Tingting Liu

Computation offloading has been extensively studied in recent years for the Internet of Vehicles (IoV), where roadside units (RSUs) are deployed to assist computation offloading. However, it is still challenging to decide when to offload regarding multiple factors, such as load differences among RSUs, a vehicle’s moving speed, and a vehicle’s energy constraint. In this article, an optimal offloading decision method is proposed based on optimal stopping theory (OST) to decide when to offload considering the aforementioned factors. First, two offloading decision problems with and without energy constraint are constructed to find the optimal RSU that can minimize expected cost, where the expected cost is determined by the decision on offloading to the current RSU or continuing observing the next RSU. Then, OST is utilized to solve these two problems. Specifically, a sequence of thresholds is precalculated based on the OST. An offloading decision can be made by comparing the load of the current RSU with the threshold. Moreover, some facts on a vehicle’s moving speed in the environment without energy constraint and the number of observations on RSUs in the environment with energy constraint are revealed. What is more, the optimal moving speed that can minimize the expected cost is also provided. Finally, extensive simulations are conducted to demonstrate the effectiveness of the proposed method. The effects of a vehicle’s moving speed and the number of observations on the performance of the proposed method are also verified. Compared to the benchmarks, the proposed method can achieve superior performance in terms of cost and hit ratio, and has comparable performance with the best offloading method, which has full RSUs’ load information. Moreover, the proposed method is robust to the estimation deviation of RSUs’ load distribution.

Privacy preservation and communication efficiency are two critical issues in making intelligent transportation systems (ITSs) come true. Owning to the recent advances on differential privacy (DP) and intelligent flexible position antenna (IFPA) technologies, it is promising that the aforementioned critical issues can be relieved by introducing DP and IFPA into vehicular federated networks. Still, it faces many challenges to implement such vehicular networks. On the one hand, imposing further protections on vehicular federated networks leads to extra energy consumption of vehicular users, and they may be unwilling to employ DP technology. On the other hand, vehicular user's mobility leads to continuous changes to the network topology, and therefore IFPA optimization should adapt to the variations of network topology. In this paper to overcome those challenges, an integrated compensation minimization and IFPA optimization algorithm is first proposed. The convergence upper bound of this algorithm is analyzed and some facts are revealed that there exists a good value of FL round which can strike a balance between the learning performance and training time, and also more participant vehicles will further minimize the convergence upper bound. Then, a user selection problem is formulated, subject to a limited compensation budget. A truncated scheme is proposed to optimize the vehicular user set. Finally, an integrated user selection and IFPA optimization algorithm is proposed to holistically address the implementation of IFPA-assisted privacy-preserved vehicular federated networks. Extensive simulations are conducted to demonstrate the effectiveness of the proposed algorithms. The simulation results show that the proposed algorithm can achieve significant advantages in minimizing the compensation cost and communication energy consumption, and the proposed integrated user selection and IFPA optimization algorithm can run smoothly and converge while the network topology is continuous changing.

This paper first develops an analytical framework to investigate the performance of uplink (UL)/downlink (DL) decoupled access in cellular vehicle-to-everything (C-V2X) networks, in which a vehicle's UL/DL can be connected to different macro/small base stations (MBSs/SBSs), separately. Using the stochastic geometry analytical tool, the UL/DL decoupled access C-V2X is modeled as a Cox process, and we obtain the following theoretical results, i.e., 1) the probability of different UL/DL joint association cases i.e., both the UL and DL are associated with the different MBSs or SBSs, or they are associated with different types of BSs; 2) the distance distribution of a vehicle to its serving BSs in each case; 3) the spectral efficiency of UL/DL in each case; and 4) the UL/DL coverage probability of MBS/SBS. The analyses reveal the insights and performance gain of UL/DL decoupled access. Through extensive simulations, the accuracy of the proposed analytical framework is validated. Both the analytical and simulation results show that UL/DL decoupled access can improve spectral efficiency. The theoretical results can be directly used for estimating the statistical performance of a UL/DL decoupled access C-V2X network.

Owning to the powerful support of 5 G/B5G technologies, it is promising that, in a vehicular network, the ever-increasing demands for artificial intelligence applications along with privacy concerns can be fulfilled by introducing federated learning (FL) into vehicular networks. However, in such a 5 G/B5G cellular-based vehicular network, distributed clients, i.e., vehicles, usually suffer from intermittent connectivity to the uplink base station (BS) due to the small cell networking and their mobility, and thereby the FL execution time and the learning performance cannot be guaranteed. In order to ensure a required execution time and the FL performance, we propose to investigate the client selection schemes in the uplink-downlink decoupled 5 G/B5G networks. Due to the bursty feature of uplink transmissions, we first incorporate a flexible delay analytical tool, i.e., martingale theory, to characterize the access delay of each vehicle, and represent it in the form of delay violation probability. Then, given an energy budget, adjustable parameters are optimized to maximize the global learning performance. Supposing that all the clients participate in the FL and they can outperform the systematic FL performance requirement, the client selection methods under different criteria are further investigated. Especially, we propose a low-complexity algorithm to select clients in terms of enhancing energy efficiency. Numerical results are provided to demonstrate the effectiveness of the proposed client selection method in both guaranteeing the FL execution time and meeting the systematic learning performance. The energy efficiency of the vehicular FL system can also be enhanced by employing the proposed method.

