, and robustness are qualities needed for Smart City. In this paper, we propose a digitalization strategy for the City of Osijek. Smart City digitalization strategy aims to solve problems of emerging urbanization, improve administration by reducing energy and water consumption, carbon emissions, pollution, and city waste management. To develop an information system based on blockchain technology, the administration structure and the current state of information systems are analyzed, and new solutions are presented.
... Early research focused on using a blockchain network as an external component of systems that are mainly used as databases for key storage and WSN management [24,25]. Blockchain networks have been used in combination with WSNs for data security, sensor node authentication [26], removing single points of failure in WSNs [27], and secure data accumulation [28]. ...
Article
Full-text available
Currently, the computational power present in the sensors forming a wireless sensor network (WSN) allows for implementing most of the data processing and analysis directly on the sensors in a decentralized way. This shift in paradigm introduces a shift in the privacy and security problems that need to be addressed. While a decentralized implementation avoids the single point of failure problem that typically applies to centralized approaches, it is subject to other threats, such as external monitoring, and new challenges, such as the complexity of providing decentralized implementations for data mining algorithms. In this paper, we present a solution for privacy-aware distributed data mining on wireless sensor networks. Our solution uses a permissioned blockchain to avoid a single point of failure in the system. Contracts are used to construct an onion-like structure encompassing the Hoeffding trees and a route. The onion-routed query conceals the network identity of the sensors from external adversaries, and obfuscates the actual computation to hide it from internally compromised nodes. We validate our solution on a use case related to an air quality-monitoring sensor network. We compare the quality of our model against traditional models to support the feasibility and viability of the solution.
... Similarly, Islam et al. used FL and IoD to create a blockchain-based data accumulation scheme. This scheme enhanced the security of IoD against cyber threats [78]. Zhang et al. investigated a robust semi-supervised learning-based FL scheme in IoD for automatic image recognition. ...
Article
Full-text available
As a result of the advancement in the fourth industrial revolution and communication technology, the use of digital twins (DT) and federated learning (FL) in the industrial Internet of Things (IIoT), the Internet of Vehicles (IoV), and the Internet of Drones (IoD) is increasing. However, the deployment of DT and FL for IoV is challenging. In this survey, we focus on DT and FL for IIoT, IoV, and IoD. Initially, we analyzed the existing surveys. In this paper, we present the applications of DT and FL in IIoT, IoV, and IoD. We also present the open research issues and future directions.
... The first unmanned aerial vehicles (UAVs), commonly called drones, were used for civil activities (i.e., cargo drone) in 2014. In the last few years, their usage has increased significantly [1][2][3] in, for instance, humanitarian [4] and healthcare [5] settings, as well as in environmental emergencies due to climate change (fires [6], storms, landslides [7], etc.). Nowadays, in all countries, the civil defense employs drones for search and rescue operations in natural disasters whenever the areas to be monitored are dangerous for the safety of rescuers. ...
Article
Full-text available
In this paper, we report on the validation of an autonomous drone-based delivery system equipped with a smart capsule for the transportation of blood products in urban areas. The influence of some thermo-mechanical parameters, such as altitude, acceleration/deceleration, external temperature and humidity, on the specimens’ integrity were analyzed. The comparison of the results carried out by hemolytic tests, performed systematically on samples before and after each drone flight, clearly demonstrated that the integrity of blood is preserved and no adverse effects took place during the transport; these results can be addressed to the smart-capsule properties, which allows integrating real-time quality monitoring and control of the temperature experienced by blood products and mechanical vibrations. In addition, we demonstrated this transport system reduces the delivery time considerably. A risk analysis (i.e., HFMEA) was applied to all delivery processes to assess possible criticalities. To the best of our knowledge, this is the first time a drone-based delivery system of blood products in an urban area has been validated to be employed in a future clinical scenario.
... Therefore, blockchain technology, one application form of DLT, is used. More specifically, some advantages motivated us to adopt blockchain in the proposed framework, including (1) maintaining the trust and secure data exchange among peer-to-peer networks [48,68]; (2) allowing traceability across the entire network [69]; (3) provide insightful consensus-based decision-making process [70]; and (4) deliver efficient solutions by utilising the decentralisation feature of blockchain technology [71]. Different open source blockchain and DLT technologies could be used including HeperLedger, Ethereum, Corda, Quorm, and Openchain. ...
Article
Full-text available
Digital twins (DTs) play a vital role in revolutionising the healthcare industry, leading to more personalised, intelligent, and proactive healthcare. With the evolution of personalised healthcare, there is a significant need to represent a virtual replica for individuals to provide the right type of care in the right way and at the right time. Therefore, in this paper, we surveyed the concept of a personal digital twin (PDT) as an enhanced version of the DT with actionable insight capabilities. In particular, PDT can bring value to patients by enabling more accurate decision making and proper treatment selection and optimisation. Then, we explored the progression of PDT as a revolutionary technology in healthcare research and industry. However, although several research works have been performed for smart healthcare using DT, PDT is still at an early stage. Consequently, we believe that this work can be a step towards smart personalised healthcare industry by guiding the design of industrial personalised healthcare systems. Accordingly, we introduced a reference framework that empowers smart personalised healthcare using PDTs by bringing together existing advanced technologies (i.e., DT, blockchain, and AI). Then, we described some selected use cases, including the mitigation of COVID-19 contagion, COVID-19 survivor follow-up care, personalised COVID-19 medicine, personalised osteoporosis prevention, personalised cancer survivor follow-up care, and personalised nutrition. Finally, we identified further challenges to pave the PDT paradigm toward the smart personalised healthcare industry.
