Search

Live video streaming has become one of the key applications in mobile wireless networks. To offload the bandwidth requirement in both backhaul and radio access networks, the integration of Mobile Edge Computing (MEC) and multicasting have become a natural candidate. However, less attention has been paid to the user Quality of Experience-driven (QoE-driven) approach to optimize the radio resource management of multicasting in mobile wireless networks. In this work, we study the optimal radio resource management, including modulation and coding scheme (MCS) selection, radio resource blocks allocation, and Forward Error Correction (FEC), for multicasting in LTE networks with the assistance of MEC. We formulate it as a convex optimization problem and propose a weighted sub-gradient (WSG) method to find the near-optimal solution. In addition, we also propose a heuristic algorithm based on the concept of Maximizing marginal Gain and Minimizing marginal Loss (MGML). Our simulation results show that both approaches are able to achieve near-optimal solutions and outperform previous work, including MSML [11] and OLM [14] . Our simulation results also show that WSG yields the best QoE fairness index while MGML yields the best system utility in most scenarios.

To support wireless Internet of things (IoT) devices, this paper presents a new solution which combines wireless power transfer and mobile edge computing. Specifically, we consider one mobile device, which first harvests energy from radio frequency signals sent by a base station and then offloads all or part of its data to be processed to the base station. The process of energy harvesting and offloading span over multiple fading blocks. The target is to maximize the average amount of processed data in unit time. To achieve this target, we optimize the stopping rule for energy harvesting (i.e., when to stop energy harvesting and start offloading) and the number of fading blocks for data offloading. To solve the formulated problem optimally, we decompose it into two levels. In the lower level, the stopping rule for energy harvesting is optimized given a fixed number of fading blocks for offloading. The associated lower-level problem is solved optimally based on a series of special properties of the problem. In the upper level, the number of fading blocks for offloading is optimized. Efficiency of our work with fully offloading mode and partially offloading mode is shown by using simulation.

The vehicular edge computing is considered an enabling technology for intelligent and connected vehicles since the optimization of communication and computing on edge has a significant impact on driving safety and efficiency. In this paper, with the road traffic assignment to “proactively” reshape the spatiotemporal distribution of resource demands, we investigate the joint service migration and mobility optimization problem for vehicular edge computing. The goal is to meet the service delay requirements of vehicular edge computing with minimum migration cost and travel time. As service migration and mobility optimization are coupled, the joint scheduling problem suffers from the curse of dimensionality, which cannot be solved in real time by centralized algorithms. To this end, a multi-agent deep reinforcement learning (MADRL) algorithm is proposed to maximize the composite utility of communication, computing, and route planning in a distributed way. In the MADRL algorithm, a two-branch convolution based deep Q -network is constructed to coordinate migration action and routing action. Extensive experimental results show that the proposed algorithm is scalable and substantially reduces service delay, migration cost and travel time as compared with the existing baselines.

In this correspondence, we jointly optimize the caching decision and radio and computation resource allocation for video service in mobile edge computing (MEC) enabled network to maximize the system revenue. To tolerate the uncertainty in the network traffic and avoid “hard constraint” based on constant content request rate, we employ robust optimization to obtain optimal caching decisions, and then accordingly allocate the radio and computation resources for video transcoding. Simulation results demonstrate the effectiveness of the proposed scheme.

Based on the functional characteristics of inter-vehicle communication networks under the current framework, a privacy-preserving authentication framework that combines fifth-generation communication technology (5G) and edge computing technology is proposed. The proposed framework is different from that used in the previous 802.11p-based inter-vehicle communication network architecture, and we use device-to-device technology to achieve communication between vehicles. One difference between a 5G-enabled model and a traditional model for vehicle ad hoc networks is that secure communication between vehicles in a 5G-enabled model is challenging. The authentication protocol proposed in this study can be divided into two parts. The first part involves authenticating and selecting an edge computing vehicle, wherein a fuzzy logic mathematical method is used in the selection process. The second part is the mutual authentication between edge computing and ordinary vehicles. In this process, the exchange of security information between vehicles in a group is possible. Simultaneously, the identity privacy and traceability of the vehicle are ensured. Moreover, the proposed signature scheme is proved to be secure under a random oracle model. Performance evaluation results demonstrate that our proposed scheme has relatively lower computational and communication overhead than existing schemes.

