No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Secure Federated Learning in 5G Mobile Networks
... Secret sharing seems to be a common scenario in FL. Isaksson et al. [39] integrated secret sharing into the 5G Network Data Analytics framework, enabling Network Functions to use independent random sample masks to mask their local updates. It optimises the communication costs of secret sharing protocols by generating masks from pairs of shared secrets that add masks and cancel masks. ...
... Hao et al. [25] integrated Distributed DP technology and FL for sample sets in industries with large numbers of data nodes to protect user privacy while guaranteeing ease of the data. In the 5G era, the integration of cloud and network will make the issue of network security even more important, and a solution is proposed to use MPC to ensure that updates are aggregated in a privacy-protected manner [39]. ...
Federated learning (FL) is a distributed machine learning framework that solves the problem of data island while ensuring the privacy of users' data. FL only collects midway parameters with relevant features to train a learning model. The parameters of the exchange model, however, may implicitly reveal specific characteristics or sensitive data of the user, as the gradient may be cracked. Several kinds of literature have noted the problem and proposed solutions such as homomorphic encryption, differential privacy, and secure multi-party computation. Some researchers have discussed FL, but they have some limitations to the summary of privacy protection and application. Based on that, this paper aims to provide a systematic literature review for current approaches of privacy protection methods in FL. The paper presents and discusses the differences between the various solutions, and classifies and analyses the solutions in different practical application scenarios. Moreover, some suggestions for trade-offs in privacy protection are provided.
... We highlight 41 general threats for 5G networks, regardless of vertical applications (Dutta and Hammad, 2020). An adversary can maliciously (1) use legitimate orchestrator access to manipulate the configuration and run a compromised network function, (2) take advantage of malicious insiders attacks, (3) perform unauthorized access (e.g., to confidential data (Isaksson and Norrman, 2020) and to RFID tags (Rahimi et al., 2018)), (4) tampering, (5) perform resource exhaustion, (6) turn services unavailable, (7) analyze or (Bordel et al., 2021)), (9) perform attacks for resource shortages, (10) extract users private information using a shared service in an unauthorized manner, (11) compromise security controls, (Vidal et al., 2018). ...
Background: The deployment of 5G infrastructure is one of the vectors for new application scenarios since it enables enhanced data bandwidth, low latency, and comprehensive signal coverage. This communication system supports various vertical applications such as smart health, smart cities, smart grids, and transportation systems. However, these applications bring new challenges to 5G networks due to specific requirements for such scenarios. Furthermore, as software-based technologies, including network slicing, software-defined networks, network function virtualization, and multi-access edge computing, are a fundamental part of the 5G architecture, the network can expose these applications to new security and privacy concerns. Results: This study summarizes existing literature on 5G vertical applications security. We highlight vulnerabilities, threats, attacks, and solutions for 5G vertical applications. We conducted a systematic literature mapping to discuss security and privacy challenges regarding the 5G vertical applications. We reviewed 389 papers from 2,349 produced by searching with a curated search query and discussed vulnerabilities, threats, attacks, and solutions for 5G vertical applications. Conclusions: Smart cities, Industry 4.0, smart transportation, public services, smart grids, and smart health are vertical applications with relevant security concerns. We observed the need for more research since the 5G and vertical applications continuously evolve.
... In a decentralized network like MANET, this requirement poses a challenge because collecting and aggregating data centrally without violating privacy concerns is impractical. Federated learning (FL) emerges as a promising solution to this problem [25] [26]. It allows individual nodes to train local models on their own data and then share model updates, rather than raw data, with neighboring nodes. ...
Mobile ad hoc networks (MANETs) face significant challenges in maintaining secure and efficient communication owing to their dynamic nature and vulnerability to security threats. Traditional routing protocols often struggle to adapt to rapidly changing topologies and potential malicious nodes, compromising network performance and security. This study addresses these challenges by proposing FLSTMT-LAR (Federated Learning Long Short-Term Memory Trust-aware Location-aided Routing), a novel framework that integrates multiobjective optimization with LSTM-based trust prediction for robust routing decisions, implements a decentralized federated learning mechanism for collaborative trust model updates while preserving node privacy, incorporates dynamic trust assessment using LSTM networks for accurate temporal behavior pattern analysis, and provides an adaptive routing decision mechanism that effectively balances multiple performance objectives including trustworthiness, energy efficiency, and network latency. We evaluate this framework against existing protocols across various scenarios, including different network densities, mobility patterns, and malicious node percentages. Results demonstrate FLSTMT-LAR’s superior performance in high-threat environments, achieving up to 80% packet delivery ratio compared with 45% for traditional approaches. In mobile scenarios, it shows improved adaptability, maintaining consistent performance as network density increases. MOO, particularly nondominated sorting genetic algorithm III, effectively balances conflicting network objectives, offering a 15% improvement in overall network performance compared with single-objective approaches. These findings highlight the potential of integrating advanced machine learning and optimization techniques in MANET routing protocols, paving the way for secure, efficient, and adaptive network communications in challenging environments.
