Harnessing the benefits of blockchain, such as decentralization, immutability, and transparency, to bolster the credibility and security attributes of federated learning (FL) has garnered increasing attention. However, blockchain-enabled FL (BFL) still faces several challenges. The primary and most significant issue arises from its essential but slow validation procedure, which selects high-quality local models by recruiting distributed validators. The second issue stems from its incentive mechanism under the transparent nature of blockchain, increasing the risk of privacy breaches regarding workers’ cost information. The final challenge involves data eavesdropping from shared local models. To address these significant obstacles, this paper proposes a Blockchain-enabled Incentivized and Secure Federated Learning (BIT-FL) framework. BIT-FL leverages a novel loop-based sharded consensus algorithm to accelerate the validation procedure, ensuring the same security as non-sharded consensus protocols. It consistently outputs the correct local model selection when the fraction of adversaries among validators is less than$1/2$with synchronous communication. Furthermore, BIT-FL integrates a randomized incentive procedure, attracting more participants while guaranteeing the privacy of their cost information through meticulous worker selection probability design. Finally, by adding artificial Gaussian noise to local models, it ensures the privacy of trainers’ local models. With the careful design of Gaussian noise, the excess empirical risk of BIT-FL is upper-bounded by$\mathcal {O}(\frac{\ln n_{\min}}{ n_{\min}^{3/2}}+\frac{\ln n}{n})$, where$n$represents the size of the union dataset, and$n_{{\min}}$represents the size of the smallest dataset. Our extensive experiments demonstrate that BIT-FL exhibits efficiency, robustness, and high accuracy for both classification and regression tasks.

Network management relies on extensive monitoring of network state to analyse network behavior, design optimizations, plan upgrades, and conduct troubleshooting. Network monitoring collects various data from network devices through different protocols and interfaces such as NETCONF and Syslog, and from monitoring tools such as Zeek and Osquery. To unify and automate the monitoring workflow across the network, this paper identifies and discusses the data collection requirements for network management, reviews different monitoring approaches, and proposes an efficient data collection platform that addresses the requirements through an extensible and lightweight protocol. The platform design is demonstrated through an adaptive collection of data for network management based on digital twin technology.

Real-time traffic has stringent requirements in terms of latency, and deadline guarantees on packet delivery play a vital role in real-time IoT applications. Deadline-aware wireless scheduling of real-time traffic has been a long-standing open problem, despite significant efforts using analytical methods. Departing from the conventional approaches, this work studies deadline-aware traffic scheduling by taking an offline reinforcement learning (RL) approach to train scheduling algorithms, ready to be used for online scheduling. To address the challenges therein, we propose a network-aided offline RL (NA-ORL) framework for deadline-aware scheduling, by making use of the fact that the network dynamics follows a well-defined physics model. Specifically, in NA-ORL the initialization of the scheduling policy is obtained through behavior cloning with a good model-based scheduling algorithm, and the network-aided actor–critic (A–C) method is utilized to train a better scheduling policy with carefully designed states and reward function, thanks to its nature of policy improvement. Building on NA-ORL, we further devise a network-aided offline meta-RL (NA-MRL) algorithm to deal with the nonstationary network dynamics. Extensive experimental results demonstrate that the proposed NA-ORL and NA-MRL algorithms can achieve better performance over adaptive mixing over nondominated links (AMIX-ND) and largest-deficit-first (LDF), in various scenarios for the deadline-aware wireless scheduling.

With the development of next-generation wireless networks, the Internet of Things (IoT) is evolving towards the intelligent IoT (iIoT), where intelligent applications usually have stringent delay and jitter requirements. In order to provide low-latency services to heterogeneous users in the emerging iIoT, multi-tier computing was proposed by effectively combining edge computing and fog computing. More specifically, multi-tier computing systems compensate for cloud computing through task offloading and dispersing computing tasks to multi-tier nodes along the continuum from the cloud to things. In this paper, we investigate key techniques and directions for wireless communications and resource allocation approaches to enable task offloading in multi-tier computing systems. A multi-tier computing model, with its main functionality and optimization methods, is presented in detail. We hope that this paper will serve as a valuable reference and guide to the theoretical, algorithmic, and systematic opportunities of multi-tier computing towards next-generation wireless networks.