In this article, we investigate the cellular assisted mobile edge computing (MEC) via non-orthogonal multiple access (NOMA), where a group of edge-computing users (EUs) exploit NOMA to simultaneously offload their computation-workloads to an edge-server (ES), and conventional cellular-user (CU) allows the EUs to reuse its authorized frequency channel for the NOMA-transmission. We firstly characterize the transmit-powers of the CU and EUs and then formulate a joint optimization of the EUs’ offloaded computation-workloads, offloading-duration (i.e., how long for reusing the CU's channel), as well as the ES's computation-resource allocations (for processing different EUs’ offloaded workloads), with the objective of minimizing the overall-latency in completing all EUs’ computation-requirements, subject to the CU‘s and EUs’ limited power and energy capacities as well as the ES's limited computation-resource capacity. Despite the strict non-convexity of the formulated joint optimization problem, we propose an efficient algorithm to compute the optimal offloading solution. With the optimal offloading solution for the scenario of one CU, we further investigate the scenario of multiple CUs and investigate the optimal pairing of the EUs for reusing different CUs’ channels for computation-offloading. Taking into account the coupling effect due to the ES's limited computation-resource, we formulate a joint optimization of the EU-pairing and the ES's capacity allocation of the computation-resource for accommodating different EU-pairs. Despite the difficulty due to the mixed binary and non-linear non-convex programming of the formulated problem, we propose an efficient layered algorithm for solving the problem. Numerical results are provided to validate the accuracy of our proposed algorithms. We also show the performance advantage of our NOMA-assisted offloading in comparison with conventional orthogonal multiple access (OMA) based computation offloading.

In this letter, we study an unmanned aerial vehicle (UAV)-mounted mobile edge computing network, where the UAV executes computational tasks offloaded from mobile terminal users (TUs) and the motion of each TU follows a Gauss-Markov random model. To ensure the quality-of-service (QoS) of each TU, the UAV with limited energy dynamically plans its trajectory according to the locations of mobile TUs. Towards this end, we formulate the problem as a Markov decision process, wherein the UAV trajectory and UAV-TU association are modeled as the parameters to be optimized. To maximize the system reward and meet the QoS constraint, we develop a QoS-based action selection policy in the proposed algorithm based on double deep Q-network. Simulations show that the proposed algorithm converges more quickly and achieves a higher sum throughput than conventional algorithms., Accepted with minor revisions by IEEE TVT

Multi-access Edge Computing (MEC) is recently acknowledged as one of the key pillars for the next revolution of mobile communications area, where the convergence of IT and telecommunications network provides the low latency and computation capability for cellular base stations (BSs). As a result, this is a great opportunity for mobile network operators by deploying new services and applications at BSs. Nevertheless, huge capital and operational cost can challenge mobile operators for the deployment of new BSs and MEC micro-datacenters. Colocation Edge Computing (ColoMEC) is a new concept where multiple operators share not only the same BS tower but also their radio and computation resources colocated at the edge sites. In order to reduce the operational cost of a ColoMEC system, the limited bandwidth at over-utilized colocation BSs can be extended by sharing the bandwidth among BSs, while shared MEC micro-datacenters can be scaled based on the arrival traffic loads. Thus, sharing the BS infrastructure, bandwidth, and MEC micro-datacenters among the co-located mobile operators can be an economical solution to provide high-performance services with low expenses by exploiting the temporal and spatial difference in traffic loads. Turning this vision into reality, we study a joint bandwidth allocation sharing and MEC micro-datacenter scaling in ColoMEC management problem ( $\mathbf {ColoMEC-MP}$ ). To solve $\mathbf {ColoMEC-MP}$ problem, we propose an algorithm based on proximal block coordinate descent technique by iteratively solving the decoupled convex subproblems (i.e., user association, bandwidth allocation, and MEC micro-datacenter scaling) with additional proximal terms. To improve the convergence of the proposed algorithm, we propose a greedy initialization for the user association which is based on the link capacity at each user. Our simulation demonstrates the superiority of the algorithm in terms of the operational cost compared with fixed service rate of shared MEC micro-datacenters strategies.