... For instance, in [73], the authors proposed a dedicated FL blockchain to ensure secure FL and create a marketplace for solving federated learning problems. Te study in [74] integrated the FL into the 3GPP 5G data analytics architecture for much lower communication. In [75], the authors explored combining an intelligent refecting surface (IRS) and UAV to form an aerial IRS system, providing comprehensive 360-degree panoramic full-angle refection and fexible deployment of the IRS system. ...
Wireless technologies are growing unprecedentedly with the advent and increasing popularity of wireless services worldwide. With the advancement in technology, profound techniques can potentially improve the performance of wireless networks. Besides, the advancement of artificial intelligence (AI) enables systems to make intelligent decisions, automation, data analysis, insights, predictive capabilities, learning, and adaptation. A sophisticated AI will be required for next-generation wireless networks to automate information delivery between smart applications simultaneously. AI technologies, such as machines and deep learning techniques, have attained tremendous success in many applications in recent years. Hances, researchers in academia and industry have turned their attention to the advanced development of AI-enabled wireless networks. This paper comprehensively surveys AI technologies for different wireless networks with various applications. Moreover, we present various AI-enabled applications that exploit the power of AI to enable the desired evolution of wireless networks. Besides, the challenges of unsolved research in this area, which represent the future research trends of AI-enabled wireless networks, are discussed in detail. We provide several suggestions and solutions that help wireless networks be more intelligent and sophisticated to handle complicated problems. In summary, this paper can help researchers deeply understand the up-to-the-minute wireless network designs based on AI technologies and identify interesting unsolved issues to be pursued in their research in a fast way.
... Federated learning can help in improving QoE modeling. FL is used in many user scenarios, such as active storage of multimedia contents [12], protecting the confidentiality of local updates in 5G networks [13], optimizing mobile edge computing, etc. However, the application of FL approaches for QoE modeling is quite limited in the literature. ...
... FL and transfer learning. In [87], FL is combined with NWDAF in a setup where NFs may use specific ML models to train local datasets. Afterwards, the updated models are sent as events using data collection procedures. ...
The full deployment of sixth-generation (6G) networks is inextricably connected with a holistic network redesign able to deal with various emerging challenges, such as integration of heterogeneous technologies and devices, as well as support of latency and bandwidth demanding applications. In such a complex environment, resource optimization, and security and privacy enhancement can be quite demanding, due to the vast and diverse data generation endpoints and associated hardware elements. Therefore, efficient data collection mechanisms are needed that can be deployed at any network infrastructure. In this context, the network data analytics function (NWDAF) has already been defined in the fifth-generation (5G) architecture from Release 15 of 3GPP, that can perform data collection from various network functions (NFs). When combined with advanced machine learning (ML) techniques, a full-scale network optimization can be supported, according to traffic demands and service requirements. In addition, the collected data from NWDAF can be used for anomaly detection and thus, security and privacy enhancement. Therefore, the main goal of this paper is to present the current state-of-the-art on the role of the NWDAF towards data collection, resource optimization and security enhancement in next generation broadband networks. Furthermore, various key enabling technologies for data collection and threat mitigation in the 6G framework are identified and categorized, along with advanced ML approaches. Finally, a high level architectural approach is presented and discussed, based on the NWDAF, for efficient data collection and ML model training in large scale heterogeneous environments.
... As service policies of ML-CFs, the CNN-based anomaly detection algorithm is utilized to finger out a suitable traffic classification and redirection. The authors of [17] came up with a plan for protecting end-user privacy by using federated learning (FL) and described how to include it into the 3GPP5G network data analytics (NWDA) architecture. They also added a multi-party computation protocol to protect the confidentiality of local updates. ...
... B EYOND 5G networks (B5G), or so-called "6G", are considered as key enabler of a wide range of new pervasive services related to different vertical industries, including eHealth, automotive, energy, and manufacturing [1]. This is possible by the support of massive number of coexisting network slices with different requirements and functionality. ...
... While there may be a benign reason that the free-riding client does not have training data to contribute to the FL model updates, there may be also selfish reason that this client may avoid incurring the cost of spending its communication and computation resources, or a malicious reason that this client may aim to steal the global model. Wireless systems can support various applications of FL such as in mobile edge networks [11], Internet of Things (IoT) [14], 5G [15], [16], and 6G [17]. FL can be performed over wireless links [18]- [22] and multi-hop wireless networks [23]. ...