Multi-tier computing effectively enables flexible computation and communication resource sharing by offloading computation-intensive tasks to nearby servers along the cloud-to-thing continuum. In essence, multi-tier computing networks can distribute computing, storage, and communication functions anywhere between the cloud and the endpoint to take full advantage of the resources available along this continuum, thus extending the traditional cloud computing architecture to the edge of the network. With multi-tier computing, some application component processing, such as delay-sensitive components, can take place at the edge of the network, while other components, such as time-tolerant and computation-intensive components, can be performed in the cloud. To best meet user requirements, centralized cloud computing with extensive resources, secure environments, and powerful algorithms is still needed, but also must be complemented by distributed fog and edge computing with shared resources, accessible environments, and simple algorithms for real-time decision-making. Given heterogeneous computing resources and collaborative service architectures, future multi-tier computing networks will be capable of supporting a full range of computing and networking services for different environments and applications. This Special Issue aims to provide a forum for the latest advances in multi-tier computing for next-generation wireless network research, innovations, and applications. Multi-tier computing enables low-latency processing by allowing data to be processed at the network edge close to end devices. It also facilitates the distribution of fog/edge nodes to collect data from end devices. Therefore, multi-tier computing effectively complements the cloud computing architecture.

Multi-tier computing effectively enables flexible computation and communication resource sharing by offloading computation-intensive tasks to nearby servers along the cloud-to-thing continuum. In essence, multi-tier computing networks can distribute computing, storage, and communication functions anywhere between the cloud and the endpoint to take full advantage of the resources available along this continuum, thus extending the traditional cloud computing architecture to the edge of the network. With multi-tier computing, some application component processing, such as delay-sensitive components, can take place at the edge of the network, while other components, such as time-tolerant and computation-intensive components, can be performed in the cloud. To best meet user requirements, centralized cloud computing with extensive resources, secure environments, and powerful algorithms is still needed, but also must be complemented by distributed fog and edge computing with shared resources, accessible environments, and simple algorithms for real-time decision-making. Given heterogeneous computing resources and collaborative service architectures, future multi-tier computing networks will be capable of supporting a full range of computing and networking services for different environments and applications. Multi-tier computing enables low-latency processing by allowing data to be processed at the network edge close to end devices. It also facilitates the distribution of fog/edge nodes to collect data from end devices. Therefore, multi-tier computing effectively complements the cloud computing architecture.

Providing differentiated services to meet the unique requirements of different use cases is a major goal of the fifth generation (5G) telecommunication networks and will be even more critical for future 6G systems. Fulfilling this goal requires the ability to assure quality of service (QoS) end to end (E2E), which remains a challenge. A key factor that makes E2E QoS assurance difficult in a telecommunication system is that access networks (ANs) and core networks (CNs) manage their resources autonomously. So far, few results have been available that can ensure E2E QoS over autonomously managed ANs and CNs. Existing techniques rely predominately on each subsystem to meet static local QoS budgets with no recourse in case any subsystem fails to meet its local budgets and, hence will have difficulty delivering E2E assurance. Moreover, most existing distributed optimization techniques that can be applied to assure E2E QoS over autonomous subsystems require the subsystems to exchange sensitive information such as their local decision variables. This paper presents a novel framework and a distributed algorithm that can enable ANs and CNs to autonomously "cooperate" with each other to dynamically negotiate their local QoS budgets and to collectively meet E2E QoS goals by sharing only their estimates of the global constraint functions, without disclosing their local decision variables. We prove that this new distributed algorithm converges to an optimal solution almost surely, and also present numerical results to demonstrate that the convergence occurs quickly even with measurement noise.

Recently, vehicle teleoperation (remote driving) has been gaining momentum in the industry and is playing an important role in helping self-driving cars navigate through tricky situations they cannot handle on their own. Today, vehicle teleoperation relies fully on human teleoperators. Future teleoperation systems, powered by artificial intelligence and advanced networks, such as 5G, can potentially allow a subset of the automated driving intelligence to be off-loaded from the vehicle into the cloud. Over time, the growing level of remote driving automation can enable new automated driving solutions that are not feasible today. This article presents a vision of such intelligent teleoperation and discusses its benefits and challenges.