We consider Device-to-Device (D2D)-enabled mobile edge computing offloading scenario, where a device can partially offload its computation task to the edge server or exploit the computation resources of proximal devices. Keeping in view the millisecond-scale latency requirement in 5G service scenarios and the spectrum scarcity, we focus on minimizing the sum of task execution latency of all the devices in a shared spectrum with interference. In particular, we provide an integrated framework for partial offloading and interference management using orthogonal frequency-division multiple access (OFDMA) scheme. Accordingly, we formulate total latency minimization as a mixed integer nonlinear programming (MINLP) problem by considering desired energy consumption, partial offloading, and resource allocation constraints. We use decomposition approach to solve our problem and propose a novel scheme named Joint Partial Offloading and Resource Allocation (JPORA). With aim to reduce the task execution latency, JPORA iteratively adjusts data segmentation and solves the underlying problem of quality of service (QoS)-aware communication resource allocation to the cellular links, and interference-aware communication resource allocation to D2D links. Extensive evaluation results demonstrate that JPORA achieves the lowest latency as compared to the other baseline schemes, meanwhile limiting the local energy consumption of user devices.

The dynamic environment due to traffic mobility and wireless communication from/to vehicles make identity authentication and trust management for privacy preservation based on vehicular edge computing (VEC) an increasingly important problem in vehicular networks. However, existing authentication schemes mainly focus on communication between a single trusted edge computing node and multiple vehicles. This framework may suffer the bottleneck problem due to the single edge computing node, and the performance depends heavily on its resources. In this paper, a blockchain empowered group-authentication scheme is proposed for vehicles with decentralized identification based on secret sharing and dynamic proxy mechanism. Sub-authentication results are aggregated for trust management based blockchain to implement collaborative authentication. The edge computing node with a higher-reputation stored in the tamper-proof blockchain can upload the final aggregated authentication result to the central server to achieve the decentralized authentication. This work analyzes typical attacks for this scheme and shows that the proposed scheme achieves cooperative privacy preservation for vehicles while also reducing communication overhead and computation cost.

Nowadays, the ever increasing demand of mobile computing applications has triggered the shift from the centralized mobile Cloud Computing toward mobile Edge Computing (MEC). The main feature of MEC is to move computing and storage to the network edges (e.g., radio access point/base stations), enabling resource-limited mobile devices to support computation-intensive and latency-critical applications. This paper deals with the performance evaluation of a MEC system providing computational capabilities to users within a limited area, according to the software as a service (SaaS) paradigm. In particular, a Markov multi-server queuing system model with requests reneging is proposed in order to accomplish the performance evaluation of the MEC system and derive the minimum number of processors to be allocated in order to fulfill specific service requirements in terms of resulting requests dropping probability. Finally, the pertinence of the proposed Markov queueing model is validated by comparing the obtained analytical predictions with numerical results derived by resorting to extensive computer simulations carried out under the assumption of realistic operating conditions.

Recently, the rapid advance of vehicular networks has led to the emergence of diverse delay-sensitive vehicular applications such as automatic driving, auto navigation. Note that existing resource-constrained vehicles cannot adequately meet these demands on low / ultra-low latency. By offloading parts of the vehicles’ compute-intensive tasks to the edge servers in proximity, mobile edge computing is envisioned as a promising paradigm, giving rise to the vehicular edge computing networks (VECNs). However, most existing works on task offloading in VECNs did not take the load balancing of the computation resources at the edge servers into account. To address these issues and given the high dynamics of vehicular networks, we introduce fiber-wireless (FiWi) technology to enhance VECNs, due to its advantages on centralized network management and supporting multiple communication techniques. Aiming to minimize the processing delay of the vehicles’ computation tasks, we propose a software-defined networking (SDN) based load-balancing task offloading scheme in FiWi enhanced VECNs, where SDN is introduced to provide supports for the centralized network and vehicle information management. Extensive analysis and numerical results corroborate that our proposed load-balancing scheme can achieve superior performance on processing delay reduction by utilizing the edge servers’ computation resources more efficiently.

