Unmanned Aerial Vehicle base stations (UBSs) are able to assist a cellular IoT network to provide content delivery service for ground users. This paper studies the UBS deployment problem in a cache-enabled UBS-assisted cellular network. The UBS deployment problem is first formulated as a joint optimization problem with an objective to minimize the average content delivery delay of all users in a service area. We decompose the problem into three sub-problems: content caching deployment, position deployment, and BS association, and propose a multi-agent proximal policy optimization (MAPPO)-based UBS deployment algorithm to solve the sub-problems. Specifically, the proposed algorithm uses edge agents on UBSs and a two-layer MAPPO algorithm including a cache layer and a position layer to solve the content caching deployment sub-problem and the UBS position deployment sub-problem. Meanwhile, it uses a central agent in a core network and a proximal policy optimization (PPO) algorithm to solve the BS association sub-problem. Simulation results demonstrate that the proposed UBS deployment algorithm can significantly improve the network performance in terms of the average content delivery delay and the cache hit ratio of all users in the network.

Uncrewed aerial vehicle base stations (UBSs) can be used to assist a cellular Internet of Things (IoT) network to provide content delivery service for ground users. This article studies the UBS deployment problem in a cellular IoT network for content delivery and formulates the problem as a joint mixed-integer linear programming problem with an objective to minimize the average content delivery delay of all users in a service area in a time frame. The formulated problem is decomposed into three subproblems: a content caching deployment problem, a UBS position deployment problem, and a BS association problem. A deep reinforcement learning-based UBS deployment (DRL-UD) algorithm is proposed to solve the problem. In the DRL-UD algorithm, an Informer-based user pattern prediction algorithm is introduced to predict the content request pattern and mobility pattern of users. Based on the prediction of user patterns, a two-layer proximal policy optimization (TLPPO)-based UBS deployment algorithm is introduced to solve the three subproblems using a cache layer, a position layer, and an implicit enumeration method, respectively. Simulation results show that the proposed DRL-UD algorithm can significantly reduce the average content delivery delay and increase the cache hit ratio of all users in the network.

Metaverse can provide powerful human-centric interactive experiences for users and metaverse application content is the fundamental component that supports the metaverse. Considering that metaverse users are more sensitive to the delay of content delivery service, unmanned aerial vehicles (UAV) can be used as aerial base stations (BSs) to assist a cellular network to provide better content delivery service for delay-sensitive metaverse users as UAV base stations (UBSs) have big potential for line-of-sight (LoS) transmission and can be deployed closer to metaverse users than macro base stations (MBSs). This paper studies the content delivery performance analysis of a cache-enabled UBS-assisted cellular network for metaverse users. Analytical models are derived for investigating the content delivery performance of the network in terms of the content delivery success probability of the network and the average content delivery delay of a metaverse user. In deriving the analytical models, a more realistic repulsive point process is considered for modeling the location distribution of MBSs, and an air-to-ground (A2G) channel model with both a LoS link and a non-line-of-sight (NLoS) link is considered. Moreover, the cache hit probability of a UBS using a probabilistic caching strategy is also taken into consideration. A BS association strategy for delay-sensitive metaverse users based on the strongest average received power at a user and the cache hit probability of a UBS is proposed. In addition, the association probabilities with the association strategy are derived for different types of base stations. A lower bound of the content delivery success probability and an upper bound of the average content delivery delay are obtained based on the derived analytical models. The numerical results justify the effectiveness and advantage of the proposed BS association strategy and show that there exist an optimal UBS height and an optimal value of the number of UBSs, which result in the optimal content delivery performance. The obtained results can provide theoretical guidance for the deployment of UBSs in a cache-enabled UBS-assisted cellular network to provide better human-centric content delivery service for metaverse user.

This paper analyzes the coverage performance of a backhaul-limited UBS-assisted cellular network and focuses on the downlink coverage probability of the network, taking into account the effects of both access links and backhaul links. Based on stochastic geometry, a theoretical model is derived to build the relationship between the downlink coverage probability and relevant system parameters. In deriving the model, the distributions of UAV base stations (UBSs) and macro-base stations (MBSs) are modeled as two independent Poisson point processes (PPPs). For access links and backhaul links, both line-of-sight (LoS) links and non-line-of-sight (NLoS) links are considered. Moreover, an association strategy in which a user selects a BS providing the largest average received power for connection is considered. The derived theoretical model is validated through simulation results. Based on numerical results, the impacts of relevant system parameters and UAV parameters on the coverage probability of the network are investigated. The results indicate that backhaul links have a big impact on the coverage performance of the network and are expected to be guaranteed. Moreover, the derived theoretical model can be used to provide a theoretical basis for the deployment of UBSs

Unmanned Aerial Vehicle base stations (UBSs) can be used to assist a ground cellular network to enhance its network services for cellular users. This paper studies the coverage performance analysis of a cache-enabled UBS-assisted cellular network. Analytical models are derived for investigating the overall coverage probability of the network and the average achievable rate of a cellular user. In deriving the analytical models, an air-to-ground (A2G) channel model with both a line-of-sight (LoS) link and a non-line-of-sight (NLoS) link, a BS association strategy based on the strongest average received power, and a cache model with a probabilistic caching strategy are considered. The cache hit probability of UBSs based on the cache model is also taken into consideration. Moreover, the association probabilities with the association strategy are derived for different types of base stations. The derived analytical models are validated through simulation results and the impacts of system parameters on the coverage performance of the network are investigated through numerical results. Compared with existing relevant work, the novelty of this work is that the effects of both an access link and a backhaul link are taken into account in the coverage performance analysis. The obtained results can provide theoretical guidance for the deployment of UBSs in a cache-enabled UBS-assisted cellular network.

Unmanned Aerial Vehicles (UAV) deployed in the air can be used as base stations (BSs) to provide uplink and downlink transmissions for ground cellular users in a target area. This paper considers the coverage recovery problem in a cellular network with UAVs as base stations when one or more UAVs go offline due to energy replenishment or extreme environment. A coverage analysis of the target area is first presented in order to obtain the condition on the coverage radius of a UAV BS (UBS) for producing a full coverage of the target area. In the case of UAV offline, the coverage radius of a UBS needs to be adjusted to recover the full coverage of the target area, which can be implemented by adjusting either the altitude or the transmission power of the UBS. Based on the obtained full coverage condition on the UBS coverage radius, a coverage recovery analysis is further presented for determining the adjustment range of the altitude or transmission power of a UBS. For this purpose, the relationship between the coverage radius and the altitude and that between the coverage radius and the transmission power are analyzed. Through numerical results, it is demonstrated that the full coverage of a target area can effectively be recovered by adjusting either the altitude or transmission power of a UBS in the case of UBS offline.