In this paper, we develop a novel active video content caching scheme (RCC) based on recommendation system and consistent hash for device-to-device(D2D) communication. The RCC scheme is constructed by a cache placement scheme, a consistent hash algorithm, an optimal video segmentation scheme and an optimal video library segmentation scheme. To begin with, a well-designed cache placement scheme based on a mobile model with helper notes and video segmentation is proposed to reduce the redundancy of the videos and save users' cache. In order to solve the interruption problem caused by segmentation, a consistent hash algorithm is introduced to improve the success probability of D2D communication. According to the recommendation system, all users' predictive score for all videos can be calculated, which results that users' most interesting video files rather than popular video files as previous work can be obtained and cached in advance to improve the hitting probability. Furthermore, an optimal video segmentation scheme and an optimal video library segmentation scheme are developed to minimize the transmission delay and maximize the hitting probability respectively. Simulation results show that compared with other traditional caching schemes, the proposed RCC scheme can reach about 70% reduction in outage probability, 40% reduction in system latency, and 10% improvement in hitting probability, all of which can achieve the best performance.

In this paper, we study a wireless powered cognitive internet of things (IoT) network, where cognitive radio (CR) and non-orthogonal multiple access (NOMA) technologies are exploited to improve spectral efficiency, and radio frequency based energy harvesting (RF-EH) technology is integrated to achieve the sustainable IoT network. To ensure the freshness of information delivery, we investigate the age of information (AoI) as a performance metric, and formulate a long-term average AoI minimization problem under energy sustainability constraint, in which the working mode and transmit power of the secondary devices (SDs) are jointly optimized. Then, we reformulate it as a decentralized Markov decision process (Dec-MDP) with continuous action space. Accordingly, a deep reinforcement learning (DRL) framework is exploited, and a multi-agent twin delayed deep deterministic policy gradient algorithm with dual action selection mechanism (MATD3-DAS) is proposed, which adopts the centralized training and decentralized execution (CTDE) framework and exploits both actor and critic networks to select actions for improving exploration ability. Simulation results show that the proposed algorithm can significantly reduce the long-term average AoI, where the decrements approach 9.58% and 52.34% compared with the MATD3 algorithm and TD3-DAS algorithm with centralized training and centralized execution (CTCE).

Feature points loss and images mismatch in the variation of light intensity, weak texture and large angle rotation for the feature points extraction of ORB-SLAM2 are severe. To deal with the problem, a feature points extraction algorithm based on adaptive information entropy, i.e., Adaptive Information Entropy Feature (AIEF) algorithm is proposed. According to the information entropy, the image blocks with less information are removed and those with more texture image information and larger gradient are selected. Then an adaptive algorithm is used to automatically calculate the optimal threshold of the image information entropy. The image blocks are homogenized to avoid that the extracted feature points are too dense and getting stuck is prevented, which makes the algorithm more robust. Finaly validation is performed using the Oxford standard data set and the performances of the AIEF algorithm are compared with those of the SIFT, SURF, and ORB-SLAM2 algorithms. Experimental results on the Oxford standard data set demonstrate that the AIEF algorithm outperforms the traditional counterparts in terms of processing time, number of feature points, correct matching number and correct matching rate.

The number of computation-intensive apps and content requests within the Internet of Vehicles (IoV) has dramatically expanded with the swift advancement of 5G communication technologies and next-generation Internet technology. To address the issue of excessive response latency caused by the surge in data traffic, edge caching (EC) technology has proven to be effective. However, accurately predicting the popularity of content in different regions is challenging due to the high degree of mobility of the vehicles. Additionally, as vehicles typically do not remain in the same region for a long period of time, making the content popularity within a region highly dynamic. To address this issue, we propose a deep reinforcement learning-based mobile-aware edge cooperative caching scheme (MAECCD). Our approach utilizes an AutoEncoder to capture user features from local vehicle user data and accurately forecast content popularity in both vehicle-local and edge regions. Furthermore, the MAECCD incorporates a cooperation model between edge nodes to optimize cache utilization and employs techniques in Deep Reinforcement Learning (DRL) to refresh cached content and reduce transmission delay. Experimental results indicate that our MAECCD scheme outperforms other baseline algorithms in terms of cache hit ratio and content transmission delay.

Device-to-device (D2D) communication, as one of the promising candidates for the fifth generation mobile network, can afford effective service of new mobile applications and business models. In this paper, we study the resource management strategies for D2D communication underlying the cellular networks. To cater for green communications, our design goal is to the maximize ergodic energy efficiency (EE) of all D2D links taking into account the fact that it may be tricky for the base station (BS) to receive all the real-time channel state information (CSI) while guaranteeing the stability and the power requirements for D2D links. We formulate the optimization problem which is difficult to resolve directly because of its non-convex nature. Then a novel maximum weighted ergodic energy efficiency (MWEEE) algorithm is proposed to solve the formulated optimization problem which consists of two sub-problems: the power control (PC) sub-problem which can be solved by employing convex optimization theory for both cellular user equipment (CUE) and D2D user equipment (DUE) and the channel allocation (CA) sub-problem which can be solved by obtaining the weighted allocation matrix. In particular, we shed light into the impact on EE metric of D2D communication by revealing the nonlinear power relationship between CUE and DUE and taking the QoS of CUEs into account. Furthermore, simulation results show that our proposed algorithm is superior to the existing algorithms.