Interactive services of mobile edge computing demand low latency in task handling, which cannot be easily satisfied due to limited per-user computing power on an edge server. Fortunately, a user can establish links with multiple base stations and the co-located edge servers in 5 G and beyond. By leveraging multi-connectivity, tasks of a computing service can be dispatched to multiple edge servers. This approach can also improve the quality of services, since tasks allocated to edge servers can be prioritized according to link quality. To exploit the benefits of multi-connectivity, a task scheduling problem is formulated to minimize latency and maximize reliability of the application, subject to task dependency and resource constraints. This problem is hard to solve due to NP-hardness and time-varying conditions. To this end, a pointer network based scheme called PtrTasking is developed to obtain a deep-learning model for the schedules and then infer new schedules on-line. To support time-varying conditions (e.g., a new application or evident change of computing load), a training strategy called PtrTrain is designed to retrain PtrTasking in a fast and efficient way. Experiments based on real-world datasets demonstrate that PtrTasking significantly improves latency and reliability, as compared to the baseline schemes.

Edge computing is promising for connected vehicles. As vehicles move, their resource demands for edge servers vary. Thus, it is necessary to reserve edge servers dynamically to meet variable demands. Existing schemes of edge-server reservation usually rely on statistical information of resource demands to make reservations; they are infeasible for connected vehicles, since such schemes are not adaptive to time-varying demands. To this end, a spatio-temporal reinforcement learning scheme called DeepReserve is developed to learn variable demands and then conduct edge-server reservation. Its design is based on the deep deterministic policy gradient algorithm of deep reinforcement learning (DRL), but is featured with several enhancements. First, the fully-connected neural network in DRL is replaced by a convolutional LSTM (ConvLSTM) network to extract spatio-temporal features of resource demands, which highly improves the prediction accuracy of resource demands. Thus, the actions in DRL (i.e., reservation decisions) can adapt to future demands. Second, an action amender is designed to ensure the actions selected by the neural network follow the spatio-temporal correlation. Finally, a training method called DR-Train is designed to stabilize the training procedure for different traffic patterns. DeepReserve is evaluated through extensive experiments on real-world datasets. Results show that it outperforms state-of-the-art approaches.

Federated learning (FL) provides a promising way to train a machine learning model among mobile devices without collecting their raw data to a central node. During training, proper devices are selected to participate in the training process to avoid model unfairness. In a mobile edge network, participant selection must be considered together with three factors: non-iid datasets possessed by devices, tunable local iterations on devices, and radio resource allocation to counter the impact of time-varying channel conditions on parameter transmissions. Since datasets of devices are given, to ensure model fairness and achieve fast convergence in the FL training process, participants, local iterations, and radio resources must be scheduled jointly in each iteration of FL training. In this paper, the joint scheduling problem is analyzed and formulated. Since it is NP-hard, a heuristic scheduling method called PALORA is designed to conduct joint scheduling of participants, local iterations, and radio resources. PALORA consists of three sequentially interactive function blocks: 1) a pointer network embedded deep reinforcement learning method to select participants, 2) an estimation algorithm to determine the numbers of local iterations, and 3) a breadth-first search method to allocate radio resources to the selected participants. PALORA is evaluated via extensive simulations based on real-world datasets. Results show that it significantly outperforms benchmark approaches.

Considering network slicing in a cellular network, one of the most intriguing tasks is slice enforcement over air interfaces across multiple cells. The challenges lie in several aspects. First, resources allocated to different slices must achieve soft isolation at the link level. Second, users’ diverse QoS requirements must be satisfied even when communication links experience fading and interference. Third, long-term slicing policies must be conformed, no matter how unbalanced they are. To address these challenges, link-level slice enforcement is first formulated as a resource allocation problem that minimizes radio resource consumption while ensuring link-level soft slice isolation, guaranteeing users’ diverse QoS requirements, and conforming to slicing policies. Next, this problem is tackled via a deep reinforcement learning (DRL) based approach, through which LinkSlice is designed as an iterative two-stage algorithm. The first stage determines transmission rates for each link based on DRL. It is embedded with a graph neural network (GNN) to characterize link interference. Based on the transmission rates from the first stage, the second stage allocates resources to each slice. Performance results show that LinkSlice converges quickly to a near-optimal solution. It gracefully tackles the three challenges of link-level slice enforcement while further improving throughput by 18.5%.

Edge computing is promising to provide computational resources for connected vehicles. Resource demands for edge servers vary due to vehicle mobility. It is then challenging to reserve edge servers to meet variable demands. Existing schemes rely on statistical information of resource demands to determine edge server reservation. They are infeasible in practice, since the reservation based on statistics cannot adapt to time-varying demands. In this paper, a spatio-temporal reinforcement learning scheme called DeepReserve is developed to learn variable demands and then reserve edge servers accordingly. DeepReserve is adapted from the deep deterministic policy gradient algorithm with two major enhancements. First, by observing that the spatio-temporal correlation in vehicle traffic leads to the same property in resource demands of CVs, a convolutional LSTM network is employed to encode resource demands observed by edge servers for inference of future demands. Second, an action amender is designed to make sure an action does not violate spatio-temporal correlation. We also design a new training method, i.e., DR-Train, to stabilize the training procedure. DeepReserve is evaluated via experiments based on real-world datasets. Results show it achieves better performance than state-of-the-art approaches that require accurate demand information.

Abstract In this work, an artificial neural network (ANN) based time‐varying channel modeling framework is proposed, including a playback model and a prediction model. The purpose of the ANN‐based modeling framework is to playback 5G measured radio channels at certain measurement positions, and further predict large scale channel parameters (LSCPs) at unmeasured positions with limited amount of measurement data. 28 GHz channel measurements were also conducted at a high‐voltage substation for the first time worldwide to meet with 5G radio system deployment for China Energy Internet. Meanwhile, the performance of the playback channels is evaluated by comparison with the measurements and traditional geometry based stochastic modeling (GBSM) simulated channels. An optimized radial basis function (ORBF) ANN is applied in the prediction model, and the predicted LSCPs are compared with the other approaches, which shows that the ORBF has the best performance. This work offers a solution to predict radio channels and parameters in case of big measured or simulated channel datasets.

Abstract In this paper, grey genetic optimization model (GGOM) is proposed for predicting insufficient channel parameters without increasing the amount of measurement data. Based on the millimetre wave 28 GHz indoor measurement data for both LOS and NLOS scenarios, the GGOM model is compared with traditional back propagation (BP) and grey model (GM) to analyse channel parameters like delay spread, excess delay and azimuth spread. Results show that the fitness of GGOM is better than the grey model in improving the stability of system. It works well with insufficient data (size less than 30) in most cases as it is set regardless of the specific scene and measurement data. This is verified by QuaDRiGa platform by generating uniformly distributed and interpolated data between the experimental measurement data. GGOM fits best with the measurement data compared with other prediction methods in channel characterization. Moreover, the mean absolute percentage error (MAPE) for GGOM is the least compared with GM and BP methods. The proposed GGOM model has good performance in modeling insufficient data of propagation channel, practically.