... Inspired by this, a global authentication blockchain can be established in the UAV network to prevent unauthorized access by malicious external nodes. Islam, Anik et al. [29] proposed a joint learning base data accumulation scheme combining drones and blockchain for remote areas where IoT devices face network scarcity and potential cyber threats. With the consent of the mining nodes, the federated learning is simply stored on the blockchain to obtain models, reducing the storage overhead of the blockchain. ...
Article
Full-text available
Due to the high mobility of nodes and the complexity of the mission environment, mission-oriented UAV networks are not only subject to frequent topology changes, but also to the risk of being compromised, hijacked and corrupted. As a result, an operating UAV network is essentially a Byzantine distributed system whose physical structure and node trustworthiness change over time. How to implement the global management of UAV networks to achieve a rational allocation of UAV network resources and reconfiguration of trusted networks is a problem worthy of in-depth study. The method proposed in this paper introduces a lightweight storage blockchain in the UAV network through two-stage consensus, firstly performing data consensus on the local state records of the nodes, then performing decision consensus on the data consensus results using algorithms such as fuzzy K-Modes clustering and global trustworthiness assessment, and finally recording the decision consensus results into a new block as the new configuration information of the UAV network. A lightweight storage blockchain-assisted trusted zone routing protocol (BC_TZRP) is designed to dynamically and adaptively build configurable trusted networks in a way that the blockchain continuously adds new blocks. Using QualNet simulation experimental software, an experimental comparison between the classical routing protocol for mobile self-organizing networks and the traditional consensus algorithm for blockchains is conducted. The results show that the approach has significant advantages in terms of packet delivery rate, routing overhead and average end-to-end delay, and can effectively improve the overall working life and fault tolerance of the UAV network.
... In [100], an FL method and a Blockchain framework were successfully combined and a secure drone-aided data accumulation IoT scheme was proposed, namely FBI. To surpass connectivity issues in remote areas, the drones were used as intermediate nodes with onboard dew servers [99] and preserved the end-to-end communication between IoT devices and edge servers. ...
Article
Full-text available
As the Internet of Things (IoT) ecosystem evolves, innovative applications with stringent demands with respect to latency will emerge. To handle computation-intensive tasks in a timely manner, data offloading to Mobile Edge Computing (MEC) servers has been suggested. On the other hand, prospective IoT networks are expected to include Unmanned Aerial Vehicles (UAVs) to enhance coverage and connectivity, while retaining reliable communication links with ground nodes in urban, suburban, and rural terrain. Nevertheless, the evolution of UAV-aided MEC-enabled IoT presupposes the mitigation of security threats through the implementation of efficient and robust countermeasures. As UAVs inherently have certain limitations in terms of energy, computational, and memory resources, designing lightweight security solutions is required. This paper provides an overview of the UAV-aided MEC-enabled IoT and a detailed presentation of use cases and application scenarios, where security is of utmost importance. Subsequently, up-to-date research works on security solutions for the UAV-aided MEC-enabled IoT are comprehensively presented. To this end, the adoption of information-theoretic techniques that ensure adequate Physical-Layer Security (PLS) is discussed along with sophisticated security approaches based on emerging technologies, such as Blockchain and Machine Learning (ML). In addition, research studies on software- and hardware-based methods for the identification and authentication of network nodes are presented. Finally, this paper provides future perspectives in this research domain, stimulating further work.
... Federated Learning (FL) is introduced with the promise of privacy [5]. In FL, data is trained on the user's end and only the weight of the trained model is collected from the user's end [6]. ...
Conference Paper
New diseases (e.g., monkeypox) are showing up and taking the form of a pandemic within a short time. Early detection can assist in reducing the spread. However, because of privacy-sensitive data, users do not share it continually. Thus, it becomes challenging to employ modern technologies (e.g., deep learning). Moreover, cyber threats encircle both communication and data. This paper introduces a blockchain-based data acquisition scheme during the pandemic in which federated learning (FL) is employed to assemble privacy-sensitive data as a form of the trained model instead of raw data. A secure training scheme is designed to mitigate cyber threats (e.g., man-in-the-middle-attack). An experimental environment is formulated based on a recent pandemic (i.e., monkeypox) to illustrate the feasibility of the proposed scheme.
... Planning in UAVs, considering the indicators of their reliability and effectiveness in missions, is presented in [63][64][65]. Security issues regarding the utilization of UAV-based networks in IoT scenarios are considered in [66,67]. ...