The papers in this special section focus on fog computing and services. The emerging Internet of Things (IoT) and rich cloud services have helped create the need for fog computing (also known as edge computing), in which data processing occurs in part at the network edge or anywhere along the cloud-to-endpoint continuum that can best meet user requirements, rather than completely in a relatively small number of massive clouds. Fog computing could address latency concerns, devices’ limited processing and storage capabilities and battery life, network bandwidth constraints and costs, and many security and privacy concerns that arise from the emerging IoT.

Today's GPS navigators have difficulty determining altitude such as whether a car is under or on an overpass. In this paper, we show how to use the barometers, which come with off-the-shelf smartphones, to improve vertical positioning accuracy. We will present performance testing results from different real road scenarios to show the feasibility of the proposed method.

Fog is an emergent architecture for computing, storage, control, and networking that distributes these services closer to end users along the cloud-to-things continuum. It covers both mobile and wireline scenarios, traverses across hardware and software, resides on network edge but also over access networks and among end users, and includes both data plane and control plane. As an architecture, it supports a growing variety of applications, including those in the Internet of Things (IoT), fifth-generation (5G) wireless systems, and embedded artificial intelligence (AI). This survey paper summarizes the opportunities and challenges of fog, focusing primarily in the networking context of IoT.

Vehicles face growing security threats as they become increasingly connected with the external world. Hackers, researchers, and car hobbyists have compromised security keys used by the electronic control units (ECUs) on vehicles, modified ECU software, and hacked wireless transmissions from vehicle key fobs and tire monitoring sensors, using low-cost commercially available tools. However, the most damaging security threats to vehicles are only emerging. One such threat is malware, which can infect vehicles in a variety of ways and cause severe consequences. Defending vehicles against malware attacks must address many unique challenges that have not been well addressed in other types of networks. This paper identifies those vehicle-specific challenges, discusses existing solutions and their limitations, and presents a cloud-assisted vehicle malware defense framework that can address these challenges.

Dedicated Short-Range Communications (DSRC) has been designed to support vehicular communications. In the U.S., DSRC operates in the 5.9 GHz licensed spectrum band. Its physical (PHY) and medium access control (MAC) layers, defined in the IEEE 802.11p standard, are based on the IEEE 802.11 family of Wi-Fi standards. Vehicular communication environments differ significantly from the sparse and low-velocity nomadic use cases of a typical Wi-Fi deployment. Thus, there are many challenges to adapt Wi-Fi technologies to support the unique requirements of vehicular communications such as achieving high and reliable performance in highly mobile, often densely populated, and frequently non-line-of-sight environments. The automotive and the communications industries, academia, and governments around the world have been devoting tremendous efforts to address these challenges, and significant achievements have been made. Remaining challenges can be addressed by the future versions of DSRC. In this paper, we investigate the current technologies used by DSRC to support vehicle safety communications, analyze existing and possible DSRC performance enhancements that can be realized in the near term, and provide a few initial thoughts on the DSRC evolution path.

Random graphs are often used to model vehicular networks. However, their applicability has been limited because it is difficult to express and instantiate parameters of graph models of vehicular networks using real-life data. In this paper, we consider using random geometric graphs to model vehicular networks where vehicle movements are constrained to a road system. We show that vehicles form a random geometric graph with edge probability p that can be expressed as a closed-form expression or as an algorithmically computable expression with parameters that are known or easily measurable in real life. This enables one to answer essential questions, such as questions related to routing and placement of mobile nodes required to detect malicious parties in vehicular communications, as a function of practically measurable and computable parameters.

Motivated by vehicular networking applications, we study a novel type of privacy-preserving public-key infrastructures where the server that distributes public and private keys to clients need not be trusted by clients to protect its secret data against intruders. We target three main requirements for these public-key infrastructures: privacy preservation of the client's identity (or, anonymity), traceability of malicious messages to clients corrupted by an attacker rather than honest clients (or, traceability) and communication and computation time constant with respect to the number of users (or, efficiency). This combination of properties was not achieved in previously studied areas such as broadcast or multicast encryption, group signatures and ring signatures. This paper designs a public key infrastructure that achieves satisfactory performance on all three requirements, based on key pools and a novel probabilistic key revocation and update strategy. Perhaps surprisingly, we use a careful design of our revocation protocol to achieve a combination of properties for public-key infrastructures that was not previously achieved.

