Intelligent metasurfaces are emerging as a key technology for future wireless systems, enabling programmable control of electromagnetic wave propagation. Compared to conventional single-layer reconfigurable intelligent surfaces (RIS), stacked intelligent metasurfaces (SIM) introduce multiple reconfigurable layers to provide more flexible and precise beamforming. This paper investigates the integration of SIM into a downlink non-orthogonal multiple access (NOMA) system to improve spectral efficiency while maintaining low hardware complexity. The proposed system combines maximum ratio transmission (MRT) precoding at the base station, SIM-assisted analog beamforming, and NOMA-based power allocation. To maximize the system sum rate, we perform joint optimization of SIM phase shifts and user power levels through an alternating optimization (AO) framework, where each variable is updated iteratively while the other is held fixed. We evaluate three SIM-assisted strategies: NOMA, water-filling, and uniform power allocation. Simulation results demonstrate that the SIM-NOMA configuration achieves the 40% sum rate improvements, outperforming the other schemes while leveraging the low-cost wave-domain processing capabilities of SIM.

This article investigates a near-field secure simultaneous wireless information and power transfer (SWIPT) network employing a non-orthogonal multiple access (NOMA) scheme. In this network, a base station equipped with an extremely largescale array antenna (ELAA) communicates with and transfers power to multiple single-antenna zero-energy devices (ZEDs) within the near-field region, while an eavesdropper attempts to wiretap the communication by intercepting the information signals. Specifically, we aim to maximize the overall secrecy sum rate of the ZEDs while ensuring a minimum energy harvesting criterion at the ZEDs. Consequently, a non-convex resource optimization problem is formulated and solved using an iterative approach, leveraging weighted sum-rate maximization via minorization-maximization (WSR-MM) with second-order cone programming (SOCP) transformation and general convex approximations. Finally, numerical results are presented to demonstrate the performance of the proposed near-field secure SWIPT system under varying network parameters.

Smart wireless communications enabled by reconfigurable intelligent surfaces (RISs) have gained significant research interest in the areas of localization and sensing over the past few years. This paper investigates an unconventional approach for 3D direction-of-arrival (DoA) estimation of multiple aerial user targets using an RIS-based communication architecture. In particular, the measurements required for DOA estimation at the receivers are optimized through a deep reinforcement learning framework. The core of the proposed method lies in formulating the DoA estimation problem as a Markov decision process (MDP), which is optimized via a proximal policy optimization algorithm for its optimization. Considering a practical RIS setup with 2-bit states at each unit element, we demonstrate significant improvements in DoA estimation accuracy, in terms of reduced root mean squared error (RMSE) for various simulation scenarios of the system.

This paper investigate an unconventional integration of integrated sensing and communication (ISAC) with mobile broadband ultra-reliable low-latency communication (mBRLLC). Specifically, a dual-functional radar-communication access point (DFRC-AP) with extremely large aperture arrays (ELAA) serve multiple communication nodes and perform near-field radar sensing while meeting strict URLLC requirements. Unlike conventional works focusing on rate maximization or latency-reliability trade-offs, the prime objective is to maximize the worst-case achievable rate while considering sensing and block-length constraints. A hybrid beamforming design problem is formulated which is solved using a sub-optimal alternating optimization (AO) framework, combining convex approximations and manifold optimization. Extensive simulations demonstrate fast convergence and robust performance under varying system parameters, showcasing the potential of the proposed framework to improve communication and sensing efficiency in next-generation networks. The results demonstrate that hybrid beamforming offers a optimal trade-off between communication and sensing performance with improved computational efficiency and hardware cost when comapred to fully-digital precoding. Moreover, the near-field considerations enhances sensing and mBRLLC performance with more users.

As the proliferation of Internet of Things (IoT) devices accelerates, next-generation wireless networks face unprecedented demands for secure, efficient, and scalable communication frameworks. This paper investigates the integration of non-orthogonal multiple access (NOMA) with reconfigurable intelligent surfaces (RIS) and full-duplex (FD) operations to address these challenges while mitigating the adverse effects of residual hardware impairments (HWI) that cause signal distortion. The proposed RIS-aided FD-NOMA system is designed to enhance resilience and secrecy in IoT communication networks, optimizing the secrecy rate while adhering to power constraints for active beamforming at the base node (BN) and unit-modulus requirements for passive beamforming via RIS. Employing an alternate optimization (AO) framework, the complex joint optimization problem is divided into tractable subproblems solved through generalized convex approximations to achieve near-optimal solutions. Numerical results validate the superiority of the proposed model, demonstrating substantial performance improvements over conventional IoT systems without RIS support or with half-duplex NOMA protocols. The findings underscore the transformative potential of RIS-aided FD-NOMA systems in securing IoT networks against hardware imperfections while meeting their stringent connectivity and security demands.

