Unmanned aerial vehicles (UAVs) are emerging as a promising platform for enabling integrated sensing and communication (ISAC) in multi-cell systems due to their deployment flexibility. However, this flexibility also introduces significant challenges, particularly co-channel interference at the UAV receiver. In this paper, we propose a novel adaptive co-channel interference mitigation model for UAV-enabled multi-cell systems. Specifically, the proposed model consists of two key components: a cost function and an update algorithm. First, we derive a new cost function that incorporates both magnitude and phase errors–critical metrics for guiding the estimated signal toward the desired signal. Second, the cost function is extended to formulate a parameter update algorithm, whose effectiveness is analyzed using both geometric and entropy-based approaches. Simulation results demonstrate that the proposed method outperforms state-of-the-art techniques, establishing it as a robust solution for interference mitigation in UAV-enabled multi-cell ISAC systems.

This paper studies the use of multiple planar fluid antennas at a full-duplex base station (BS) for integrated sensing and communication (ISAC). In this model, the BS communicates with a downlink user, an uplink user, and performs target sensing simultaneously. Our objective is to maximize the communication sum-rate of the up and downlink users while meeting the sensing and power constraints. Given that the problem is non-convex, we first reformulate the problem using the fractional programming (FP) framework. After that, we iteratively optimize the beamforming vectors of the BS, the uplink transmit power from the user, and the antenna positions of both transmit and receive fluid antenna systems (FASs) at the BS. In particular, the transmit and receive beamforming vectors are optimized by utilizing the majorization-minimization (MM) framework, and a closed-form solution for the uplink transmit power is derived. To optimize the BS antenna positions, we transform the problems into convex quadratically constrained quadratic programs (QCQP) by using Taylor series expansion. The subproblems can then be solved based on the successive convex approximation (SCA). Simulation results show that FAS can greatly improve the communication rate compared to the traditional fixed-position antenna (FPA) system.

Network slicing has become a primary function of 5G/6G network resource management. However, the existing slicing schemes have not sufficiently discussed the reconfiguration optimization schemes brought by user behavior changes and mobile network environment fluctuations, leading to excessive service interruption rates and slice reconfiguration costs in dynamic environments. To address this problem, this paper proposes an End-to-end Steady-state Adaptive slicing method for Dynamic network state and load (ESAD). To realize the steady-state slicing decisions, ESAD takes the steady-state degree of network slicing and reconfiguration cost as the objective and constructs the slicing reconfiguration probability evaluation function based on the service load dynamics function and the time-varying function of the network channel conditions. To improve the predictability and steady-state degree of the slicing decision, ESAD introduces an ensemble deep learning method to predict the load service fluctuation based on the user behavior model and employs reinforcement learning to compute the channel dynamics boundary, which guides the slicing decision to balance the network dynamics factors. Experiments on quality of service assurance for 5G cloud game rendering class prove that ESAD can reduce reconfiguration probability and long-term reconfiguration cost by 49.45%–58.50% while improving system QoS assurance and capacity.

This paper investigates the capacity performance of an uplink fluid antenna system (FAS), in which the base station (BS) is equipped with multiple fluid antennas and each user has a single fluid antenna. We aim to maximize the capacity of the system by optimizing the transmit power, and the user and BS antenna positions. Beginning with simple cases where the number of paths or the number of BS antennas is small, we reveal that the capacity is independent of the antenna positions. Then we give an upper bound on the capacity for the case where the BS has a single fluid antenna. After that, we show that in the optimal case, all users should transmit at the maximum power. Moreover, we propose an alternative algorithm to iteratively optimize the antenna positions at the BS and user sides. When keeping the user antenna positions fixed, the BS antenna positions are updated alternatively using a discrete exhaustive search in the single-user case. By transforming the capacity maximization problem into a difference-of-convex (DC) form, the majorization-minimization (MM) algorithm can also be applied to jointly optimize the BS antenna positions when there is a single user in the system. For the multiuser scenario, the antenna positions at the BS side are optimized utilizing the gradient descent method. We show that the user antenna positions can also be optimized using the discrete exhaustive search or the MM algorithm. Simulation results show that FAS can greatly improve the system capacity compared to traditional fixed-position antenna systems.

In recent years, the operation and maintenance of Intelligent Computing Center(ICC) have been moving toward the direction of cloud-native, intelligent, and green. Applications deployed in ICC are increasingly adopting containerization technology and microservice architecture. Although the adoption of microservice architecture is expected to simplify the development, deployment, and maintenance of software, the resource competition of microservices can lead to the degradation of application performance and thus affect the quality of user experience. In addition, dependencies between microservices can introduce the cascading effect. These issues make it challenging for operators to achieve efficient resource management while maintaining the quality of user experience.In this paper, we analyze the impact of dependencies between microservices on application performance. We model the service dependency graph as a weighted graph and propose a hybrid resource allocation method based on Deep Q-network. With the goal of maximizing the application performance while minimizing resources usage, our method evaluates the importance of microservices based on the service dependency graph to set the reward function. Using a reinforcement learning algorithm, our method adjusts the compute and network resources allocated to an application based on dynamically changing available resources. Experimental results show that our method can converge quickly and improve the application performance by approximately 68% while reducing the use of compute and network resources.

Reconfigurable intelligent surfaces (RISs) are deemed as a potential technique for the future of the Internet of Things (IoT) due to their capability of smartly reconfiguring the wireless propagation environment using a large number of low-cost passive elements. To benefit from RIS technology, the problem of RIS-assisted channel state information (CSI) acquisition needs to be carefully considered. Existing channel estimation methods usually ignored the different channel characteristics of direct channel and reflected channels. In fact, the reflected channel can be smartly configured by adjusting the phase shifts of the RIS, which is different from the direct channel due to the different path loss exponents between the transmitter and receiver. Therefore, it is necessary to further develop a RIS-assisted channel estimation to determine the direct and reflected channels, respectively. In this paper, we study a RIS-assisted channel estimation that jointly exploits the properties of the direct and the reflected channel to provide more accurate CSI. The direct channel is estimated using weighted$\ell_1$norm minimization, while the reflected channel is modeled based upon the robust$\ell_{1,\tau}$norm minimization to sequentially estimate the channel parameters. Moreover, by combining the gradient descent and the alternating minimization method, a flexible and fast algorithm is developed to provide a feasible solution. Simulation results demonstrate that an RIS-aided MIMO system significantly reduces the active antennas/RF chains compared to other benchmark schemes.

This paper investigates a reconfigurable intelligent surface (RIS)-aided unmanned aerial vehicles (UAVs) system with non-orthogonal-multiple access (NOMA), where the transmit signals from multiple UAVs to ground users are strengthened through a RIS. An innovative framework is designed to minimize the total power consumption of the system, by jointly optimizing the position of UAVs, RIS reflection coefficients, active beamforming vectors and decoding order. To solve this problem, we first consider the sub-solution of the UAV’s location which can be achieved via the successive convex approximation (SCA) and maximum ratio transmission (MRT). By applying the Gaussian randomization procedure, we then yield the closed-form solution for RIS phase coefficients. Subsequently, the transmit power is obtained by the standard convex optimization methods. Finally, a dynamic-order decoding scheme is proposed to optimize the decoding order. Simulation results show that the resource allocation scheme can obviously reduce the total power consumption compared to the benchmark schemes.