Abstract As vehicular networks approach deployment phases, there is wide recognition for challenges and pressing needs for solutions with respect to the areas of security, privacy, and performance. The requirement of participation in applications such as traffic safety, combined with the intrinsically ad hoc network environment, poses rather novel challenges to the modeling, design, and analysis of security solutions for vehicular networks. One particularly stringent requirement in this area is that of protecting the privacy of vehicle owners (i.e., their anonymity and their vehicle's location unlinkability) during their participation in a vehicular network such as in traffic safety applications. In this paper, we present novel models of concrete anonymity and unlinkability requirements for vehicular networks that can be built using state‐of‐the‐art options for communication network deployment (e.g., road‐side short‐range radio networks and hotspot networks). One key aspect of our modeling consists of recognizing the existence and impact of additional certification authorities managed by vehicle manufacturers. In particular, we consider a simple variant of a previously proposed public‐key infrastructure (PKI) for vehicular networks and present two techniques to augment it so that it provides improved anonymity and unlinkability properties. The resulting vehicular‐network key infrastructures satisfy desirable combinations of anonymity, unlinkability, bad actor detection, and performance requirements. Copyright © 2010 John Wiley & Sons, Ltd.

Vehicular networks require secure communication, especially for safety applications. A public key infrastructure using a Combinatorial Certificate Scheme was implemented in the US Vehicle Infrastructure Integration (VII) Proof-of- Concept (PoC) trial to secure V2V communication and preserve vehicle privacy. This paper analyzes the privacy and scalability of the Combinatorial Certificate approach for a nationwide network of 200 million vehicles. It examines the tradeoffs between privacy, the ability to efficiently detect and remove bad actors, and the need to minimize the impact on innocent vehicles due to revocation and replacement of compromised shared certificates. Key findings include the level of vehicle anonymity that exists in situations of low vehicular density and the impact that certificate revocations have on innocent vehicles. A refinement to the Combinatorial Certificate Scheme is described that improves the innocent vehicle re-key quota lifetime by an order of magnitude.

In this paper, we propose a peer-to-peer service discovery protocol specially designed for mobile ad-hoc network (MANET) environment. The proposed mobile service discovery protocol (MSDP) uses a dynamic clustering algorithm to group nodes in a MANET, and utilizes distributed hash tables (DHTs) to efficiently cache service information in a peer-to-peer manner. Simulation results under different scenarios show that MSDP has steady performance and is able to reduce message overhead efficiently

Vehicular networks have attracted extensive attentions in recent years for their promises in improving safety and enabling other value-added services. Security and privacy are two integrated issues in the deployment of vehicular networks. Privacy-preserving authentication is a key technique in addressing these two issues. We propose a random keyset based authentication protocol that preserves user privacy under the zero-trust policy, in which no central authority is trusted with the user privacy. We show that the protocol can efficiently authenticate users without compromising their privacy with theoretical analysis. Malicious user identification and key revocation are also described

As vehicular networks approach practicality, there is wide recognition for security challenges in their use, and pressing need for security solutions. The intrinsically mobile and ad-hoc nature of vehicular networks pose rather novel challenges to the modeling, design and analysis of security solutions for them. One particularly stringent requirement in this area is that of protecting the anonymity of vehicle owners during their participation in a vehicular network, such as in traffic safety applications. In this paper we present novel models of concrete anonymity requirements for vehicular networks. We consider a case study of a simple variant of a public-key infrastructure for vehicular networks and present two techniques to augment it so that it provides improved anonymity properties. The resulting vehicular-network key infrastructures satisfy desirable combinations of anonymity, unlinkability, bad actor detection, and efficiency requirements.