Next-generation wireless applications are expected to enable extended ultra-reliable low-latency communication (URLLC) to support high data rates along with ultra-reliability and low-latency features beyond the capabilities of existing core services. There is a need to transition from conventional architectures to more efficient and robust multiple-access schemes to meet these consolidated requirements in resource-constrained systems. This study explores the utilization of simultaneous transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) in non-orthogonal multiple access (NOMA) systems to enable spectrally efficient URLLC, even under the imperfect channel state information. In particular, we focus on maximizing spectral efficiency by jointly designing robust beamforming at the base station and STAR-RIS subject to given URLLC requirements. Due to the non-convexity of the formulated problem, we propose an alternating optimization framework that obtains sub-optimal solutions to the problems of beamforming design at the BS and STAR-RIS, respectively by exploiting$\mathcal {S}-$procedure and successive convex approximation. Simulation results confirm that the STAR-RIS-NOMA system can significantly boost the spectral efficiency by 10-15% compared to conventional reflecting-only RIS while guaranteeing the strict URLLC requirements. Specifically, among all the possible modes of STAR-RIS, the time-splitting mode provides better spectral efficiency than other modes owing to its better interference management.

This paper investigates a digital twin (DT) and reconfigurable intelligent surface (RIS)-aided mobile edge computing (MEC) system under given constraints on ultra-reliable low latency communication (URLLC). In particular, we focus on the problem of total end-to-end (E2E) latency minimization for the considered system under the joint optimization of beamforming design at the RIS, power, bandwidth allocation, processing rates, and task offloading parameters using DT architecture. To tackle the formulated non-convex optimization problem, we first model it as a Markov decision process (MDP). Later, we adopt deep deterministic policy gradient (DDPG) based deep reinforcement learning (DRL) algorithm to solve it effectively. We have compared the DDPG results with proximal policy optimization (PPO), modified PPO (M-PPO), and conventional alternating optimization (AO) algorithms. Simulation results depict that the proposed DT-enabled resource allocation scheme for the RIS-empowered MEC network using DDPG algorithm achieves up to 60% lower transmission delay and 20% lower energy consumption compared to the scheme without an RIS. This confirms the practical advantages of leveraging RIS technology in MEC systems. Results demonstrate that DDPG outperforms M-PPO and PPO in terms of higher reward value and better learning efficiency, while M-PPO and PPO exhibit lower execution time than DDPG and AO due to their advanced policy optimization techniques. Thus, the results validate the effectiveness of the DRL solutions over AO for dynamic resource allocation w.r.t. reduced execution time.

As a revolutionary paradigm for green ultra-reliable low-latency communication (URLLC), reconfigurable intelligent surfaces (RISs) have been considered as a prominent architecture for enabling next-generation communication systems. Recently, a novel RIS framework, called simultaneous transmitting and reflecting (STAR-RIS), has been proposed to facilitate both transmission and reflection through the meta-material surface, leading to full-space coverage and even better beamforming flexibility than conventional RIS. This paper investigates an energy-efficient resource allocation design scheme for a STAR-RIS-aided downlink system under various STAR-RIS modes to deliver energy-efficient URLLC services by jointly optimizing the beamforming at the base station (BS) and STAR-RIS, subject to the given requirements on the rate, packet-error probability, and latency. Owing to the non-convex and NP-hard nature of the formulated problem, we propose an alternating optimization framework that obtains suboptimal solutions to the problems of beamforming design at the BS and STAR-RIS, respectively, in an iterative manner by exploiting fractional programming and successive convex approximation approaches. Simulation results confirm that the TS, ES, and MS modes of STAR-RIS achieve approximately$30\%-50\%$,$20\%-40\%$, and$10\%-15\%$, respectively better performance than a conventional reflecting-only RIS while guaranteeing strict reliability and latency requirements of URLLC. Specifically, among all the possible modes of STAR-RIS, the time-splitting mode renders an effective solution due to its better interference management.