With the evolution of industrial technologies, the Internet of Things and blockchain show enormous potential. Different institutes with heterogeneous devices indicate diversity in their private information, including service demand and risk-bearing capability. Low efficiency and rising transaction prices of the public blockchain have been significant barriers to these potential service purchasers. The sidechain is recognized as one of the scalability solutions to a public chain. However, it is challenging to design the economic incentive mechanism and study the transaction fee pricing for transaction services on a sidechain due to the uncertainty of the network's performance and the miner's effort. Considering the asymmetric information (i.e., anonymous private information) and uncertainty, we propose a random-contract-based scheme to maximize the sidechain service provider's revenue and assign the service buyers the feasible service price under the framework of a sidechain linked to the public blockchain. Furthermore, we show that random contracts outperform the deterministic contracts under the increasing absolute risk-aversion assumption. We investigate service pricing in various scenarios using Geometric Brownian Motion. The simulation findings reveal that random contracts can increase the sidechain's revenue by 24.70%, compared with deterministic contracts. The efficient service payments have been reduced by 44.65%, compared with the main chain's cost.

Novel empirical path-loss and root mean square delay spread (RDS) models for smart office scenarios are proposed. The effects of person density on the path loss and RDS are investigated based on the extensive measurements at 2.3–2.5 GHz. First, both of the measured path loss and RDS data are modeled as the dual log-distance functions. It is caused by the regular structure and furniture in the office environment. Second, in the proposed path-loss model, the path-loss exponents and the additional attenuation factor are modeled as quadratic functions of the person density. Meanwhile, the RDS is found to be uncorrelated with the person density. These phenomena reveal that the persons in the environments can be regarded as absorbers rather than scatters. Then, the accuracy of the proposed models is validated by the measured data and compared with two traditional models. Finally, the effect of the persons’ movements on the path loss and RDS is investigated, and the proposed models are extended to millimeter wave bands by a ray tracing technology. The proposed models and results can provide necessary information for link budget and algorithm design for the Internet of Things smart office scenarios.

A sharded blockchain with the Proof-of-Stake (PoS) consensus protocol has advantages in increasing throughput and reducing energy consumption, enabling the resource-limited participants to manage transactions and in a decentralized way and obtain rewards at a lower cost, e.g., Internet-of-Things (IoT) users. However, the latest PoS (e.g., Casper) requires a steep security deposit, which is the key to provide more robust security guarantees than Proof of Work, but not practical for the owners of heterogeneous IoT devices. This article considers any individual and institute who owns the IoT devices as the potential participant and focuses on designing the proper security deposits in a practical scenario with hidden information and hidden action. To bridge blockchain and the IoT users, we study the problem of balancing the security incentive and the economic incentive under two cases: 1) stake oriented and 2) effort oriented. We propose two joint models under the contract theory framework to efficiently address the problems: 1) joint adverse selection and moral hazard and 2) joint adverse selection and tournament. Both optimal contracts can provide a maximized profit for blockchain. The optimal rewards and security deposits for different types of participants can be determined accordingly. Simulations indicate that the proposed models can overcome asymmetric information and offer feasible contracts. Moreover, it demonstrates that both joint models can provide an economic incentive for the participants without reducing security incentives for the sharded blockchain.

In order to provide high quality services for Internet of Things (IoT) users with ever-changing requirements, a novel multi-access edge computing (MEC) based network paradigm is proposed in this letter. In the proposed framework, the service providers rent resources from resource providers (RPs), and then deploy them into MEC nodes (MNs). In this way, the MNs are labeled as temporal service points, which is flexible and cost-effective. Aiming to address the resource trading and price negotiation problem between the RPs and MNs, a matching with contracts model is leveraged in our proposed paradigm. Next, we further propose a distributed price negotiation algorithm to solve the problem. Finally, the simulations are presented to demonstrate the superiority of our proposed algorithm.

Proactive handover can avoid frequent handovers and reduce handover delay, which plays an important role in maintaining the quality of service (QoS) for mobile users in millimeter-wave vehicular networks. To reduce the communication cost of training the learning model for proactive handover, we propose a federated learning (FL) framework. The proposed FL framework can accommodate the limited storage capacity of each user, increase the number of users who participate in the FL, and adapt to the dynamic mobility pattern. Simulation results validate the effectiveness of the proposed FL framework. Compared to reactive handover schemes, the proposed handover scheme can reduce the unnecessary handovers and improve the QoS of users simultaneously.