Long-time operation of traction systems under severe working environment easily induces various faults in high-speed trains, and thereby puts passengers in an unsafe position. Thanks to widely equipped sensors in high-speed trains and the technical development of advanced data analysis techniques, real-time fault detection (FD) using only online data is an inevitable trend to improve safety and reliability of high-speed trains. In this paper, an edge computing-aided FD framework for traction systems in high-speed trains is proposed; the superior advantages include: 1) its implementation can be easily carried out without system information such as accurate mathematical models of traction systems; 2) it has the well-deserved expansibility without being redesigned for the control structure of traction systems; 3) it nearly does not occupy resource of traction control units, and it does not overburden the information transmission systems. This highly intelligent FD framework using edge computing is firstly speculated in details, and its core theory is then applied to a dSPACE platform of high-speed trains for verification.

The blockchain-based video systems are designed to build a decentralized and flexible video ecosystem by enabling a direct interaction among users, video providers (VPs), and service providers. In blockchain-based video systems, the heterogeneous qualities and formats of the video streams usually require massive computational resources to transcode them into different versions and formats to meet distinct requirements of users. However, current blockchains cannot handle massive and heterogeneous video streaming due to limited computing capacity and long transaction times. To deal with this issue, in this paper, leveraging mobile edge computing (MEC) technology, we propose a blockchain-based MEC architecture, where small base stations (SBSs) allocate their computation as well as communication resources for providing video streaming in a distributed and secure manner. Moreover, to improve the operation efficiency, we use a series of smart contracts to enable a self-organized video transcoding and delivery service without a centralized controller. Then, users, SBSs, and VP could adjust their strategies according to the transactional information on blockchain. Moreover, we formulate the video transcoding and delivery problem as a three-stage Stackelberg game. We analyze the sub-game equilibrium in each stage and the interplays of the three-stage game. Last, we propose an iterative algorithm to solve the problem. Simulation results show that the proposed approach could obtain the good performance in terms of average time to finality (TTF), average access delay, and network cost.

Cooperative vehicles are better able to detect the environment and self-localize than a single vehicle. Cooperative vehicles can quickly cover the entire environment by communicating and cooperating with each other and can also reduce localization and mapping error by merging the cooperative vehicle information from observation and navigation. In this paper, we propose a novel algorithm for an effective solution of navigation and mapping for cooperative vehicles in an unknown environment. We present an improved centralized and collaborative monocular simultaneous localization and mapping (CCM-SLAM) approach. The proposed algorithm can accurately compute the transformation matrix for cooperative vehicle maps and reduce the communication delay, data loss among vehicles and decrease the bandwidth demand. The quaternion and credibility similarity transformation (QC-Sim(3)) method we proposed is used to accurately merge the matched maps and accomplish loop closures. The sending messages at variable frequencies (SMVF) method we proposed and an improved detection and resending lost data (I-DRLD) method we proposed can improve the accuracy of pose estimation. SMVF solves the time-delay problem by sending messages to the vehicles at flexible frequencies while I-DRLD detects and resends the lost data. We also adopt Intra-frame Feature Compression (IFC) to decrease the bandwidth demand in the process of the transmitting data. The experiments demonstrate the superiority of our proposed algorithm compared with the state-of-the-art methods.