This paper investigates an unconventional superposition scheme, i.e., rate-splitting multiple access (RSMA) to maximize the overall user fairness and high system performance gain for ultra-reliable low-latency communication (URLLC), enhanced mobile-broadband (eMBB) traffic coexistence in uplink scenarios. In particular, we focus on maximizing the worst-case performance of uplink eMBB and URLLC users when multiplexed in a given resource block using an effective rate-splitting approach among multiple sub-messages. Subsequently, a multi-objective optimization problem (MOOP) is formulated to jointly maximize the worst-case rate and minimize the worst-case packet-error probability (PEP) for eMBB and URLLC users, respectively, using effective power splitting and successive interference cancellation (SIC) decoding of the sub-messages. To solve the non-convexity of the formulated MOOP, we adopt a priori articulation scheme combined with the weighted product approach to transforming the MOOP into a single objective optimization problem (SOOP) and later, solve it using a low complex differential evolution (DE)-based meta-heuristic algorithm. We derive an optimal decoding strategy for sub-messages to ensure better user fairness among eMBB-URLLC traffic. Numerical simulations demonstrate the superiority of the considered RSMA-based superposition for hybrid eMBB-URLLC traffic over conventional slicing and superposition techniques. Moreover, the adopted weighted product method-based DE algorithm outperforms the state-of-art solutions.

Re-configurable intelligent surface (RIS)-aided communication has been envisaged as a frontier scheme to enable ultra-high spectral efficiency (SE) and energy efficiency (EE) for next-generation communication. This paper investigates an unconventional framework of RIS-aided full-duplex (FD) multi-user multiple-input multiple-output (MIMO) communication and analyzes its resource efficiency (RE), a preferable performance metric for realizing trade-off between SE and EE maximization. In particular, we focus on the RE maximization problem via a joint optimization of transmit covariance, optimal receive covariance, and phase-shift matrices for each RIS subject to the given constraint on the power budget. To solve the formulated non-convex problem, we propose two optimization approaches: a) policy gradient-based deep-reinforcement learning (DRL) algorithm based on a Markov decision process formulation for a stochastic-time varying channel and b) alternate optimization (AO) algorithm based on general approximations and majorization-minimization (MM) for static channel conditions. Simulation results validate the out-performance of the considered RIS-aided FD-MIMO system compared to the counterpart system with half-duplex (HD) mode and without RIS case. The proposed DRL algorithm achieves comparable RE performance with reduced computational complexity and running time compared to the traditional AO-based algorithm.

Next-generation wireless applications are expected to enable extended ultra-reliable low latency communication (xURLLC) to support high data rates along with ultra-high reliability and low end-to-end latency features beyond the capabilities of existing core services. These consolidated data rates and URLLC requirements in resource-constrained systems necessitate the shift from conventional architectures to more powerful and robust multiple access schemes. This paper investigates a multi-reconfigurable intelligent surface (RIS)-assisted rate-splitting multiple access (RSMA) to prompt an unconventional xURLLC service called mobile broadband reliable low latency communication (mBRLLC) for high spectral efficiency under finite block-length (FBL) transmission constraints. To enable spectral-efficient resource allocation, we formulate a sum throughput maximization problem for joint optimization of precoder design at the base-station (BS), block-length of common and private symbols of each user, and passive beamforming at each RIS. To solve the NP-hardness and non-convexity of the formulated problem, we use an alternating optimization technique to decouple the original problem into three sub-problems: active beamforming at the BS, block-length optimization, and passive beamforming at each RIS which are solved using general convex approximations. Simulations demonstrate the effectiveness of the proposed resource allocation algorithm over conventional schemes. The considered RSMA system achieves high data rates even with lower latency and higher reliability. Additionally, the investigation encompasses the evaluation of RIS deployment implications, the analysis of the worst-case latency scenario, and the assessment of the influence of channel estimation errors.

