D2D Architecture Research Articles

Towards a 3-tiered space-air-ground network with reinforcement learning

A Space-Air-Ground Integrated Network (SAGIN) is a novel networking concept that integrates satellite networks, aerial networks and terrestrial networks into a 3-tiered network. It has been developed as an adaptable computing and traffic model in the present decade. In addition to various benefits, there are some unprecedented challenges in SAGIN, and reliability is one among them. The wireless communication network requires stable communication between Unmanned Aerial Vehicles (UAVs) and Unmanned Ground Vehicles (UGVs). For ensuring reliable communication links, Massive Multiple-Input Multiple-Output (MaMIMO) is used with the deployed aerial vehicles, and the mobility of UGVs can be controlled by UAV, providing device-to-device (D2D) communication between vehicular nodes, so that no interruption can occur. This work involves developing a 3-Tier D2D Architecture, comprised of network links UGVs, network links UAVs, and the combined model of both and also links of the satellite groups at lower orbits. The UAV senses the environment and transmits the data to its operator for appropriate decision-making. This work focuses on incorporating D2D communications in SAGIN for ensuring reliability via Reinforcement Learning (RL) based on Markov process. We derive the optimal numbers of transmitting nodes used for communication links, represent the Markov State transition and provide the Bellman equation for our model. The main objective is to maximize Energy Efficiency (EE), under the limitations of Spectral Efficiency (SE). Performance evaluations are employed to assess reliability over the links in the corresponding proposed architecture. The simulation results with water-filling optimization show that the proposed model achieves enhanced EE and SE per D2D link.

A Cooperative Approach to Extend Cellular Coverage via D2D Architecture based on OLSR Protocol

The access part of all cellular network’s generation suffers from common concerns related to dead spots (zones that are not covered by the network) and hot spots (zones where the number of users is higher compared to network resources). During the last decade, lots of research proposals have tried to overcome cellular problems through multi-hop D2D architecture, which is a new paradigm allowing the direct communication between devices in cellular network to enhance network performances and improve user QoS. In this paper, we propose a multi-hop D2D architecture based on the OLSR protocol to extend cellular coverage. Cell-OLSR, which is the proposed adaptation of OLSR for our architecture, allows the exchange of cellular parameters between nodes to choose the best proxy device to forward data to the cellular base station (BS).

Device-to-Device (D2D) Communication as a Bootstrapping System in a Wireless Cellular Network

Device-to-Device (D2D) communication is considered to be a promising form of technology to distribute video contents to nearby users and devices at a high data rate and with few delays. In this paper, we propose utilizing the D2D services and cache of mobile devices as a bootstrapping system in a cellular network. The main objective of a bootstrapping-D2D (BD2D) system is to improve the quality of service in terms of achieving the minimum startup-time of popular video files in a cellular network. To achieve our objective, we propose a new caching concept for D2D communication in which each mobile device can cache and share the initial video portion or initial segments of the popular video files to other proximate users via a D2D communication link. We formulate an optimization problem that maximizes the cache-hit ratio of initial segments of the most popular video contents. The basic aim is to move the initial segments as close as possible to the wireless users. We also propose a network-assisted D2D architecture called mobile content delivery network-assisted-BD2D and algorithms that assist the BD2D system in the discovery of devices that have cached the initial segments of the requested video files, in cache placement decisions, and also in the registration of potential D2D devices at the core network. We evaluate the performance of the BD2D system and propose algorithms through extensive simulations under real wireless network conditions, such as interference, fading, and shadowing. Our concluding simulation results demonstrate that bootstrapping the initial segments using the short range BD2D system can significantly reduce the average startup-time of video files in comparison with bootstrapping the initial segments from the traditional cellular system.

A Novel Distributed Q-Learning Based Resource Reservation Framework for Facilitating D2D Content Access Requests in LTE-A Networks

With the rapid increase in mobile data traffic (especially video content traffic), the number of content access requests is also increasing. In this context, device-to-device (D2D) communication has been an effective technology in increasing the spectral efficiency and reducing the load by offloading the mobile data traffic in cellular networks. In order to reduce the load on eNodeBs (eNBs), centralized area controller (CAC) has been proposed earlier to take content aware decisions for content access requests. In this paper, we propose to exploit CAC in a distributed manner. Distributed D2D controller (DDC) is responsible for arbitrating and scheduling D2D transfers. Thus, every content access request from a user equipment (UE) is first served by a neighboring DDC and in case the content is not available in the region, the request is forwarded to the eNB. Further, it has been observed that the inclusion of DDCs in the existing D2D architecture reference model requires the addition of new reference points. The same has been highlighted in this paper. Due to the spatiotemporal correlation in the mobile data traffic, a learning algorithm can be employed that determines the number of required resources to fulfill the D2D data rate requirements. Hence, we propose a Q-learning based algorithm that learns the expected number of resource blocks (RBs) required to meet the data rate requirements at each DDC. In addition, a cache content management policy is proposed that exploits the popularity of the contents in order to increase the chances of D2D communication. With extensive simulations, we observe that the proposed Q-learning algorithm indeed learns by reserving near-optimal number of RBs to serve the data rate requirement at each DDC. Further, we observe that the one- and two-hop modes of D2D transfer effectively reduces the load on eNB by transferring a maximum 49% of the required data to the UEs.