Article
Full-text available
This paper suggests a methodology (conception and principles) for building two-mode monitoring systems (SMs) for industrial facilities and their adjacent territories based on the application of unmanned aerial vehicle (UAV), Internet of Things (IoT), and digital twin (DT) technologies, and a set of SM reliability models considering the parameters of the channels and components. The concept of building a reliable and resilient SM is proposed. For this purpose, the von Neumann paradigm for the synthesis of reliable systems from unreliable components is developed. For complex SMs of industrial facilities, the concept covers the application of various types of redundancy (structural, version, time, and space) for basic components—sensors, means of communication, processing, and presentation—in the form of DTs for decision support systems. The research results include: the methodology for the building and general structures of UAV-, IoT-, and DT-based SMs in industrial facilities as multi-level systems; reliability models for SMs considering the applied technologies and operation modes (normal and emergency); and industrial cases of SMs for manufacture and nuclear power plants. The results obtained are the basis for further development of the theory and for practical applications of SMs in industrial facilities within the framework of the implementation and improvement of Industry 4.0 principles.
... Integrating IoT-enabled healthcare systems enables the realization of many robust telehealth patient care applications and raises concerns over the secure and efficient communication of critical personal data [7][8][9]. The large scale of highly valued personal data increases privacy and security risks, and requires innovative networking techniques to maintain industrial-level communication efficiency [10,11]. ...
Article
Full-text available
Dynamic and smart Internet of Things (IoT) infrastructures allow the development of smart healthcare systems, which are equipped with mobile health and embedded healthcare sensors to enable a broad range of healthcare applications. These IoT applications provide access to the clients’ health information. However, the rapid increase in the number of mobile devices and social networks has generated concerns regarding the secure sharing of a client’s location. In this regard, federated learning (FL) is an emerging paradigm of decentralized machine learning that guarantees the training of a shared global model without compromising the data privacy of the client. To this end, we propose a K-anonymity-based secure hierarchical federated learning (SHFL) framework for smart healthcare systems. In the proposed hierarchical FL approach, a centralized server communicates hierarchically with multiple directly and indirectly connected devices. In particular, the proposed SHFL formulates the hierarchical clusters of location-based services to achieve distributed FL. In addition, the proposed SHFL utilizes the K-anonymity method to hide the location of the cluster devices. Finally, we evaluated the performance of the proposed SHFL by configuring different hierarchical networks with multiple model architectures and datasets. The experiments validated that the proposed SHFL provides adequate generalization to enable network scalability of accurate healthcare systems without compromising the data and location privacy.
... Initially, blockchain was perceived as the mechanism for storing financial data and as the intermediary for various financial transactions. However, now we witness the trend of shifting it toward other domains, including the IoT environment, such as for healthcare services with support of Internet of Skills [26], automated manufacturing processes [27], secure data aggregation [28], mixed reality content VOLUME 4, 2016 sharing [29], COVID pandemic monitoring [30], etc. ...
Article
Full-text available
The future of networking will be driven by the current emerging trends of combining the physical and virtual realities in cyberspace. Considering the ambient pandemic challenges, the role of virtual and augmented reality will definitely grow over time by transforming into the paradigm of the Metaverse of Things, where each person, thing or other entity will simultaneously exist within multiple synchronized realities. In this paper, we propose a novel framework for future metaverse applications composed of multiple synchronized data flows from multiple operators through multiple wearable devices and with different quality requirements. A new service quality model is proposed based on a customizable utility function for each individual data flow. The proposed approach is based on dynamic fine-grained data flow allocation and service selection using non-fungible tokens, which can be traded over the blockchain among users and operators in a decentralized mobile network environment.
... Dew computing can provide multiple services that are equivalent to cloud computing, such as data in dew, platform in dew, infrastructure in dew, web in dew, software in dew, storage in dew, and database in dew [71]. Studies show that dew computing-assisted drones [72] and a federated learning-based blockchain can be useful in IoT-aware drone employments [73]. ...
Article
Full-text available
Building a widely distributed hotspot network is a very tedious task due to its complexity. Providing security, fully distributed network services, and a cost-conscious impact are the major challenges behind this goal. To overcome these issues, we have presented a novel distributed hotspot network architecture with five layers that can provide large-scale hotspot coverage as an assimilated result. Our contributions to this new architecture highlight important aspects. First, scalability can be increased by including many Internet of Things (IoT) devices with sensors and Wi-Fi and/or LoraWAN connectivity modules. Second, hotspot owners can rent out their hotspots to create a distributed hotspot network in which the hotspots can act as an ordinary data gateway, a full-fledged hotspot miner, and a light-weight hotspot miner to earn crypto tokens as rewards for certain activities. Third, the advantages of Wi-Fi and LoraWAN can be seamlessly leveraged to achieve optimal coverage, higher network security, and suitable data transmission rate for transferring sensor data from IoT devices to remote application servers and users. Fourth, blockchain is used to enhance the decentralized behavior of the architecture that is presented here by providing immutability and independence from a centralized regulator and making the network architecture more reliable and transparent. The main feature of our paper is the use of the dew-computing paradigm along with hotspots to improve availability, Internet backhaul-agnostic network coverage, and synchronous update capability, and dew-aware leasing to strengthen and improve coverage. We also discuss the key challenges and future roadmap that require further investment and deployment.