The low-resolution reality of the hardware elements associated with massive mmWave antenna or reflector arrays is associated with the performance degradation of the wireless link when it is not properly controlled. In particular, the unintended angular radiations of the transmission or reflection arrays (e.g., transmission in non-intended directions) would invalidate the usual assumptions of information secrecy, even with perfect channel state information (CSI) knowledge at the transmitter, in the presence of low-resolution hardware. In this paper, we study a hybrid beamforming design for reconfigurable intelligent surface (RIS)-assisted multi-user multiple-input multiple-output (MU-MIMO) downlink (DL) communication, from the prospect of information secrecy maximization, wherein the array element phase rotations belong to the known discrete space. To address the NP-hard and non-convex nature of the problem at hand, we propose an iterative procedure by re-structuring the obtained discrete-domain problem into a tractable form which solves the problem numerically and guarantees the convergence to a stationary point. Further, we confirm the accuracy of the proposed optimization algorithm by an exhaustive search method based on graphical simulations. The minimal performance disparity that exists between the proposed algorithm and the considered digital beamforming (DBF) scheme as the upper bound validates the hybrid beamforming design. Moreover, the proposed work highlights the superiority of discrete-aware design over various existing baseline schemes, demonstrating the significant gains attainable by adopting discrete space design from the outset. Additionally, the proposed solution discusses the improvement in secrecy system performance by deploying RIS with an increased number of reflecting elements and thereby restricting the effect of eavesdroppers on secure communication.

As a revolutionary paradigm for green communication architecture for next-generation, reconfigurable intelligent surfaces (RISs) has been considered for simultaneous wireless information and power transfer (SWIPT). Nevertheless, the performance gain achieved by the conventional passive RISs is limited due to the multiplicative fading effect. In this paper, we investigate an unconventional framework of active reconfigurable intelligent surface (ARIS) aided multi-user (MU) multi-input multi-output (MIMO) system to captivate better performance for the SWIPT system. Particularly, we focus on the problem of power minimization via joint beamforming design at the base station (BS) and the ARIS for the considered SWIPT system under statistical channel estimation error (CEE) while guaranteeing the minimum rate requirement and the minimum energy-harvested constraints for information and energy receivers, respectively. Owing to the non-convex and NP-hard nature of the formulated problem, we first utilize a minimum mean square error (MMSE) approach to transform the problem into its simplified form, and later utilize an alternating optimization framework which solves the problems of beamforming design at the BS and the ARIS independently in an iterative manner using general approximations. Simulation results confirm that the ARIS can significantly reduce the required transmission power by 50-60% when compared to passive RIS while satisfying given QoS constraints for SWIPT system under the CEE model.

Next-generation wireless applications are expected to enable extended ultra-reliable low latency communication (URLLC) to support high data rates along with ultra-high reliability and low end-to-end latency features beyond the capabilities of existing core services. This paper investigates simultaneous transmission and reflection re-configurable intelligent surface (STARRIS) aided non-orthogonal multiple access (NOMA) systems to enable spectral-efficient short-packet communication under imperfect channel state information. In particular, we focus on the problem of spectral-efficiency maximization via joint beamforming design at the base station and STAR-RIS subject to given URLLC requirements. Owing to the non-convexity of the formulated problem, we propose an alternating optimization framework that obtains sub-optimal solutions to the robust beamforming design, respectively by exploiting$S$-procedure and successive convex approximation. Simulation results confirm that the STAR-RIS-NOMA system can significantly boost the spectral efficiency by 10-15% compared to conventional reflecting-only RIS while guaranteeing the strict URLLC requirements.

The proliferation of multiple devices and acceleration of spectral efficiency has become a pivotal requirement for the unprecedented connectivity and performance of vehicle-to-everything (V2X) networks. This paper investigates an unconventional framework of reconfigurable intelligent surface (RIS)-integrated full-duplex (FD) rate-splitting multiple access (RSMA) communication systems, which aims to maximize the spectral efficiency of uplink (UL) and downlink (DL) vehicles in V2X network. In particular, a robust spectral-efficient design for the considered RIS-integrated FD-RSMA system via joint beamforming design and power allocation at UL vehicles under imperfect channel state information is investigated. To tackle the non-convexity of the original sum-rate maximization problem, we adopt a deep reinforcement learning (DRL)-based proximal policy optimization (PPO) algorithm which leverages Markov decision process formulation. Simulation results demonstrate the effectiveness of the integration of RIS, RSMA, and FD schemes for V2X networks over half-duplex (HD) and multi-user linear precoding schemes. Furthermore, the superiority of the proposed PPO algorithm is validated over the counterpart deep deterministic policy gradient algorithm (DDPG).