This paper presents a game theoretic framework for participation and free-riding in federated learning (FL), and determines the Nash equilibrium strategies when FL is executed over wireless links. To support spectrum sensing for NextG communications, FL is used by clients, namely spectrum sensors with limited training datasets and computation resources, to train a wireless signal classifier while preserving privacy. In FL, a client may be free-riding, i.e., it does not participate in FL model updates, if the computation and transmission cost for FL participation is high, and receives the global model (learned by other clients) without incurring a cost. However, the free-riding behavior may potentially decrease the global accuracy due to lack of contribution to global model learning. This tradeoff leads to a non-cooperative game where each client aims to individually maximize its utility as the difference between the global model accuracy and the cost of FL participation. The Nash equilibrium strategies are derived for free-riding probabilities such that no client can unilaterally increase its utility given the strategies of its opponents remain the same. The free-riding probability increases with the FL participation cost and the number of clients, and a significant optimality gap exists in Nash equilibrium with respect to the joint optimization for all clients. The optimality gap increases with the number of clients and the maximum gap is evaluated as a function of the cost. These results quantify the impact of free-riding on the resilience of FL in NextG networks and indicate operational modes for FL participation.
... FL is defined as "a machine learning setting where multiple entities (clients) collaborate in solving a machine learning problem, under the coordination of a central server or service provider: Each client's raw data is stored locally and not exchanged or transferred; instead, focused updates intended for immediate aggregation are used to achieve the learning objective" [3]. The FL is used in many application scenarios, such as proactive caching of multimedia contents [13], protecting the confidentiality of local updates in 5G networks [14], and optimizing mobile edge computing, caching and communication by intelligently utilizing the collaboration among devices [15]. However, the application of the FL approaches to the QoE modelling is quite limited in the literature. ...
The Federated Learning (FL) approach can be exploited to build a solution to data sparsity and privacy protection issues (e.g., utilization of user-sensitive data) in Quality of Experience (QoE) modelling. In this paper, we investigate whether it is possible to obtain improvements in FL-based inference by grouping data sources to build separate inference systems. To this, we adopted an experimental based approach: firstly, we identified different clusters of users, from a public QoE dataset, based on user-related QoE influence factors and the distributions of the quality rating scores provided by the users; secondly, we developed a Cluster-Based FL QoE pre-dictor and conducted experimental tests to compare the QoE prediction performance with that obtained by a centralised learning approach and a standard FL approach. The obtained results show that the proposed approach achieved the best QoE prediction performance (in terms of accuracy, precision, recall, and F1-Score), followed respectively by the standard FL and the centralised approach.
... It has to noticed that the HAP computation infrastructure can be implemented by resorting to the function virtualization approach through different virtualization technologies, e.g., virtual machines, containers, hyper-visors, for performing the FL process. Moreover, the interaction with FL-clients can happen through predefined interfaces (e.g., implementing the REST API technology) allowing a smarter interaction [39]. However, such considerations are beyond the scope of this work, that instead mainly focuses on the optimization of the joint FL-offloading framework. ...
With the advent of smart vehicles, several new latency-critical and data-intensive applications are emerged in Vehicular Networks (VNs). Computation offloading has emerged as a viable option allowing to resort to the nearby edge servers for remote processing within a requested service latency requirement. Despite several advantages, computation offloading over resource-limited edge servers, together with vehicular mobility, is still a challenging problem to be solved. In particular, in order to avoid additional latency due to out-of-coverage operations, Vehicular Users (VUs) mobility introduces a bound on the amount of data to be offloaded towards nearby edge servers. Therefore, several approaches have been used for finding the correct amount of data to be offloaded. Among others, Federated Learning (FL) has been highlighted as one of the most promising solving techniques, given the data privacy concerns in VNs and limited communication resources. However, FL consumes resources during its operation and therefore incurs an additional burden on resource-constrained VUs. In this work, we aim to optimize the VN performance in terms of latency and energy consumption by considering both the FL and the computation offloading processes while selecting the proper number of FL iterations to be implemented. To this end, we first propose an FL-inspired distributed learning} framework for computation offloading in VNs, and then develop a constrained optimization problem to jointly minimize the overall latency and the energy consumed. An evolutionary Genetic Algorithm is proposed for solving the problem in-hand and compared with some benchmarks. The simulation results show the effectiveness of the proposed approach in terms of latency and energy consumption.
... The security and privacy of DL applications over 5G edge networks were surveyed in [152]. To protect the sensitive data available to network functions or network slices of a 5G network and protect the confidentiality of local DL model updates, the authors in [153] proposed a Multi-Party Computation (MPC) protocol on top of FL within 5G networks. IoT-based traffic data was subjected to FL by the authors in [154], where private data was trained using an SVM RBF kernel function, and the global training module was administered by a secure blockchain smart contract. ...
