Mehyar Najla

While the integration of flying base stations (FlyBSs) into future mobile networks has received plenty of attention, a backhaul link (i.e., the link between a static base station and the FlyBS) is often either fully disregarded or oversimplified. However, the backhaul link and an access link between the FlyBS and users should be managed together to exploit radio resources efficiently. Thus, we introduce a novel framework considering the FlyBSs with a realistic backhaul to maximize the sum capacity of the users. First, we propose a scheme for an association of the users and a transmission power allocation. Thus, we derive a closed-form expression for the optimal allocation of the FlyBSs’ transmission power to individual users to utilize the radio resources at the backhaul and access links in an efficient way. Second, we develop an algorithm for a repositioning of the FlyBSs and a reallocation of the FlyBSs’ transmission power to further improve the overall sum capacity. Third, we design a scheme reusing the access links by multiple users in the coalitions to reduce the FlyBSs’ transmission power. The reduced transmission power allows to further increase the sum capacity of the users via an additional repositioning of the FlyBSs. Alternatively, the reduced transmission power also lowers the level of interference experienced by the underlying devices not communicating via the FlyBSs. Our proposal increases the sum capacity of the users by up to 60% while suppressing the interference to the underlying devices by up to 7.7 dB compared to the state-of-the-art schemes.

Transparent flying relay stations (FlyRSs), represented by transparent relays mounted on unmanned aerial vehicles (UAVs), have the potential to improve cellular network's capacity and coverage at little extra complexity and energy cost, especially when compared with non-transparent relays. As the transparent relays do not transmit reference signals, they do not lend themselves easily to channel estimation. This makes solving the problems of user association and positioning of transparent FlyRSs much harder. We propose a solution enabling an efficient association of users to the FlyRSs and determining suitable positions of the FlyRSs. Surprisingly, this can be done knowing neither the qualities of the channels linking the FlyRSs and the users nor the users' location information. Our approach involves the users being grouped into clusters based on the channels to nearby static base stations via agglomerative hierarchical clustering. Then, 3D positions of one FlyRS per cluster are determined by deep neural networks. The proposal improves the users' sum capacity with respect to existing solutions that rely on the knowledge of users' positions.

Device-to-device communication (D2D) is expected to accommodate high data rates and to increase the spectral efficiency of mobile networks. The D2D pairs can opportunistically exploit channels that are not allocated to conventional users in a dedicated mode. To increase the sum capacity of D2D pairs in the dedicated mode, we propose a novel solution that allows the reuse of multiple channels by multiple D2D pairs. In the first step, the bandwidth is split among D2D pairs so that each pair communicates at a single channel that guarantees a minimal capacity for each pair. Then, the channel reuse is facilitated via a grouping of the D2D pairs into coalitions. The D2D pairs within one coalition mutually reuse the channels of each other. We propose two approaches for the creation of the coalitions. The first approach reaches an upper-bound capacity by optimal coalitions determined by the dynamic programming. However, such approach is of a high complexity. Thus, we also introduce a low-complexity algorithm, based on the sequential bargaining, reaching a close-to-optimal capacity. Moreover, we also determine the transmission power allocated to each reused channel. Simulations show that the proposed solution triples the sum capacity of the state-of-the-art algorithm with the highest performance.

Device-to-device (D2D) communication, which enables a direct connection between users while bypassing the cellular channels to base stations (BSs), is a promising way to offload the traffic from conventional cellular networks. In D2D communication, optimizing the resource allocation requires the knowledge of D2D channel gains. However, such knowledge is hard to obtain at reasonable signaling costs. In this paper, we show this problem can be circumvented by tapping into the information provided by the estimated cellular channels between the users and surrounding BSs as these channels are estimated anyway for a normal operation of the network. While the cellular and D2D channel gains exhibit independent fast fading behavior, we show that average gains of the cellular and D2D channels share a non-explicit relation, which is rooted into the network topology, terrain, and buildings setup. We propose a deep learning approach to predict the D2D channel gains from seemingly independent cellular channels. Our results show a high degree of convergence between the true and predicted D2D channel gains. Moreover, we demonstrate the robustness of the proposed scheme against environment changes and inaccuracies during the offline training. The predicted gains allow to reach a near-optimal capacity in many radio resource management algorithms.