Proactive handoff mechanisms involving secure pre-authentication and proactive configuration help reduce the delay and transient data loss for real-time communication during movement between homogeneous or heterogeneous access networks. Pre-authentication is meant to perform authentication with a network before a mobile moves into the network. To achieve secure pre- authentication with a target neighboring network, a mobile needs to obtain an IP address of the authentication server from the target network when the mobile is still outside the target network and then to establish a security association with the authentication agent in the target network. This requires the mobile to discover the parameters of various network elements in the target network ahead of time so that the mobile can communicate with these network elements to establish proactive security associations. We describe several approaches for a mobile to discover the network elements in target networks before moving into these target networks. We also describe how network discovery can help provide fast-handoff using secure pre-authentication and proactive IP address acquisition during handover between different access networks.

IEEE 802.11 wireless LANs have been widely deployed. How to efficiently balance the traffic loads among access points (APs), which will lead to improved network utilization and higher quality of user experience, has become an important issue. A key challenge is how to select an AP to use during the handoff process in a way that will achieve overall load balance in the network. The conventional approaches typically use signal-to-noise ratio (SNR) as the criterion without considering the load of each AP. In this paper, we propose two effective new AP selection algorithms. These algorithms estimate the AP traffic loads by observing and estimating the IEEE 802.11 frame delays and use the results to determine which AP to use. The algorithms can be implemented in software on mobile stations alone. We will show that better performance can be achieved when the APs provide assistance in delay measurements; however, the improvement is not significant. There is no need to exchange information among APs. The proposed algorithms are fully compatible with the existing standards and systems and they are easy to implement. We will present extensive simulation results to show that the proposed algorithms can significantly increase overall system throughput and reduce average frame delay.

This paper considers how uncontrolled mobiles can be used to collaboratively accomplish sensing tasks. Uncontrolled mobiles are mobile devices whose movements cannot be easily controlled for the purpose of achieving a task. Examples include sensors mounted on mobile vehicles of people to monitor air quality and to detect potential airborne nuclear, biological, or chemical agents. We describe an approach for using uncontrolled mobile devices for collaborative sensing. Considering the potentially large number of mobile sensors that may be required to monitor a large geographical area such as a city, a key issue is how to achieve a proper balance between performance and costs. We present analytical results on the rate of information reporting by uncontrolled mobile sensors needed to cover a given geographical area. We also present results from testbed implementations to demonstrate the feasibility of using existing low-cost software technologies and platforms with existing standard protocols for information reporting and retrieval to support a large system of uncontrolled mobile sensors

Emergence of mobile handheld devices with WiFi (IEEE 802.11) interfaces such as WiFi phones, WiFi-cellular dual-mode phones and PDAs (personal digital assistants), allow users to take full advantages of heterogeneous radio technologies. WiFi, however, was not originally designed for energy-constrained handheld devices. As a result, the standby times of a handheld device with a WiFi interface is significantly lower than what people typically experience with today's cellular phones. This paper presents and evaluates a system-level power management method devices with WiFi, or other high energy-consuming network interfaces, to significantly increase their standby times without modifications to WiFi and upper layer protocols and implementations.

In order to support session-based real-time communication in a highly mobile environment it is desirable to limit end-to-end delay, jitter and packet loss at a certain threshold level. This paper describes a framework of media-independent pre-authentication (MPA), a new handover optimization mechanism that has a potential to address issues on existing mobility management protocols and mobility optimization mechanisms to achieve these values. MPA is a mobile-assisted, secure handover optimization scheme that works over any link-layer and with any mobility management protocol. This paper also presents an initial implementation of MPA and performance results to show how existing protocols could be leveraged to realize the functionalities of MPA and provide the desired results.

The paper describes the architecture of toll-free IP (TIP) and the use of the diameter base protocol defined by IETF to implement the key protocol interfaces in the architecture. TIP is a new approach to providing no-charge-to-user ("toll-free") access over IP-based access networks (e.g., IEEE 802.11 and 802.16 wireless local area networks). Usage of IP access networks to communicate with a toll-free IP destination (identified by, for example, a toll-free IP address, Internet Domain Name, or Universal Resource Locator) is paid by the toll-free destination's owner. TIP allows one to use an IP access network without a service subscription, and without the need to authenticate with the access network or service provider (hence, no need for terminal software or hardware for authentication with access networks). It allows one to provide services over IP-based access networks without any business arrangement with the network operators or aggregators. TIP can be easily incorporated into, and hence coexist with, existing network access control frameworks.