... LN node updated model is aggregated in the blockchain network, a separate distributed network. BC network is also responsible for maintaining authentication of LN [24], GW, and CIoT. ...
Article
Full-text available
Resource constraint Consumer Internet of Things (CIoT) is controlled through gateway devices (e.g., smartphones, computers, etc.) that are connected to Mobile Edge Computing (MEC) servers or cloud regulated by a third party. Recently Machine Learning (ML) has been widely used in automation, consumer behavior analysis, device quality upgradation, etc. Typical ML predicts by analyzing customers’ raw data in a centralized system which raises the security and privacy issues such as data leakage, privacy violation, single point of failure, etc. To overcome the problems, Federated Learning (FL) developed an initial solution to ensure services without sharing personal data. In FL, a centralized aggregator collaborates and makes an average for a global model used for the next round of training. However, the centralized aggregator raised the same issues, such as a single point of control leaking the updated model and interrupting the entire process. Additionally, research claims data can be retrieved from model parameters. Beyond that, since the Gateway (GW) device has full access to the raw data, it can also threaten the entire ecosystem. This research contributes a blockchain-controlled, edge intelligence federated learning framework for a distributed learning platform for CIoT. The federated learning platform allows collaborative learning with users’ shared data, and the blockchain network replaces the centralized aggregator and ensures secure participation of gateway devices in the ecosystem. Furthermore, blockchain is trustless, immutable, and anonymous, encouraging CIoT end users to participate. We evaluated the framework and federated learning outcomes using the well-known Stanford Cars dataset. Experimental results prove the effectiveness of the proposed framework.
... Other options aim to increase the security of the FL system. A good illustration is a federated accumulation system [71] based on blockchain and drones. In this case, a cuckoo filter is used to verify requests, and then a nonce timestamp is used to authenticate them. ...
Article
Full-text available
New technologies bring opportunities to deploy AI and machine learning to the edge of the network, allowing edge devices to train simple models that can then be deployed in practice. Federated learning (FL) is a distributed machine learning technique to create a global model by learning from multiple decentralized edge clients. Although FL methods offer several advantages, including scalability and data privacy, they also introduce some risks and drawbacks in terms of computational complexity in the case of heterogeneous devices. Internet of Things (IoT) devices may have limited computing resources, poorer connection quality, or may use different operating systems. This paper provides an overview of the methods used in FL with a focus on edge devices with limited computational resources. This paper also presents FL frameworks that are currently popular and that provide communication between clients and servers. In this context, various topics are described, which include contributions and trends in the literature. This includes basic models and designs of system architecture, possibilities of application in practice, privacy and security, and resource management. Challenges related to the computational requirements of edge devices such as hardware heterogeneity, communication overload or limited resources of devices are discussed.
... Anil Islam, et al. [25] proposed an approach to a federated learning-based blockchain embedded data accumulation scheme using drones for the Internet of Things. The work mainly focuses on securing remote regions using drones and blockchain technology. ...
Article
Full-text available
With the growing demand for smart, secure, and intelligent solutions, Industry 4.0 has emerged as the future of various applications. One of the primary sectors that are becoming more vulnerable to security assaults like ransomware is the healthcare sector. Researchers have proposed various mechanisms in smart and secure health care systems with this vision in mind. Existing systems are vulnerable to security attacks on medical data. It is required to build a real-time diagnosis device using a cyber-physical system with blockchain technology in a considerable manner. The proposed work’s main purpose is to build secure, real-time preservation and tamper-proof control of medical data. In this work, the Bayesian grey filter-based convolution neural network (BGF-CNN) approach is used to enhance accuracy and reduce time complexity and overhead. Additionally, PSO and GWO optimization techniques are used to improve network performance. As an outcome of the proposed work, the privacy preservation of medical data is improved with a high accuracy rate by a blockchain-based cyber-physical system using a deep neural network (BGF Blockchain). To summarize, the proposed system helps in the privacy preservation of medical data along with a reduction in communication overhead using the Bayesian Grey Filter–CNN.
... Current solutions for network intrusion detection were also observed: malware recognition and network attack detection include deep learning [40,41], ensemble learning [42,43], multistage deep learning [44], metaheuristic methods [45], a federated learning-based blockchain-embedded data accumulation scheme [46], and federated transfer learning for bearing fault diagnosis with discrepancy-based weighted federated averaging [47]. ...