The current pandemic caused by the COVID-19 virus requires more effort, experience, and science-sharing to overcome the damage caused by the pathogen. The fast and wide human-to-human transmission of the COVID-19 virus demands a significant role of the newest technologies in the form of local and global computing and information sharing, data privacy, and accurate tests. The advancements of deep neural networks, cloud computing solutions, blockchain technology, and beyond 5G (B5G) communication have contributed to the better management of the COVID-19 impacts on society. This paper reviews recent attempts to tackle the COVID-19 situation using these technological advancements.
... Open challenges in federated learning are that efficiency comes at the cost of accuracy and that the cooperative parties may more easily inject backdoors into the global model because the training data is hidden [88]. Within the scope of 5G research, federated learning has been integraded [141] into the 3GPP 5G Network Data Analytics (NWDA) function. Further, blockchain based approaches [140] have been proposed to protect integrity of federated networks and to detect malicious cooperating parties. ...
Machine learning (ML) is expected to solve many challenges in the fifth generation (5G) of mobile networks. However, ML will also open the network to several serious cybersecurity vulnerabilities. Most of the learning in ML happens through data gathered from the environment. Un-scrutinized data will have serious consequences on machines absorbing the data to produce actionable intelligence for the network. Scrutinizing the data, on the other hand, opens privacy challenges. Unfortunately, most of the ML systems are borrowed from other disciplines that provide excellent results in small closed environments. The resulting deployment of such ML systems in 5G can inadvertently open the network to serious security challenges such as unfair use of resources, denial of service, as well as leakage of private and confidential information. Therefore, in this article we dig into the weaknesses of the most prominent ML systems that are currently vigorously researched for deployment in 5G. We further classify and survey solutions for avoiding such pitfalls of ML in 5G systems.
Beyond 5G networks, (B5G) introduce the emergence of a smart service and an automate management archetype of running network slices, such as the network data analytics function (NWDAF). These new architectures provide a large usage of advanced machine learning algorithms, in order to dynamically build efficient decisions. In this perspective, the Distributed Learning (D‐), known as A Distributed NWDAF Architecture (D‐NWDAF) has proved its efficiency in not only building collaborative deep learning models, among several network slices, but also ensuring the privacy and isolation of such network slices. Notwithstanding, FL is vulnerable to different attacks, where FL nodes (Leaf NWDAF: AI‐VNFs * ) may upload malicious updates to the FL central (root NWDAF) entity so that it can cause a construction failure of the FL global model and affect the global performance. Moreover, attacks detection of FL‐based solutions give “black‐box” decisions about the performance of running network slices and the attacks detected. In other words, detection solutions do not provide any details about why and how such ML attacks decisions were made. Thus, such decisions cannot be properly trusted and comprehended by B5G slice managers. To resolve this issue, we leverage Dimensional Reduction (DR) and Explainable Artificial Intelligence (XAI) algorithms which aim to detect attacks and improve the transparency of black‐box FL attacks detection‐making process. In the present article, we design a novel DR‐XAI‐powered framework to detect attacks and explain the FL‐based attacks detection. We first build a deep learning model in a federated way, to predict key performance indicators (KPI) of network slices. Then, we try to detect the malicious FL nodes, using DR and several XAI models, such as RuleFit, in order to enhance the level of trustiness, transparency, and the explanation of the FL‐based attacks detection, while adhering the data privacy, to different B5G network stakeholders.
In this chapter, we first introduce the preliminaries of FL. Inparticular, we introduce the federated averaging and two personalized FL algorithms. Then, we introduce four important performance metrics to quantify the FL performance over wireless networks and analyze how wireless factors affect these metrics. Finally, we present the research directions and industry interest of designing communication efficient FL over wireless networks.
The importance of reliable and secure wireless communication cannot be overstated, especially in emergency situations that require the use of smart wearables. To ensure the provision of robust and secure communication services for emergency applications, we propose a decentralized and real-time approach to accessing data from wearable technology. This approach involves integrating blockchain technology and federated-learning-enabled beyond 5G/6G (B5G/6G) network architecture. We also discuss the potential benefits of using massively populated and real-time-streaming, data-intensive, federated-learning-enabled B5G/6G systems to provide emergency services with real-time access to relevant information. By integrating blockchain and responsible AI frameworks, we aim to ensure the security and trustworthiness of the communication channels. We believe that our proposed solution will address sovereignty concerns and have the potential to transform the way emergency responses are handled, resulting in faster and more efficient responses that ultimately save lives.
