With the development of MEC services, MEC brokers will emerge to facilitate the purchase and management of resources for individual MEC users. Both data communication and computing resources offered by MEC service providers can be purchased by pay-as-you-go (PAYG) or reserved plans. Besides data and computing plans for each type of resource, we also consider combo plans specifically designed for MEC services covering both resources. In this paper, we propose a smart online aggregated reservation (SOAR) framework for MEC brokers to minimize their cost of reserving resources for multiple users without the knowledge of future demands. In our framework, a task aggregation algorithm is designed to aggregate the users’ demands in each PAYG billing cycle to improve the plan utilization, and plan reservation algorithms are proposed to decide when to reserve which plans. The performance gap (competitive ratio) between SOAR and optimal solution which knows all future demands in advance, is analyzed and derived in closed-form. The performance gap is proved to be the minimum among all deterministic online algorithms. Trace-driven simulations verify the cost advantage of our SOAR framework, which can save nearly 40 percent of cost for users through the brokerage service.

In this paper, we examine new plan pricing schemes for multimedia applications that offload computing-intensive tasks to computing servers incurring both communication and computing costs. Pricing schemes offered to users include: 1) a pay-as-you-go payment for usage of communication or computing resources, 2) an upfront data (resp. computing) plan for unlimited usage of communication (resp. computing) resources during a period, and 3) an upfront combo plan for unlimited usage of both communication and computing resources during a period. We aim to solve an online plan reservation problem: the amount of resources needed by a task is only known when it arrives, i.e., the future is unknown. However, even if the resource usage of future tasks is known in advance, the plan reservation problem is NP-hard and thus challenging. To tackle this problem, we propose a randomized online reservation (ROR) scheme to reserve plans probabilistically, where the probability is determined by the recent usage of resources. The performance gap (competitive ratio) between our proposed scheme and the optimal solution is analyzed and derived in closed-form, and this gap is proved to be the minimum among all online algorithms which do not know the usage of future tasks. Trace-driven simulations verify the cost advantage of ROR and characterize how different prices of plans influence users' plan reservation strategies.

A promising solution to address the spectrum shortage in 5G cellular systems is the deployment of millimeter wave (mmWave) heterogeneous networks (HetNets). However, a key challenge for mmWave HetNets is to manage the user mobility and handovers among mmWave small cells exploiting highly directional antennas and conventional microwave macro cells. In this paper, we propose a new efficient handover decision algorithm based on a Markov Decision Process (MDP) to optimize the overall service experience of users in mmWave HetNets. By utilizing user's mobility information (velocity and location), the proposed algorithm avoids excessive handovers and effectively tackles beamforming misalignments, and signal blockages in mmWave small cells. To improve the computational efficiency of the MDP, we apply the action elimination method by exploiting unique handover properties of mmWave HetNets. While maintaining optimality, theoretical analysis shows that our proposed handover decision algorithm can reduce the computational complexity by 0.25M|Bs| + 0.25M/(M - 1) times, where M is the number of base stations and |Bs| is the number of beamwidth options for mmWave beamforming. Mobility-trace driven numerical results demonstrate the optimality of our proposed algorithm compared with other benchmark schemes and its low computational complexity over the traditional approach.