Article
Full-text available
The intrusion detection system (IDS) must be able to handle the increase in attack volume, increasing Internet traffic, and accelerating detection speeds. Network flow feature (NTF) records are the input of flow-based IDSs that are used to determine whether network traffic is normal or malicious in order to avoid IDS from difficult and time-consuming packet content inspection processing since only flow records are examined. To reduce computational power and training time, this paper proposes a novel pre-processing method merging a specific amount of NTF records into frames, and frame transformation into images. Federated learning (FL) enables multiple users to share the learned models while maintaining the privacy of their training data. This research suggests federated transfer learning and federated learning methods for NIDS employing deep learning for image classification and conducting tests on the BOUN DDoS dataset to address the issue of training data privacy. Our experimental results indicate that the proposed Federated transfer learning (FTL) and FL methods for training do not require data centralization and preserve participant data privacy while achieving acceptable accuracy in DDoS attack identification: FTL (92.99%) and FL (88.42%) in comparison with Traditional transfer learning (93.95%).
... The proposed approach yields higher diagnosis accuracy without feature engineering and ensures data privacy in real-life deployable scenarios. Islam et al. [44] proposed an FL-based secure data-collection method from IoT devices using drones and blockchain. The proposed approach yields better results in proof of concept experiments, highlighting multiple benefits such as data collection, storage, privacy preservation, security, and execution time. ...
Article
Full-text available
Federated learning (FL) is one of the leading paradigms of modern times with higher privacy guarantees than any other digital solution. Since its inception in 2016, FL has been rigorously investigated from multiple perspectives. Some of these perspectives are extensions of FL’s applications in different sectors, communication overheads, statistical heterogeneity problems, client dropout issues, the legitimacy of FL system results, privacy preservation, etc. Recently, FL is being increasingly used in the medical domain for multiple purposes, and many successful applications exist that are serving mankind in various ways. In this work, we describe the novel applications and challenges of the FL paradigm with special emphasis on the COVID-19 pandemic. We describe the synergies of FL with other emerging technologies to accomplish multiple services to fight the COVID-19 pandemic. We analyze the recent open-source development of FL which can help in designing scalable and reliable FL models. Lastly, we suggest valuable recommendations to enhance the technical persuasiveness of the FL paradigm. To the best of the authors’ knowledge, this is the first work that highlights the efficacy of FL in the era of COVID-19. The analysis enclosed in this article can pave the way for understanding the technical efficacy of FL in medical field, specifically COVID-19.
... Some researchers focused on the security issue of the Internet of Drones Things (IoDT), whereby lightweight blockchain is considered a security solution [25,26]. The combination of cuckoo and Hampel filters [27] is used as a type of blockchain security. Some other studies noted UAV applications in the 5G or 5GB era, from the aspect of wireless communication and its underlying physical characteristics, such as air-to-ground common, energy-efficient channels [28,29]. ...
Article
Full-text available
Recently, the unmanned aerial vehicles (UAV) under the umbrella of the Internet of Things (IoT) in smart cities and emerging communities have become the focus of the academic and industrial science community. On this basis, UAVs have been used in many military and commercial systems as emergency transport and air support during natural disasters and epidemics. In such previous scenarios, boosting wireless signals in remote or isolated areas would need a mobile signal booster placed on UAVs, and, at the same time, the data would be secured by a secure decentralized database. This paper contributes to investigating the possibility of using a wireless repeater placed on a UAV as a mobile booster for weak wireless signals in isolated or rural areas in emergency situations and that the transmitted information is protected from external interference and manipulation. The working mechanism is as follows: one of the UAVs detect a human presence in a predetermined area with the thermal camera and then directs the UAVs to the location to enhance the weak signal and protect the transmitted data. The methodology of localization and clusterization of the UAVs is represented by a swarm intelligence localization (SIL) optimization algorithm. At the same time, the information sent by UAV is protected by blockchain technology as a decentralization database. According to realistic studies and analyses of UAVs localization and clusterization, the proposed idea can improve the amplitude of the wireless signals in far regions. In comparison, this database technique is difficult to attack. The research ultimately supports emergency transport networks, blockchain, and IoT services.
... In this article, there are mainly multiple drones to be replaced, which can solve the problem of endurance. In Islam et al. (2022), the routing design of drones at each charging station is mainly discussed to serve the power supplement of the drone. In Kim and Moon (2019), drones are mainly used to guide vehicle into the parking lot, and it is guided according to the parking lot route plan. ...
Article
Full-text available
Today, many maritime nations have been promoting boat sports proactively, including sailing races. As sailing races are large-scale regattas that require massive workforces to monitor the game fairly; however, with limited match budgets and labors, Internet of Things (IoT) technology supports monitoring games has become a trend. This article proposes a robot umpire system in sailing races based on Artificial Intelligence (AI) techniques, using drones and AIoT technology to monitor sailing matches. When a large number of sailboats are in a match, and each sail along different routes, drones can monitor the entire game simultaneously. The features of this proposed approach are (1) The system recognizes images by Faster R-CNN, judging whether a sailboat uses a motor to accelerate; (2) The system detects conditions by edge computing; when cheating behaviors happen, it can notify the event holder immediately; (3) Advanced drone route plans can avoid collision incidents; (4) Improve the system recognition by federated learning. This study has implemented an experiment with real drones and installed IoT equipment on the drones for taking videos and recognizing. The experimental result has shown that the proposed approach is feasible and benefits the match's fairness. Additionally, umpires can review the violation details from the videos taken by the drones, supporting evidence for judging.