Emerging mobile devices are equipped with multiple interfaces allowing users to take advantage of heterogeneous radio technologies (e.g., cellular networks and wireless LANs). To support mobility and quality of service (QoS) for such multi-interface mobile devices, it is important to dynamically determine which interface or interfaces a mobile should use and when to switch from one interface to another. To make such decisions, it is critical for the mobile to compare the "qualities of the interfaces (QoI)", which may be measured by a variety of metrics, in real time so that the mobile can switch from one interface to another without user perceivable interruptions to on-going applications. This paper presents and analyzes new techniques for a mobile device to make rapid comparison of the quality of the interfaces using quickest sequential change detection and sequential two-sample hypotheses tests.

This paper describes an IP paging protocol that can be independent of, and also easily optimized for, any mobility management protocol. It is scalable, runs over any radio system, supports arbitrary location update and paging algorithms, and allows IP paging areas to map arbitrarily to IP subnets. Our analysis results show that the protocol can haves significantly lower signaling cost than mobile IP foreign agent paging methods such as P-MIP. This paper also identifies the shortcomings of using IPsec to secure IP paging protocols and proposes a new solution to overcome these shortcomings.

In a mobile wireless ad-hoc network, nodes move about and relay packets destined for other nodes. One of the biggest challenges in this area is the design of scalable routing protocols. A family of routing protocols has emerged that is potentially more scalable than the protocols that discover and/or maintain end-to-end routes. These protocols are location-based-the network maintains nodes' approximate geographic locations and use the information to route packets. An important component of these protocols is the management of the location information at network nodes. We present one such scheme called SLALoM which scales well in large, mobile ad-hoc networks. In particular, we prove that under a specific environment the overhead cost of SLALoM is asymptotically lower than SLURP, the only other location management scheme that has been analyzed theoretically. We also provide simulation results that show our scheme performs well in comparison to SLURP under a variety of scenarios not incorporated into the analysis.

This paper presents a new method for predicting resource requirements of future handoff calls in multimedia wireless IP networks. It allows call arrivals to be non-Poisson and non-stationary and calls to have arbitrary bandwidth needs. Statistical models, based on time series forecasting theory, that utilize only local information are developed. Our simulations show that the proposed method achieves close correlation between predicted and actual handoff resource demands even when demands fluctuate rapidly over a wide range. Furthermore, it can generate accurate dynamic reservation levels to provide quality/grade of service guarantees for handoff calls. The approach is simple and can be implemented easily.

This paper presents a new method for predicting resource requirements of and resenting resources for future handoff calls in multimedia wireless networks. It allows handoff calls to (1) follow non-Poisson and/or non-stationary arrival processes, (2) have arbitrary resource demands, and (3) have arbitrarily distributed call and channel holding times in any cell. A statistical model, based on Wiener prediction theory, that uses only local information is developed. Our simulations show that it performs well under the above conditions as designed. Furthermore, for Poisson handoff call arrivals, identical per-call bandwidth demands and exponential call and channel holding times, it generates comparable performance with an existing method optimized only for these conditions.

This paper presents two new methods that use local information alone to predict the resource demands of and determine resource reservation levels for future handoff calls in multimedia wireless IP networks. The proposed methods model the instantaneous resource demand directly. This differs from most existing methods that derive the demands from modeling the factors that impact the demands. As a result, the proposed methods allow new and handoff calls to: (1) follow non-Poisson and/or nonstationary arrival processes; (2) have arbitrary per-call resource demands; and (3) have arbitrarily distributed call and channel holding times. The first method is based on the Wiener prediction theory and the second method is based on time series analysis. Our simulations show that they perform well even for non-Poisson and nonstationary handoff call arrivals, arbitrary per-call bandwidth demands, and nonexponentially distributed call and channel holding times. They generate closely comparable performance with an existing local method and an existing collaborative method that uses information about mobiles in neighboring cells, under assumptions for which these other methods are optimized. The proposed methods are much simpler to implement than most other existing methods with fewer capabilities.