In order to help user terminal devices (UTDs) efficiently handle computation-intensive and time-delay sensitive computing task, multi-access edge computing (MEC) has been proposed. However, due to the differences among the performance of UTDs, and the resource limitation of MEC servers, the joint optimization between the offloading decisions of UTDs and the allocation of resources in network is still a focus of the research. This paper studies the joint computation offloading and resource allocation strategy in multi-user and multi-server scenarios. Firstly, we formulate the joint optimization problem of computation offloading and resource allocation as a mixed integer nonlinear programming (MINP) problem to minimize the energy consumption of UTDs, by constraining the offloading decision, channel selection, power allocation and resource allocation. Secondly, we propose a two-stage heuristic optimization algorithm based on genetic algorithms, which divides the joint optimization problem of computation offloading and resource allocation in two stages. Based on the coupling relationship between the offloading decision and the resource allocation scheme, we iteratively update the solution of the problem, and finally obtain the stable convergence solution of the optimization problem. Finally, the proposed algorithm is compared with other classical methods to prove the effectiveness.

The application of network slicing to mobile edge computing (MEC) systems has aroused great interests from both academia and industry. However, the optimization of network slicing and MEC in most existing research works only focuses on resource slicing, energy scheduling, and power allocation from the perspective of mobile devices, without considering the operator's revenue. In this paper, we propose a novel framework for network slicing in MEC systems, including slice request admission and a revenue model, to investigate the operator's revenue escalation problem while considering traffic variations. The revenue model is mainly composed of the longer-term revenue and short-term revenue. Particularly, we jointly optimize slice request admission in the long-term and resource allocation in the short-term to maximize the operator's average revenue. By employing the Lyapunov optimization technique, we develop an algorithm without requiring any prior-knowledge of traffic distributions, referred to as the DNSRA, to solve the problem. To reduce the computational complexity of directly solving the DNSRA, we decouple the optimization variables for efficient algorithm design. By this, the strategies for user association and CPU-cycle frequency are obtained in closed forms, respectively. Power allocation and subcarriers assignment are obtained by employing the successive convex approximation approach. Meanwhile, we develop a heuristic algorithm to obtain the dynamic slice request admission decision. Simulation results show that the proposed DNSRA can strike a flexible balance between the average revenue and the average delay, and can significantly increase the operator's revenue against existing schemes.

Mobile edge computing (MEC) has been envisioned as a promising technology for enhancing the computational capacities of mobile devices, by enabling computational task offloading. In this article, we employ massive multiple-input multiple-output methods to facilitate offloading in MEC. Our objective is to minimize the maximum delay for offloading and computing among the users, which requires a joint allocation of wireless and computational resources. Both perfect and imperfect channel state information (CSI) are considered. Under perfect CSI, we derive a semi-closed-form solution for the formulated problem. Under imperfect CSI, since the formulated problem is non-convex, we transform it into a convex one using a successive convex approximation technique and propose an iterative algorithm to solve it. Presented numerical results show the benefits of having a large number of antennas at the base station, and the necessity of performing joint radio and computational resource allocation.

With the development of autonomous vehicular technologies, the execution tasks become more memory-consuming and computation-intensive. Simultaneously, a certain portion of tasks are latency-sensitive, such as collaborative perception, path planning, collaborative simultaneous localization and mapping, real-time pedestrian detection, etc. Because of the limited computation resources inside vehicles and restricted transmission bandwidth, edge computing can be an effective way to assist with the tasks execution. Considering from the perspective of business, the reservation or subscription cost is cheaper than real time requests. In order to minimize the expense of consuming edge services, the desirable situation is to reserve the resources as much as needed. However, the configuration of vehicular network is variational in practice due to the diversity of road maps, different time range like peak time and off-peak time, and the various task types, which makes it challenging to figure out a general machine learning model that is suitable for any case. Therefore, to predict the resource consumption in edge nodes accurately in different scenarios, we propose a two-stage meta-learning based approach to adaptively choose the appropriate machine learning algorithms based on the meta-features extracted on database. Besides, due to the deficiency of dataset for edge resource consumption, we program in game engine unity to generate the 3D model of Manhattan area. Meanwhile, we change the factors like different road maps and number of vehicles so as to get closer to practices. In the evaluation part, we adopt root mean square error, mean absolute percentage, and mean GEH as evaluation metrics to assess the performance of each model. Also, a quantitative analysis for the total cost and waste is also conducted. Eventually, we can find that the proposed meta-learning based method outperforms the non-meta ones.

