Billions of mobile terminals will be deployed in various Internet of Things (IoT), in which millimeter-wave (mmWave) technology will be gradually applied. Accurate modeling and simulation of wireless channel is the base for efficient design and performance evaluation. This becomes more important for industrial scenarios, which might be highly dynamic and potentially different from well-investigated cellular deployment scenarios. In this work, a novel semi-deterministic mmWave dynamic channel modeling approach based on optimal neural network (ONN) principle is proposed. The ONNs are radial basis function (RBF) neural networks (NNs) trained with optimal variance parameters and are applied to predict large-scale channel parameters (LSCPs) [e.g., path loss (PL), delay spread (DS), and angle spread (AS)]. Based on the ONNs’ predicted large-scale parameters and simplified propagation environment including the layout of transmitter (TX), receiver (RX), and major scatterers, the proposed channel modeling approach can generate accurate dynamic channel parameters. The proposed approach is validated by the channel data measured at a high-voltage substation. Large-scale parameters, multipath component (MPC) distributions, and power delay profiles (PDPs) are validated. The proposed approach is demonstrated to be an accurate, fast, and robust channel modeling method, which can be used for both link-level and system-level channel simulation for future design and optimization of industrial IoT.

Accurate prediction of millimeter-wave (mmWave) channel features can guarantee the quality of service in mmWave mobile communications. In this work, combining the advantages of deep neural network (DNN) and long short-term memory (LSTM), a novel approach is proposed to predict mmWave channel features. A joint prediction problem is raised that takes both the historical states of channel features and position information into account, while a novel DNN-LSTM structure is designed for realizing prediction. The proposed approach is validated by the mobile channel measurements conducted in a railway station and compared with existing approaches. The results show that the proposed DNN-LSTM based approach enables to predict both the change trend of channel features on large scale and the fluctuation on small scale, the accuracy can be improved by more than 4.5 % compared with existing approaches.

In this work, a novel deep neural network (DNN) based approach is proposed to simultaneously identify mmWave channel states and scenarios. In addition to use channel statistical features in traditional identification approaches, the cluster-based channel features carrying scatterers information are also considered when training our proposed DNN-based identification model, which is validated by the channel measurements conducted in a substation, a waiting hall and an open office scenarios with the line-of-sight and non-line-of-sight channel states. The results show that our proposed DNN-based approach can enhance the identification accuracies greatly, which is very useful to be applied in millimeter wave channel state and scenario identifications.

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.

This paper investigated a non-orthogonal multiple access (NOMA)-based integrated satellite–terrestrial network (ISTN), where each user can select to access a terrestrial base station (BS) or the satellite according to the capacity of BS and their individual transmission requirements. A two-stage algorithm is proposed to solve the achievable sum rate maximizing resource optimization problem. In the first stage, user associations are determined based on individual preference lists and the backhaul capacities of the access points (APs). In the second stage, the power allocation, and the receiving beamforming vectors are optimized alternately. Within each iteration, the closed-form solution for the transmit power is derived. Simulation results show the effectiveness of the proposed algorithm and the benefits brought by NOMA. When the backhaul link capacity of terrestrial BSs is sufficient, users (UEs) prefer to access these BSs. Otherwise, the satellite can offer QoS guarantees to UEs. Furthermore, the overall system performance reaches its optimum when the number of UEs in the system matches the number of receive antennas at the APs.

In order to overcome the disadvantages of traditional channel simulation approaches and achieve more accurate millimetre‐wave (mmWave) channel simulation under the condition of limited measured data, a novel Long Short‐term Memory Networks (LSTM) based hybrid mmWave channel simulation approach is proposed in this work. The proposed hybrid approach takes advantages of LSTM‐based time‐varying model of path loss and large‐scale channel parameters, statistical model of intra‐cluster multipath parameters and ray tracing, which are able to accurately simulate path loss, large‐scale channel parameters, multipath parameters and channel impulse responses, respectively. Based on the mmWave channel data measured in the waiting hall of Qingdaobei Railway Station, it is verified that the simulation results of our proposed hybrid approach are in good agreement with the measured data and better than that of existing channel simulation approaches. The hybrid channel simulation approach proposed in this work has important application value for channel modelling and simulation in the case of small amount of data.