The fifth generation (5G) and beyond wireless networks are envisioned to provide an integrated communication and computing platform that will enable multipurpose and intelligent networks driven by a growing demand for both traditional end users and industry verticals. This evolution will be realized by innovations in both core and access capabilities, mainly from virtualization technologies and ultra-dense networks, e.g., software-defined networking (SDN), network slicing, network function virtualization (NFV), multi-access edge computing (MEC), terahertz (THz) communications, etc. However, those technologies require increased complexity of resource management and large configurations of network slices. In this new milieu, with the help of artificial intelligence (AI), network operators will strive to enable AI-empowered network management by automating radio and computing resource management and orchestration processes in a data-driven manner. In this regard, most of the previous AI-empowered network management approaches adopt a traditional centralized training paradigm where diverse training data generated at network functions over distributed base stations associated with MEC servers are transferred to a central training server. On the other hand, to exploit distributed and parallel processing capabilities of distributed network entities in a fast and secure manner, federated learning (FL) has emerged as a distributed AI approach that can enable many AI-empowered network management approaches by allowing for AI training at distributed network entities without the need for data transmission to a centralized server. This article comprehensively surveys the field of FL-empowered mobile network management for 5G and beyond networks from access to the core. Specifically, we begin with an introduction to the state-of-the-art of FL by exploring and analyzing recent advances in FL in general. Then, we provide an extensive survey of AI-empowered network management, including background on 5G network functions, mobile traffic prediction, and core/access network management regarding standardization and research activities. We then present an extensive survey of FL-empowered network management by highlighting how FL is adopted in AI-empowered network management. Important lessons learned from this review of AI and FL-empowered network management are also provided. Finally, we complement this survey by discussing open issues and possible directions for future research in this important emerging area.
During COVID-19, the logistics industry plays a crucial
role in managing the outbreak and maintaining basic social
consumption needs. To minimize the risk of infection, unmanned
intelligent sorting systems are suggested to be implemented. This
paper covers the developing process of an automatic stacking
system based on ABB IRB 120 robot with two auxiliary functions:
synchronous data processing and digital twin simulation. Detailed
logic and method will be introduced, and relevant figures will be
presented. The project will be evaluated and improvement will be
suggested at the conclusion of this paper.
Federated learning (FL) is a cutting-edge artificial intelligence approach. It is a decentralized problem-solving technique that allows users to train using massive data. Unprocessed information is stored in advanced technology by a secret confidentiality service, which incorporates machine learning (ML) training while removing data connections. As researchers in the field promote ML configurations containing a large amount of private data, systems and infrastructure must be developed to improve the effectiveness of advanced learning systems. This study examines FL in-depth, focusing on application and system platforms, mechanisms, real-world applications, and process contexts. FL creates robust classifiers without requiring information disclosure, resulting in highly secure privacy policies and access control privileges. The article begins with an overview of FL. Then, we examine technical data in FL, enabling innovation, contracts, and software. Compared with other review articles, our goal is to provide a more comprehensive explanation of the best procedure systems and authentic FL software to enable scientists to create the best privacy preservation solutions for IoT devices. We also provide an overview of similar scientific papers and a detailed analysis of the significant difficulties encountered in recent publications. Furthermore, we investigate the benefits and drawbacks of FL and highlight comprehensive distribution scenarios to demonstrate how specific FL models could be implemented to achieve the desired results.
In federated learning (FL), model weights must be updated at local users and the base station (BS). These weights are subjected to uplink (UL) and downlink (DL) transmission errors due to the limited reliability of wireless channels. In this paper, we investigate the impact of imperfections in both UL and DL links. First, for a multi-user massive multi-input-multi-output (mMIMO) 6G network, employing zero-forcing (ZF) and minimum mean-squared-error (MMSE) schemes, we analyze the estimation errors of weights for each round. A tighter convergence bound on the modelling error for the communication efficient FL algorithm is derived of the order of
, where…
denotes the variance of overall communication error including the quantization noise. The analysis shows that the reliability of DL links is more critical than that of UL links; and the transmit power can be varied in training process to reduce energy consumption. We also vary the number of local training steps, average codeword length after quantization and scheduling policy to improve the communication efficiency. Simulations with image classification problems on MNIST, EMNIST and FMNIST datasets verify the derived bound and are useful to infer the minimum SNR required for successful convergence of the FL algorithm.
This paper presents microstrip patch antenna designed using computer simulation technology (CST) Microwave Studio at a resonant frequency of 2.4 GHz. The antenna consists of three layers, the upper layer called metallic patch, the bottom layer called ground and the dielectric layer in between the conduction layers called substrate. The antenna has advantage of minimal weight, low profile and can maintain high performance over a wide spectrum of frequencies. As such the study focused on the novel design of a rectangular microstrip patch antenna. The performance characteristics of the antenna patch arrays elements 1X2, 1X4 and 2X2 were compared. The aim of designing antenna with improved gain, reduced losses and use for X band applications such as radar, satellite, communication, medical applications and other wireless systems was achieved. The performance of the designed antenna in terms of radiation efficiency, gain, reflection coefficients and radiation patterns were verified and found suitable for wireless local area network (WLAN) applications.