... Google Research invented FL in 2016, which has become a popular approach for machine learning [5]. FL was introduced to mitigate privacy concerns [6]. FL provides new training methods that can help to create personalized models without infringing any user's privacy [7]. ...
Conference Paper
In the modern era, the internet of vehicles (IoV) is being utilized in commercial applications and extensively explored in research. However, internal fault in IoV can cause accidents on the road. Moreover, privacy concerns can hamper the internal data sharing to build a model to detect the anomaly. Federated learning (FL) and blockchain are emerging technologies that can assist in mitigating these challenges. FL-based anomaly detection is introduced to prevent road accidents with the help of blockchain. An environment is built to conduct experiments to prove the feasibility of the proposed scheme. The performance analysis demonstrates that our presented scheme outperforms the traditional scheme while having privacy concerns.
Article
Full-text available
Handling electronic health records from the Internet of Medical Things is one of the most challenging research areas as it consists of sensitive information, which targets attackers. Also, dealing with modern healthcare systems is highly complex and expensive, requiring much secured storage space. However, blockchain technology can mitigate these problems through improved health record management. The proposed work develops a scalable, lightweight framework based on blockchain technology to improve COVID‐19 data security, scalability and patient privacy. Initially, the COVID‐19 related data records are hashed using the enhanced Merkle tree data structure. The hashed values are encrypted by lattice based cryptography with a Homomorphic proxy re‐encryption scheme in which the input data are secured. After completing the encryption process, the blockchain uses inter planetary file system to store secured information. Finally, the Proof of Work concept is utilized to validate the security of the input COVID based data records. The proposed work's experimental setup is performed using the Python tool. The performance metrics like encryption time, re‐encryption time, decryption time, overall processing time, and latency prove the efficacy of the proposed schemes.
Article
Full-text available
Digital modeling of the real world with 3D modeling technologies has recently prepared the groundwork for Metaverse. Metaverse is an ecosystem that will allow social and cultural events in the real world to be carried out in the digital world as well. Inspired by this framework, this study presents an approach called MetaRepo, where users can securely store digital assets (cryptocurrency, avatars, clothes, tickets, etc.) and use them in various activities within the metaverse universe. The motivation for the realization of this study is the security problems in the exchange, buying and selling transactions that take place in the virtual universes that have become popular in recent days. Another source of motivation is the anxiety of the object owners having their possessions stolen, lost, and transferred to another universe in the Metaverse. Blockchain technology, which can store assets in MetaRepo, has been used to address these concerns. In the developed blockchain structure, New User Engine, Transaction Centre, Authenticator Engine (Weng) and Repos models have been developed for user interaction, transaction processing and security mechanism. An exemplary metaverse universe including social activities has been designed for the testing and evaluation processes of the proposed MetaRepo approach. MetaRepo is communicated from the browser via APIs. Detailed performance analysis has been carried out for the proposed model. As a result, with MetaRepo, a mechanism is aimed at which users can communicate with different metaverse universes and platforms without the need for extra verification and security measures within the metaverse universe.
Article
Full-text available
Drone security is currently a major topic of discussion among researchers and industrialists. Although there are multiple applications of drones, if the security challenges are not anticipated and required architectural changes are not made, the upcoming drone applications will not be able to serve their actual purpose. Therefore, in this paper, we present a detailed review of the security-critical drone applications, and security-related challenges in drone communication such as DoS attacks, Man-in-the-middle attacks, De-Authentication attacks, and so on. Furthermore, as part of solution architectures, the use of Blockchain, Software Defined Networks (SDN), Machine Learning, and Fog/Edge computing are discussed as these are the most emerging technologies. Drones are highly resource-constrained devices and therefore it is not possible to deploy heavy security algorithms on board. Blockchain can be used to cryptographically store all the data that is sent to/from the drones, thereby saving it from tampering and eavesdropping. Various ML algorithms can be used to detect malicious drones in the network and to detect safe routes. Additionally, the SDN technology can be used to make the drone network reliable by allowing the controller to keep a close check on data traffic, and fog computing can be used to keep the computation capabilities closer to the drones without overloading them.
