The ongoing deployment of 5G network involves the Internet of Things (IoT) as a new technology for the development of mobile communication, where the Internet of Everything (IoE) as the expansion of IoT has catalyzed the explosion of data and can trigger new eras. However, the fundamental and key component of the IoE depends on the computational intelligence (CI), which may be utilized in the sixth generation mobile communication system (6G). The motivation of this article presents the 6G enabled network in box (NIB) architecture as a powerful integrated solution that can support comprehensive network management and operations. The 6G enabled NIB can be used as an alternative method to meet the needs of next-generation mobile networks by dynamically reconfiguring the deployment of network functions, providing a high degree of flexibility for connection services in various situations. Especially the CI technology such as evolutionary computing, neural computing and fuzzy systems utilized as a part of NIB have inherent capabilities to handle various uncertainties, which have unique advantages in processing the variability and diversity of large amounts of data. Finally, CI technology for NIB, which is widely used is also introduced such as distributed computing, fog computing, and mobile edge computing in order to achieve different levels of sustainable computing infrastructure. This article discusses the key technologies, advantages, industrial scenario applications of CI technology as NIB, typical use cases and development trends based on IoE, which provides directional guidance for the development of CI technology as NIB for 6G.

This paper studies the energy efficiency (EE) maximization problem of fixed-wing Unmanned aerial vehicles (UAVs) enabled Internet of Things (IoT) system, where an UAV is employed as an aerial relay with enough cache capacity to amplify and forward (AF) received signals between narrow band (NB)-IoT device and eNodeB. This paper optimizes the unconstrained trajectory of the UAV to maximize the EE of UAV-enabled IoT system to achieve the green communication. Due to the objective function is a fraction and is non-convex, it's hard to solve the optimization problem. Therefore, this paper alternately optimizes trajectory in source subspace and destination subspace with the other fixed and comes up with an algorithm based on successive convex approximation (SCA) method and Dinkelbach method to obtain a local optimal solution. Numerical results show that the proposed UAV trajectory design method can obtain much bigger EE than the running track (RT) trajectory and circular trajectory. Besides, the proposed cache-enabled AF strategy can obtain much bigger EE than non cache-enabled AF strategy.

This article studies the confidential transmission performance of an electric vehicle (EV) in heterogeneous network when it communicates with vehicle to grid (V2G). Based on the relay selection strategy that maximizes the signal-to-noise ratio(SNR), the electric vehicle as a legitimate user in this article uses a multi-antenna maximum ratio combining method for signal reception. Among them, a single antenna is configured for the power grid sender, relay nodes and illegal eavesdropping users. The wireless channel adopts Nakagami-m fading channel and the relay adopts decode and forward (DF) method. First, based on the stochastic geometric analysis method, statistical characteristics such as probability density function(PDF) and cumulative distribution function(CDF) of the received SNR are obtained at legitimate users and illegal eavesdropping users, respectively. Then, a functional analysis method is used to derive closed expressions for the secrecy outage probability (SOP) and non-zero security capacity probability in multi-eavesdropping user systems. Finally, the effects of the system's related parameters on SOP and non-zero security capacity probability are verified through simulations. The simulation results prove the correctness of the theoretical analysis, which can guarantee the privacy and security of electric vehicle users in heterogeneous network.

Reconfigurable Intelligent Surface (RIS) can intelligently control the wireless propagation environment by adjusting the signal phase and amplitude in real time, which is considered as one of the key technologies of 6G. Although RIS-assisted wireless communication systems greatly improve transmission efficiency under ideal conditions, there are various uncertainties in actual communication systems. In this paper, RIS-assisted communication system is investigated where the RIS cannot completely eliminate the phase error and the user location is randomly distributed with uncertainty. In addition, there are some interference sources around the user, which are simplified as co-channel interference (CCI). Then we study the performance of RIS-assisted communication system, and derive the closed-form expressions for the outage probability (OP) and channel capacity of the system. Furthermore, we analyze the effects of various parameters on the OP and channel capacity. Finally Monte Carlo simulation is carried out which verify the accuracy of the derivation.

Human pose estimation (HPE) is crucial for computer vision (CV). Moreover, it’s a vital step for computers to understand human actions and behaviours. However, the huge number of parameters and calculations in the HPE model have brought big challenges to deploy to resource-constrained mobile devices. Aiming to overcome the challenge, we propose a sparse pruning method (SPM) for the HPE model. First, L1 regularisation is added in the training phase of the original model, and network parameters of the convolution layers (CLs) and batch normalisation layers (BNLs) are sparsely trained to obtain a network structure with sparse weights. We then combine the sparse weights of filters with the scaling parameters of the BNLs to determine their importance. Finally, the structured pruning method is used to prune the sparse filters and corresponding channels. SPM can reduce the number of model parameters and calculations without affecting precision. Promising results indicate that SPM outperforms other advanced pruning methods.