The next-generation of wireless networks will enable many machine learning (ML) tools and applications to efficiently analyze various types of data collected by edge devices for inference, autonomy, and decision making purposes. However, due to resource constraints, delay limitations, and privacy challenges, edge devices cannot offload their entire collected datasets to a cloud server for centrally training their ML models or inference purposes. To overcome these challenges, distributed learning and inference techniques have been proposed as a means to enable edge devices to collaboratively train ML models without raw data exchanges, thus reducing the communication overhead and latency as well as improving data privacy. However, deploying distributed learning over wireless networks faces several challenges including the uncertain wireless environment (e.g., dynamic channel and interference), limited wireless resources (e.g., transmit power and radio spectrum), and hardware resources (e.g., computational power). This paper provides a comprehensive study of how distributed learning can be efficiently and effectively deployed over wireless edge networks. We present a detailed overview of several emerging distributed learning paradigms, including federated learning, federated distillation, distributed inference, and multi-agent reinforcement learning. For each learning framework, we first introduce the motivation for deploying it over wireless networks. Then, we present a detailed literature review on the use of communication techniques for its efficient deployment. We then introduce an illustrative example to show how to optimize wireless networks to improve its performance. Finally, we introduce future research opportunities. In a nutshell, this paper provides a holistic set of guidelines on how to deploy a broad range of distributed learning frameworks over real-world wireless communication networks.
The tactile internet (TI) is believed to be the prospective advancement of the internet of things (IoT), comprising human-to-machine and machine-to-machine communication. TI focuses on enabling real-time interactive techniques with a portfolio of engineering, social, and commercial use cases. For this purpose, the prospective 5th generation (5G) technology focuses on achieving ultra-reliable low latency communication (URLLC) services. TI applications require an extraordinary degree of reliability and latency. The 3rd generation partnership project (3GPP) defines that URLLC is expected to provide 99.99% reliability of a single transmission of 32 bytes packet with a latency of less than one millisecond. 3GPP proposes to include an adjustable orthogonal frequency division multiplexing (OFDM) technique, called 5G new radio (5G NR), as a new radio access technology (RAT). Whereas, with the emergence of a novel physical layer RAT, the need for the design for prospective next-generation technologies arises, especially with the focus of network intelligence. In such situations, machine learning (ML) techniques are expected to be essential to assist in designing intelligent network resource allocation protocols for 5G NR URLLC requirements. Therefore, in this survey, we present a possibility to use the federated reinforcement learning (FRL) technique, which is one of the ML techniques, for 5G NR URLLC requirements and summarizes the corresponding achievements for URLLC. We provide a comprehensive discussion of MAC layer channel access mechanisms that enable URLLC in 5G NR for TI. Besides, we identify seven very critical future use cases of FRL as potential enablers for URLLC in 5G NR.
The term Federated Learning was coined as recently as 2016 to describe a machine learning setting where multiple entities collaborate in solving a machine learning problem, under the coordination of a central server or service provider. Each client’s raw data is stored locally and not exchanged or transferred; instead, focused updates intended for immediate aggregation are used to achieve the learning objective. Since then, the topic has gathered much interest across many different disciplines and the realization that solving many of these interdisciplinary problems likely requires not just machine learning but techniques from distributed optimization, cryptography, security, differential privacy, fairness, compressed sensing, systems, information theory, statistics, and more. This monograph has contributions from leading experts across the disciplines, who describe the latest state-of-the art from their perspective. These contributions have been carefully curated into a comprehensive treatment that enables the reader to understand the work that has been done and get pointers to where effort is required to solve many of the problems before Federated Learning can become a reality in practical systems. Researchers working in the area of distributed systems will find this monograph an enlightening read that may inspire them to work on the many challenging issues that are outlined. This monograph will get the reader up to speed quickly and easily on what is likely to become an increasingly important topic: Federated Learning.
Federated learning (FL) is a machine learning setting where many clients (e.g. mobile devices or whole organizations) collaboratively train a model under the orchestration of a central server (e.g. service provider), while keeping the training data decentralized. FL embodies the principles of focused data collection and minimization, and can mitigate many of the systemic privacy risks and costs resulting from traditional, centralized machine learning and data science approaches. Motivated by the explosive growth in FL research, this paper discusses recent advances and presents an extensive collection of open problems and challenges. * Peter Kairouz and H. Brendan McMahan conceived, coordinated, and edited this work. Correspondence to kairouz@ google.com and mcmahan@google.com.