In this paper, we study unmanned aerial vehicle (UAV) assisted mobile edge computing (MEC) with the objective to optimize computation offloading with minimum UAV energy consumption. In the considered scenario, a UAV plays the role of an aerial cloudlet to collect and process the computation tasks offloaded by ground users. Given the service requirements of users, we aim to maximize UAV energy efficiency by jointly optimizing the UAV trajectory, the user transmit power, and computation load allocation. The resulting optimization problem corresponds to nonconvex fractional programming, and the Dinkelbach algorithm and the successive convex approximation (SCA) technique are adopted to solve it. Furthermore, we decompose the problem into multiple subproblems for distributed and parallel problem solving. To cope with the case when the knowledge of user mobility is limited, we adopt a spatial distribution estimation technique to predict the location of ground users so that the proposed approach can still be applied. Simulation results demonstrate the effectiveness of the proposed approach for maximizing the energy efficiency of UAV., 33 pages, 7 figures, IEEE single-column journal

Low latency/delay is one of the most critical requirements for the application of vehicular networks. However, frequent real-time information update caused by vehicles high mobility is liable to aggravate the delay. Meanwhile, the task migration between different vehicular edge computing (VEC) servers results in an amount of delay if the computing cannot be completed before the vehicle moves out of the coverage of the current VEC server. In this paper, the problem is concluded as when and to whom to offload the task for VEC, which is formulated as a finite horizon Markov decision process (MDP) to minimize the delay with respect to the communication, computing, handover and migration. Through characterizing the time-space correlation of vehicles mobility, the curse of dimensionality problem in MDP is resolved. Meanwhile, a general expression of the transition probabilities is derived. On this basis, the specific results of highway, 2-D street and real-data scenarios are provided as well. For practical implementation considerations, the transition probabilities are commonly uncertain primarily due to random driver behavior, inaccurate sample data and complex path environment. Under this uncertain environment,a robust time-aware MDP-based task offloading algorithm (RTMDP) is proposed, which has been proved to perform well even under the high uncertain transition probabilities by simulation results.

Mobile Edge Computing (MEC) is a new computing paradigm with great potential to enhance the performance of user equipment (UE) by offloading resource-hungry computation tasks to lightweight and ubiquitously deployed MEC servers. In this paper, we investigate the problem of offloading decision and resource allocation among multiple users served by one base station to achieve the optimal system-wide user utility, which is defined as a trade-off between task latency and energy consumption. Mobility in the process of task offloading is considered in the optimization. We prove that the problem is NP-hard and propose a heuristic mobility-aware offloading algorithm (HMAOA) to obtain the approximate optimal offloading scheme. The original global optimization problem is converted into multiple local optimization problems. Each local optimization problem is then decomposed into two subproblems: a convex computation allocation subproblem and a non-linear integer programming (NLIP) offloading decision subproblem. The convex subproblem is solved with a numerical method to obtain the optimal computation allocation among multiple offloading users, and a partial order based heuristic approach is designed for the NLIP subproblem to determine the approximate optimal offloading decision. The proposed HMAOA is with polynomial complexity. Extensive simulation experiments and comprehensive comparison with six baseline algorithms demonstrate its excellent performance.