Article
Full-text available
The communication and networking field is hungry for machine learning decision-making solutions to replace the traditional model-driven approaches that proved to be not rich enough for seizing the ever-growing complexity and heterogeneity of the modern systems in the field. Traditional machine learning solutions assume the existence of (cloud-based) central entities that are in charge of processing the data. Nonetheless, the difficulty of accessing private data, together with the high cost of transmitting raw data to the central entity gave rise to a decentralized machine learning approach called Federated Learning . The main idea of federated learning is to perform an on-device collaborative training of a single machine learning model without having to share the raw training data with any third-party entity. Although few survey articles on federated learning already exist in the literature, the motivation of this survey stems from three essential observations. The first one is the lack of a fine-grained multi-level classification of the federated learning literature, where the existing surveys base their classification on only one criterion or aspect. The second observation is that the existing surveys focus only on some common challenges, but disregard other essential aspects such as reliable client selection, resource management and training service pricing. The third observation is the lack of explicit and straightforward directives for researchers to help them design future federated learning solutions that overcome the state-of-the-art research gaps. To address these points, we first provide a comprehensive tutorial on federated learning and its associated concepts, technologies and learning approaches. We then survey and highlight the applications and future directions of federated learning in the domain of communication and networking. Thereafter, we design a three-level classification scheme that first categorizes the federated learning literature based on the high-level challenge that they tackle. Then, we classify each high-level challenge into a set of specific low-level challenges to foster a better understanding of the topic. Finally, we provide, within each low-level challenge, a fine-grained classification based on the technique used to address this particular challenge. For each category of high-level challenges, we provide a set of desirable criteria and future research directions that are aimed to help the research community design innovative and efficient future solutions. To the best of our knowledge, our survey is the most comprehensive in terms of challenges and techniques it covers and the most fine-grained in terms of the multi-level classification scheme it presents.
Article
Full-text available
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 a 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.
Article
Full-text available
By leveraging blockchain, this letter proposes a blockchained federated learning (BlockFL) architecture where local learning model updates are exchanged and verified. This enables on-device machine learning without any centralized training data or coordination by utilizing a consensus mechanism in blockchain. Moreover, we analyze an end-to-end latency model of BlockFL and characterize the optimal block generation rate by considering communication, computation, and consensus delays.
Article
Full-text available
Learning to store information over extended time intervals by recurrent backpropagation takes a very long time, mostly because of insufficient, decaying error backflow. We briefly review Hochreiter's (1991) analysis of this problem, then address it by introducing a novel, efficient, gradient-based method called long short-term memory (LSTM). Truncating the gradient where this does not do harm, LSTM can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units. Multiplicative gate units learn to open and close access to the constant error flow. LSTM is local in space and time; its computational complexity per time step and weight is O(1). Our experiments with artificial data involve local, distributed, real-valued, and noisy pattern representations. In comparisons with real-time recurrent learning, back propagation through time, recurrent cascade correlation, Elman nets, and neural sequence chunking, LSTM leads to many more successful runs, and learns much faster. LSTM also solves complex, artificial long-time-lag tasks that have never been solved by previous recurrent network algorithms.
Article
Trusted third parties (TTPs) are frequently used for serving as an authority to issue and verify transactions in applications. Although the TTP-based paradigm provides customers with convenience, it causes a whole set of inevitable problems such as security threats, privacy vulnerabilities, and censorship. The TTP-based paradigm is not suitable for all modern networks, e.g., 5G and beyond networks, which are been evolving to support ubiquitous, decentralized, and autonomous services. Driven by the vision of blockchain technologies, there has been a paradigm shift in applications, from TTP-based to decentralized-trust-based. Decentralized applications (DApps) with blockchains promise no trust on authorities, tackling the key challenges of security and privacy problems. A main thrust of blockchain research is to explore frameworks and paradigms for decentralizing applications, fostering a number of new designs ranging from network architectures to business models. Therefore, this paper provides a compact and concise survey on the state-of-the-art research of decentralizing applications with blockchain in the 5G and beyond perspective. We provide four burning 5G and beyond challenges and discuss five aspects of motivation for decentralizing applications with blockchain. Then, we define nine fundamental modules of blockchains and explain the potential influence of these modules on decentralization in depth. We also discuss the interrelation between decentralization and some desired blockchain properties. Particularly, we present the capabilities of blockchain for decentralizing applications through reviewing DApps for 5G and beyond. We clearly distinguish three blockchain paradigms and discuss how developers to make right choices for 5G and beyond. Finally, we highlight important learned lessons and open issues in applying blockchain for decentralizing applications. Lessons learned and open issues from this survey will facilitate the transformation of centralized applications to DApps.
Article
Edge-of-Things (EoT) enables the seamless transfer of services, storage and data processing from the Cloud layer to Edge devices in a large-scale distributed Internet of Things (IoT) ecosystems (e.g., Industrial systems). This transition raises the privacy and security concerns in the EoT paradigm distributed at different layers. Intrusion detection systems are implemented in EoT ecosystems to protect the underlying resources from attackers. However, the current intrusion detection systems are not intelligent enough to control the false alarms, which significantly lower the reliability and add to the analysis burden on the intrusion detection systems. In this article, we present a DaaS, Dew Computing as a Service for intelligent intrusion detection in EoT ecosystems. In DaaS, a deep learning-based classifier is used to design an intelligent alarm filtration mechanism. In this mechanism, the filtration accuracy is improved (or sustained) by using deep belief networks. In the past, the cloud-based techniques have been applied for offloading the EoT tasks, which increases the middle layer burden and raises the communication delay. Here, we introduce the dew computing features which are used to design the smart false alarm reduction system. DaaS, when experimented in a simulated environment, reflects lower response time to process the data in the EoT ecosystem. The revamped DBN model achieved the classification accuracy up to 95%. Moreover, it depicts a 60% improvement in the latency and 35% workload reduction of the cloud servers as compared to Edge intrusion detection system.