We present AGGREGATHOR, a framework that implements state-of-the-art robust (Byzantine-resilient) distributed stochastic gradient descent. Following the standard parameter server model, we assume that a minority of worker machines can be controlled by an adversary and behave arbitrarily. Such a setting has been theoretically studied with several of the existing approaches using a robust aggregation of the workers’ gradient estimations. Yet, the question is whether a Byzantine-resilient aggregation can leverage more workers to speed-up learning. We answer this theoretical question, and implement these state-of-the-art theoretical approaches on AGGREGATHOR, to assess their practical costs. We built AGGREGATHOR around TensorFlow and introduce modifications for vanilla TensorFlow towards making it usable in an actual Byzantine setting. AGGREGATHOR also permits the use of unreliable gradient transfer over UDP to provide further speed-up (without losing the accuracy) over the native communication protocols (TCP-based) of TensorFlow in saturated networks. We quantify the overhead of Byzantine resilience of AGGREGATHORto 19% and 43% (to ensure weak and strong Byzantine resilience respectively) compared to vanilla TensorFlow.
We design a novel, communication-efficient, failure-robust protocol for secure aggregation of high-dimensional data. Our protocol allows a server to compute the sum of large, user-held data vectors from mobile devices in a secure manner (i.e. without learning each user's individual contribution), and can be used, for example, in a federated learning setting, to aggregate user-provided model updates for a deep neural network. We prove the security of our protocol in the honest-but-curious and active adversary settings, and show that security is maintained even if an arbitrarily chosen subset of users drop out at any time. We evaluate the efficiency of our protocol and show, by complexity analysis and a concrete implementation, that its runtime and communication overhead remain low even on large data sets and client pools. For 16-bit input values, our protocol offers $1.73 x communication expansion for 2¹⁰ users and 2²⁰-dimensional vectors, and 1.98 x expansion for 2¹⁴ users and 2²⁴-dimensional vectors over sending data in the clear.
In recent years, a branch of machine learning called Deep Learning has become incredibly popular thanks to the ability of a new class of algorithms to model and interpret a large quantity of data in a similar way to humans. Properly training deep learning models involves collecting a vast amount of users' private data, including habits, geographical positions, interests, and much more. Another major issue is that it is possible to extract from trained models useful information about the training set and this hinders collaboration among distrustful participants or parties that deal with sensitive information. To tackle this problem, collaborative deep learning models have recently been proposed where parties share only a subset of the parameters in the attempt to keep their respective training sets private. Parameters can also be obfuscated via differential privacy to make information extraction even more challenging, as shown by Shokri and Shmatikov at CCS'15. Unfortunately, we show that any privacy-preserving collaborative deep learning is susceptible to a powerful attack that we devise in this paper. In particular, we show that a distributed or decentralized deep learning approach is fundamentally broken and does not protect the training sets of honest participants. The attack we developed exploits the real-time nature of the learning process that allows the adversary to train a Generative Adversarial Network (GAN) that generates valid samples of the targeted training set that was meant to be private. Interestingly, we show that differential privacy applied to shared parameters of the model as suggested at CCS'15 and CCS'16 is utterly futile. In our generative model attack, all techniques adopted to scramble or obfuscate shared parameters in collaborative deep learning are rendered ineffective with no possibility of a remedy under the threat model considered.
Emerging approaches to network monitoring involve large numbers of agents collaborating to produce performance or security related statistics on huge, partial mesh networks. The aggregation process often involves security or business-critical information which network providers are generally unwilling to share without strong privacy protection. We present efficient and scalable protocols for privately computing a large range of aggregation functions based on addition, disjunction, and max/min. For addition, we give a protocol that is information-theoretically secure against a passive adversary, and which requires only one additional round compared to non-private protocols for computing sums. For disjunctions, we present both a computationally secure, and an information-theoretically secure solution. The latter uses a general composition approach which executes the sum protocol together with a standard multi-party protocol for a complete subgraph of "trusted servers". This can be used, for instance, when a large network can be partitioned into a smaller number of provider domains.
Federated learning (FL) is a machine learning setting where many clients (e.g., mobile devices or whole organizations) collaboratively train a model under the orchestration of a central server (e.g., service provider), while keeping the training data decentralized. FL embodies the principles of focused data collection and minimization, and can mitigate many of the systemic privacy risks and costs resulting from traditional, centralized machine learning and data science approaches. Motivated by the explosive growth in FL research, this monograph discusses recent advances and presents an extensive collection of open problems and challenges.