Vehicular edge computing (VEC) is considered as a novel paradigm to enhance the safety of automated vehicles and intelligent transportation systems (ITS). The computation offloading strategies are the key point of VEC, and the effect of time-varying channels cannot be ignored during the task transmission period. This paper investigates the utility maximization problem with task delay requirement constraints, in which the influence of time-varying channel on the task offloading strategies during the task offloading period is considered. The time-varying fading channel leads to the time-varying spectrum efficiency (SE), so the previous offloading strategies are questionable when the additional uncertain allocated bandwidth is taken into account. To deal with it, we first propose a linearization based Branch and Bound (LBB) algorithm to solve the fixed SE problem without considering the time-varying channel characteristics. Considering the complexity of the LBB algorithm, a closest rounding integer (CRI) algorithm is proposed to solve the fixed SE problem. Then, based on the resource allocation strategies of the fixed SE problem, we propose the LBB based computation offloading (LBBCO) algorithm and the CRI based computation offloading (CRICO) algorithm to solve the original problem for both the static tasks and dynamic tasks. The proposed LBBCO/CRICO algorithms are also applicable to multi-vehicle and multi-task scenarios. Furthermore, we analyze the effect of small-scale fading on the proposed offloading strategies. The simulation results show that the average utilities of LBBCO and CRICO algorithms have a small gap by 3.93% and 6.13% only to the upper bound, respectively. Meanwhile, the proposed LBBCO and CRICO algorithms can outperform the previous state-of-the-art solution by 4.52% and 2.38%, respectively.

Unmanned aerial vehicles (UAVs) can be conveniently deployed for environmental monitoring, firefighting, disaster rescue, and so on. However, the utmost challenge is how to transfer the important and urgent information to the control center as quick as possible in face of communication and computation constraints. As one of the promising technologies, mobile edge computing (MEC) technology can be deployed on UAVs to support computation-intensive and latency-critical applications. Therefore, a joint communication and computation optimization model is established for a MEC enabled UAV network, which includes a centralized MEC enabled top-UAV and a swarm of distributed bottom-UAVs. Using stochastic geometry, the successful transmission probability results for a single link and a group of links are derived based on the three-dimensional distribution of UAV swarm. Moreover, the optimal response delay is theoretically achieved with the closed-form solutions by using stochastic geometry and queueing theory. In contrast to the conventional UAVs without MEC capabilities, the optimal response delay is achieved by using our proposed joint communication and computation optimization algorithm in the MEC enabled UAV swarm scenario. The performances of the proposed algorithm are evaluated based on the results from the simulation system and the hardware testbed.

The emergence of mobile edge computing (MEC) and unmanned aerial vehicles (UAVs) is of great significance for the prospective development of Internet of Things (IoT). The additional computation capability and extensive network coverage provide energy-limited smart mobile devices (SMDs) with more opportunities to experience diverse intelligent applications. In this paper, a computation efficiency maximization problem is formulated in a multi-UAV assisted MEC system, where both computation bits and energy consumption are considered. Based on the partial computation offloading mode, user association, allocation of central processing unit (CPU) cycle frequency, power and spectrum resources, as well as trajectory scheduling of UAVs are jointly optimized. Due to the non-convexity of the problem and the coupling among variables, we propose an iterative optimization algorithm with double-loop structure to find the optimal solution. Simulation results demonstrate that the proposed algorithm can obtain higher computation efficiency than baseline schemes while guaranteeing the quality of computation service.

Multi-access edge computing (MEC) and nonorthogonal multiple access (NOMA) are two enabling technologies in the 5 G network and beyond. MEC admits user equipments (UEs) running many more compute-intensive applications by providing computing capabilities at the network edge and within radio access networks, while NOMA enables multiple UEs to share the same resource block, thus leveraging considerable advantages such as greater spectral efficiency and a larger number of supported UEs. The state-of-the-art showed that the combination of NOMA and MEC can lower the energy consumption and/or overall latency; however, they mostly focused on single-carrier NOMA. In this paper, we investigate the computation offloading problem in multi-carrier NOMA enabled MEC systems and solve it from the cooperative game theory viewpoint using coalition formation game. Particularly, UEs are considered as game players and subcarriers are regarded as coalitions that can be used for computation offloading of multiple UEs. Based on the introduced coalition formation game, we develop a low-complexity algorithm with convergence guarantee to achieve the Nash-stable solution. Numerical results are provided to validate the effectiveness of the proposed coalition game based algorithm as well as its comparison with three baseline schemes.