Article
Unmanned aerial vehicles (UAVs) combined with artificial intelligence (AI) have opened a revolutionized way for mobile crowdsensing (MCS). Conventional AI models, built on aggregation of UAVs’ sensing data (typically contain private and sensitive user information), may arise severe privacy and data misuse concerns. Federated learning, as a promising distributed AI paradigm, has opened up possibilities for UAVs to collaboratively train a shared global model without revealing their local sensing data. However, there still exist potential security and privacy threats for UAV-assisted crowdsensing with federated learning due to vulnerability of central curator, unreliable contribution recording, and low-quality shared local models. In this paper, we propose SFAC, a s ecure f ederated learning framework for U A V-assisted M C S. Specifically, we first introduce a blockchain-based collaborative learning architecture for UAVs to securely exchange local model updates and verify contributions without the central curator. Then, by applying local differential privacy, we design a privacy-preserving algorithm to protect UAVs’ privacy of updated local models with desirable learning accuracy. Furthermore, a two-tier reinforcement learning-based incentive mechanism is exploited to promote UAVs’ high-quality model sharing when explicit knowledge of network parameters are not available in practice. Extensive simulations are conducted, and the results demonstrate that the proposed SFAC can effectively improve utilities for UAVs, promote high-quality model sharing, and ensure privacy protection in federated learning, compared with existing schemes.
Conference Paper
Machine learning techniques based on neural networks are achieving remarkable results in a wide variety of domains. Often, the training of models requires large, representative datasets, which may be crowdsourced and contain sensitive information. The models should not expose private information in these datasets. Addressing this goal, we develop new algorithmic techniques for learning and a refined analysis of privacy costs within the framework of differential privacy. Our implementation and experiments demonstrate that we can train deep neural networks with non-convex objectives, under a modest privacy budget, and at a manageable cost in software complexity, training efficiency, and model quality.
Conference Paper
In many networking systems, Bloom filters are used for high-speed set membership tests. They permit a small fraction of false positive answers with very good space efficiency. However, they do not permit deletion of items from the set, and previous attempts to extend "standard" Bloom filters to support deletion all degrade either space or performance. We propose a new data structure called the cuckoo filter that can replace Bloom filters for approximate set membership tests. Cuckoo filters support adding and removing items dynamically while achieving even higher performance than Bloom filters. For applications that store many items and target moderately low false positive rates, cuckoo filters have lower space overhead than space-optimized Bloom filters. Our experimental results also show that cuckoo filters outperform previous data structures that extend Bloom filters to support deletions substantially in both time and space.
Article
We introduce Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions. The method is straightforward to implement and is based an adaptive estimates of lower-order moments of the gradients. The method is computationally efficient, has little memory requirements and is well suited for problems that are large in terms of data and/or parameters. The method is also ap- propriate for non-stationary objectives and problems with very noisy and/or sparse gradients. The method exhibits invariance to diagonal rescaling of the gradients by adapting to the geometry of the objective function. The hyper-parameters have intuitive interpretations and typically require little tuning. Some connections to related algorithms, on which Adam was inspired, are discussed. We also analyze the theoretical convergence properties of the algorithm and provide a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework. We demonstrate that Adam works well in practice when experimentally compared to other stochastic optimization methods.
Article
One approach to identifying outliers is to assume that the outliers have a different distribution from the remaining observations. In this article we define outliers in terms of their position relative to the model for the good observations. The outlier identification problem is then the problem of identifying those observations that lie in a so-called outlier region. Methods based on robust statistics and outward testing are shown to have the highest possible breakdown points in a sense derived from Donoho and Huber. But a more detailed analysis shows that methods based on robust statistics perform better with respect to worst-case behavior. A concrete outlier identifier based on a suggestion of Hampel is given.
Communication-efficient learning of deep networks from decentralized data
  • B Mcmahan
  • E Moore
  • D Ramage
  • S Hampson
  • B A Y Arcas
B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y. Arcas, "Communication-Efficient Learning of Deep Networks from Decentralized Data," in Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, ser. Proceedings of Machine Learning Research, A. Singh and J. Zhu, Eds., vol. 54. PMLR, 20-22 Apr 2017, pp. 1273-1282.
Adaptive federated optimization
  • S J Reddi
  • Z Charles
  • M Zaheer
  • Z Garrett
  • K Rush
  • J Konečný
  • S Kumar
  • H B Mcmahan
S. J. Reddi, Z. Charles, M. Zaheer, Z. Garrett, K. Rush, J. Konečný, S. Kumar, and H. B. McMahan, "Adaptive federated optimization," in 9th International Conference on Learning Representations, ICLR 2021, Virtual Event, Austria, May 3-7, 2021, 2021.
Adaptive federated optimization
  • S J Reddi
Communication-efficient learning of deep networks from decentralized data
  • mcmahan