Machine Learning (ML) based Quality of Experience (QoE) models potentially suffer from over-fitting due to limitations including low data volume, and limited participant profiles. This prevents models from becoming generic. Consequently, these trained models may under-perform when tested outside the experimented population. One reason for the limited datasets, which we refer in this paper as small QoE data lakes, is due to the fact that often these datasets potentially contain user sensitive information and are only collected throughout expensive user studies with special user consent. Thus, sharing of datasets amongst researchers is often not allowed. In recent years, privacy preserving machine learning models have become important and so have techniques that enable model training without sharing datasets but instead relying on secure communication protocols. Following this trend, in this paper, we present Round-Robin based Collaborative Machine Learning model training, where the model is trained in a sequential manner amongst the collaborated partner nodes. We benchmark this work using our customized Federated Learning mechanism as well as conventional Centralized and Isolated Learning methods.
Deep Learning has recently become hugely popular in machine learning for its ability to solve end-to-end learning systems, in which the features and the classifiers are learned simultaneously, providing significant improvements in classification accuracy in the presence of highly-structured and large databases.
Its success is due to a combination of recent algorithmic breakthroughs, increasingly powerful computers, and access to significant amounts of data.
Researchers have also considered privacy implications of deep learning. Models are typically trained in a centralized manner with all the data being processed by the same training algorithm. If the data is a collection of users' private data, including habits, personal pictures, geographical positions, interests, and more, the centralized server will have access to sensitive information that could potentially be mishandled. To tackle this problem, collaborative deep learning models have recently been proposed where parties locally train their deep learning structures and only share a subset of the parameters in the attempt to keep their respective training sets private. Parameters can also be obfuscated via differential privacy (DP) to make information extraction even more challenging, as proposed by Shokri and Shmatikov at CCS'15.
Unfortunately, we show that any privacy-preserving collaborative deep learning is susceptible to a powerful attack that we devise in this paper. In particular, we show that a distributed, federated, or decentralized deep learning approach is fundamentally broken and does not protect the training sets of honest participants. The attack we developed exploits the real-time nature of the learning process that allows the adversary to train a Generative Adversarial Network (GAN) that generates prototypical samples of the targeted training set that was meant to be private (the samples generated by the GAN are intended to come from the same distribution as the training data). Interestingly, we show that record-level differential privacy applied to the shared parameters of the model, as suggested in previous work, is ineffective (i.e., record-level DP is not designed to address our attack).
Deeper neural networks are more difficult to train. We present a residual
learning framework to ease the training of networks that are substantially
deeper than those used previously. We explicitly reformulate the layers as
learning residual functions with reference to the layer inputs, instead of
learning unreferenced functions. We provide comprehensive empirical evidence
showing that these residual networks are easier to optimize, and can gain
accuracy from considerably increased depth. On the ImageNet dataset we evaluate
residual nets with a depth of up to 152 layers---8x deeper than VGG nets but
still having lower complexity. An ensemble of these residual nets achieves
3.57% error on the ImageNet test set. This result won the 1st place on the
ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100
and 1000 layers.
The depth of representations is of central importance for many visual
recognition tasks. Solely due to our extremely deep representations, we obtain
a 28% relative improvement on the COCO object detection dataset. Deep residual
nets are foundations of our submissions to ILSVRC & COCO 2015 competitions,
where we also won the 1st places on the tasks of ImageNet detection, ImageNet
localization, COCO detection, and COCO segmentation.
We present the SIGMA family of key-exchange protocols and the “SIGn-and-MAc” approach to authenticated Diffie-Hellman underlying its design. The SIGMA protocols provide perfect forward secrecy via a Diffie-Hellman exchange authenticated with digital signatures, and are specifically designed to ensure sound cryptographic key exchange while providing a variety of features and trade-offs required in practical scenarios (such as optional identity protection and reduced number of protocol rounds). As a consequence, the SIGMA protocols are very well suited for use in actual applications and for standardized key exchange. In particular, SIGMA serves as the cryptographic basis for the signature-based modes of the standardized Internet Key Exchange (IKE) protocol (versions 1 and 2).
This paper describes the design rationale behind the SIGMA approach and protocols, and points out to many subtleties surrounding the design of secure key-exchange protocols in general, and identity-protecting protocols in particular. We motivate the design of SIGMA by comparing it to other protocols, most notable the STS protocol and its variants. In particular, it is shown how SIGMA solves some of the security shortcomings found in previous protocols.
Communication-efficient learning of deep networks from decentralized data
- mcmahan
Privacy-aware machine learning with low network footprint
- K Vandikas
- S Ickin
Artificial intelligence and machine learning in next-generation systems
- E Fersman
- J Forgeat
Privacy in mobile networks — How to embrace privacy by design
- M Anneroth
- D Casella
Towards federated learning at scale: System design
- bonawitz
Quantification of the Leakage in Federated Learning
- li