This paper investigates the problem of spectral efficiency maximization in an underlay cognitive radio network (CRN) utilizing rate-splitting multiple access (RSMA) transmission for MISO downlink under short packet communications and imperfect channel estimation information. In particular, we focus on an effective transmit beamforming design at the cognitive base station while satisfying the requirements of ultra-reliable and low-latency communication (URLLC), interference temperature, power budget, and minimum throughput. We model the dynamic resource allocation problem as a Markov decision process (MDP) and employ deep reinforcement learning techniques, specifically the deep deterministic policy gradient (DDPG) and proximal policy optimization (PPO) algorithms, while taking into account the time-varying channel conditions. Simulation results demonstrate that the DDPG algorithm outperforms PPO at low interference temperatures for the primary receiver, while the opposite holds at high interference temperatures. Moreover, the considered RSMA system for CRN outperforms traditional multi-user linear precoding and power-domain multiple access schemes while maintaining small packet sizes and high reliability.

In this paper, we explore a novel approach of using a simultaneous transmission and reflection reconfigurable intelligent surface (STAR-RIS) to aid a multi-user (MU) multi-input multi-output (MIMO) system for simultaneous wireless information and power transfer (SWIPT) in the presence of statistical channel estimation errors (CEE). Our focus is on minimizing the power required for the SWIPT system through joint beamforming design at both the base station (BS) and STAR-RIS, while ensuring that the minimum rate and minimum energy harvesting requirements are met for information and energy receivers, respectively. Due to the non-convex and NP-hard nature of the problem, we utilize a minimum mean square error (MMSE) approach to simplify the problem and then use an alternating optimization framework to solve the beamforming design problems at the BS and STAR-RIS iteratively using general approximations. Simulation results show that the proposed algorithm provides a significant beamforming gain for STAR-RIS-aided SWIPT system over conventional RIS system while satisfying the given quality of service (QoS) constraints for SWIPT systems under the CEE.

This paper investigates an unconventional framework of active reconfigurable intelligent surface (ARIS) aided multi-user (MU) multi-input multi-output (MIMO) system for simultaneous wireless information and power transfer (SWIPT) network under statistical channel estimation error (CEE). Particularly, we focus on the problem of power minimization via joint beamforming design at BS and ARIS for considered SWIPT system while guaranteeing the minimum rate requirement and the minimum energy-harvested constraints for information and energy receivers, respectively. Owing to the non-convex and NP-hard nature of the formulated problem, we first utilize a minimum mean square error (MMSE) approach to transform the problem into its simplified form, and later utilize an alternating optimization framework which solves the problems of beamforming design at BS and ARIS independently in an iterative manner using general approximations. Simulation results confirm that ARIS can significantly reduce the required transmission power by 50-60% when compared to passive RIS while satisfying given QoS constraints for SWIPT systems under the CEE model.

This paper investigates an unconventional framework of rate-splitting multiple access (RSMA)-integrated full-duplex (FD) system to attain spectral-efficient multi-user communication. The considered FD-RSMA system divides and encodes the original messages of each downlink (DL) and uplink (UL) into two different sub-messages, and later transmits them at the same resource block, resulting in strong inter-user interference and cross-link interference, i.e., self-interference (SI) and co-channel interference (CCI). Specifically, we focus on maximizing the sum rate of the considered FD-RSMA system via joint power allocation for simultaneous UL and DL communication, subject to transmit power constraints and given quality of service (QoS) requirements. To tackle the non-convexity of the formulated problem, we adopt an iterative algorithm that employs semidefnite programming (SDP), majorization minimization (MM), and inner approximation (IA) techniques to attain near-optimal resource allocation with effective interference management. Simulation results validate that the FD-RSMA scheme outperforms conventional half-duplex, multi-user linear precoding, and non-orthogonal multiple access schemes.

In this paper, an intelligent reflecting surface (IRS) aided uplink (UL) rate-splitting multiple access (RSMA) system is investigated for dead-zone users where the direct link between the users and the base station (BS) is unavailable and the UL transmission is carried out only through IRS. In the considered RSMA system, a message of each user is split into several sub-messages and each part contributes to the rate of that user and depending upon split proportions BS decodes them using appropriate decoding order. The problem of sum-rate maximization is formulated to jointly design the optimal power allocation at each UL user, passive beamforming at the IRS under optimal decoding order of sub-messages. Due to non-convexity and discrete non-linear programming of the formulated problem, the original problem is intractable and hence, we decouple the problem into different sub-problems in which the problems of power allocation and passive beamforming are alternatively solved under using successive convex approximation and Riemaniann conjugate gradient algorithms, respectively. Moreover, the decoding order strategy is analytically derived which confirm that the optimal decoding order strategy depend upon decreasing order of channel gain of users and increasing order of split proportions of sub-messages. Later, the unified solution based on block-coordinate descent (BCD) algorithm is proposed. Simulation results validate that the proposed decoding order scheme attains performance closer to the optimal solution with low computational complexity. Moreover, the proposed IRS aided RMSA system outperforms the system with non-orthogonal multiple access (NOMA) and orthogonal multiple access (OMA) schemes in terms of achievable sum-rate throughput.