This letter focuses on the selection between radio frequency (RF) and visible light communications (VLC) bands for users exchanging data directly with each other via device-to-device (D2D) communication. We target to maximize the energy efficiency of D2D communication while the outage is minimized. Since the VLC channel can vary quickly due to the possible changes in irradiance and incidence angles, we aim to reach a quick band selection decision in a multi-user scenario based only on the knowledge of the received power and sum interference from all D2D transmitters at the individual D2D receivers. The proposed solution is based on a deep neural network making an initial band selection decision. Then, based on the DNN’s output, a fast heuristic algorithm is proposed to further improve the band selection decision. The results show that the proposal reaches a close-to-optimal performance and outperforms the existing solutions in complexity, outage ratio, and energy efficiency.

The concept of device-to-device (D2D) communication, combining common radio frequency (RF) and visible light communication (VLC), is seen as a feasible way how to cope with spectrum crunch in the RF domain and how to maximize spectral efficiency in general. In this paper, our objective is to decide when RF should be utilized or if VLC proves to be the more profitable option. The selection between RF and VLC is defined as a multi-objective optimization problem targeting primarily to minimize the outage ratio while the secondary objective is to maximize the sum capacity of D2D pairs, composed by D2D transmitters and D2D receivers. To solve this problem, we design a centralized low-complexity heuristic algorithm selecting either RF or VLC band for each D2D pair relying on the mutual interference among the pairs. For interpretation of the mutual interference among the D2D pairs, we exploit directed weighted graphs adopted from the graph theory. The simulation results show that the proposed algorithm outperforms state-of-the-art algorithms in terms of the outage ratio, sum capacity and average energy efficiency. What is more, despite a very low complexity, the proposed algorithm reaches a close-to-optimum performance provided by the exhaustive search algorithm.

Device-to-device (D2D) communication, which enables a direct connection between users while bypassing the cellular channels to base stations (BSs), is a promising way to offload the traffic from conventional cellular networks. In D2D communication, one recurring problem is that, in order to optimally allocate resources across D2D and cellular users, the knowledge of D2D channel gains is needed. However, such knowledge is hard to obtain at reasonable signaling costs. In this paper, we show this problem can be circumvented by tapping into the information provided by the estimation of the cellular channels between the users and surrounding BSs as this estimation is done anyway for a normal operation of the network. While the cellular and D2D channel gains exhibit independent fast fading behavior, we show that average gains of the cellular and D2D channels share a non-explicit correlation structure, which is rooted into the network topology, terrain, and buildings setup. We propose a machine (deep) learning approach capable of predicting the D2D channel gains from seemingly independent cellular channels. Our results show a high degree of convergence between true and predicted D2D channel gains. The predicted gains allow to reach a near-optimal communication capacity in many radio resource management algorithms.

In this letter, the goal is to maximize sum capacity of device-to-device (D2D) communication through a reuse of each radio channel by multiple D2D pairs while each D2D pair can access multiple channels. Since existing approaches cannot be easily extended to enable reuse of multiple channels by multiple D2D pairs in scenario with a high interference among the D2D pairs, we propose a novel resource allocation consisting of two phases. In an initial phase, all available channels are assigned by the Hungarian algorithm so that each channel is occupied by just one D2D pair. In a reuse phase, multiple D2D pairs are sequentially added to the individual channels according to their priority expressed by channel quality and received interference from already added D2D pairs. The proposal significantly outperforms existing solutions and reaches close to theoretical upper bound capacity despite a very low complexity of the proposed algorithm.

Future mobile networks are supposed to serve high data rates required by users. To accommodate the high data rates, a direct communication between nearby mobile terminals (MTs), known as Device-to-device (D2D) communication, can be exploited. Furthermore, a communication in high frequency bands, such as, visible light communication (VLC), is also foreseen as an enabler for the high data rates. In conventional D2D communication, pairs of the MTs should reuse the same frequencies to keep a high spectral efficiency of the system. However, this implies either interference among the D2D pairs or utilization of complex and advanced resource allocation algorithms. In this paper, we propose a new concept for D2D communication combining VLC and RF technologies in order to maximize capacity of the system. This paper provides an analysis of the operational limits for the proposed concept and investigates capacity gains introduced by the combined usage of RF and VLC bands for D2D. Moreover, we discuss several practical issues related to the proposed RF-VLC D2D concept. Performed analyses show that the RF--VLC D2D is able to improve the capacity in an indoor scenario by a factor of 4.1 and 1.5 when compared to stand-alone RF D2D and VLC D2D, respectively.