In this paper, a phase-shift coupled simultaneous transmitting/refracting and reflecting reconfigurable intelligent surface (STAR-RIS)-aided uplink (UL) rate-splitting multiple access (RSMA) system is investigated to achieve improved spectral efficiency. The considered UL RSMA system splits the rate for each user by dividing their message into multiple sub-messages and these sub-messages are transmitted to the base station (BS) via STAR-RIS as direct link between BS and user is absent. In particular, we formulate a resource allocation design problem which aim to maximize the overall rate-throughput of the considered system under the joint optimization of power allocation, decoding order, user-fairness and beamforming design at STAR-RIS for various operating modes of STAR-RIS modes, which is mixed- integer non-linear programming (MINLP). To solve the formulated non-convex complex problem, we first transform the original sum-rate maximization into its simplified form and then solved it using an alternating optimization algorithm where the sub-problems of power allocation and beamforming design under given decoding order are solved alternatively using general convex approximation and fractional programming approaches. Numerical simulation and computational complexity analysis validate that the proposed solution attains fast convergence. Moreover, the proposed RSMA scheme in STAR-RIS aided UL system outperforms the conventional nonorthogonal multiple access and orthogonal multiple access schemes in terms of overall rate-throughput and user-fairness.

In this paper, a reconfigurable intelligent surfaces (RISs)-aided millimeter wave (mmWave) uplink (UL) rate-splitting multiple access (RSMA) system is investigated which targets to achieve better rate performance and enhanced coverage capability for multiple users. The considered UL RSMA model splits the rate for each user by dividing their message into multiple parts and hence exploits all the necessary degrees of freedom to achieve maximum capacity region and high user fairness. In particular, we focus on the sum-rate maximization for considered UL RSMA system subject to joint optimization of power allocation to the UL users and beamforming design, i.e., active receive beamforming at the base-station (BS) and passive beamforming at multiple RISs. To efficiently mitigate high inter-node interference in multi-user scenario, we first provided a low-complex user pairing scheme based on k-means clustering and then develop an effective low-cost alternating optimization framework to solve the joint optimization problem sub-optimally by decoupling the problem into different sub-problems of power allocation and beamforming design. Specifically, the sub-problems of power allocation and beamforming design are solved using successive convex approximation, Riemannian manifold and fractional programming techniques. Later, the unified solution based on block coordinate descent (BCD) algorithm is proposed. Extensive numerical simulations validate that the user-clustering effectively significantly improves the performance gain and the considered RSMA system outperforms the conventional multiple schemes in terms rate and user-fairness. Also, the exploitation of spatial correlation among each RIS elements i.e., non-diagonal phase-matrices at each RIS achieve better performance that conventional diagonal phase-matrices setting.

This paper investigates unmanned aerial vehicles (UAVs)-assisted full-duplex (FD) non-orthogonal multiple access (NOMA) system based cellular network, aiming to improve overall sum-rate throughput of the system through dynamic user clustering, optimal UAV placement and power allocation. Since each UAV operates in FD mode, self-interference (SI), co-channel interference (CCI), inter-UAV interference (IUI) and intra-node interference (INI) dominate the system’s performance. Consequently, we propose an unconventional two-stage dynamic user clustering for user nodes (UNs) to reduce the cross-interference in multi-UAV aided FD-NOMA system. Particularly, all UNs are initially clustered into$K$clusters using k-means clustering in the first stage where each cluster is served by an UAV. Furthermore, each cluster is further divided into sub-clusters and each sub-clusters are operated in FD-NOMA scheme. Finally, to control interferences, a sum-rate throughput maximization problem is formulated for each UAV to jointly optimize uplink and downlink power allocation and UAV placement. The joint optimization problem is non-convex and difficult to solve directly, for which we decoupled the original problem by addressing UAV placement and power allocation separately. We first fix the UAV position and then solve the problem iteratively using successive convex approximation (SCA) method. By utilizing brute-force search algorithm, an optimal UAV placement is later performed which corresponds to maximum possible sum-rate throughput. Simulation results demonstrate that the proposed solution for the considered FD-NOMA system outperforms the conventional schemes.