Downlink Power Allocation Research Articles

Joint In-Orbit Computation and Communication for Minimizing Download Time From LEO Satellites

Downloading a large amount of data from a low Earth orbit satellite to a ground station can be challenging due to the limited contact window, dynamic channel quality, solar energy supply, and thermal management without an atmosphere. Considering such dynamics, this paper proposes a joint design of in-orbit computation and communication for download time minimization. We combine the non-convex thermal constraints and energy constraints into unified energy budget constraints with upper bound approximation, and computational efficiency is achieved by decomposing the resulting large-scale problem into a non-convex communication sub-problem, a convex computation sub-problem solvable with interior point method and a master problem that optimizes the energy budget allocation between computation and communication. The communication sub-problem is solved with a generalized-benders-decomposition-based algorithm that decouples downlink scheduling and power allocation based on a closed-form solution of optimal dual variables in the power allocation primal problem. And the master problem is solved with ternary search by proving the minimal download time is quasi-convex with respect to the energy budget allocation between computation and communication. Simulation results demonstrate that the proposed solution effectively reduces the download time, especially under strict energy constraints and severe channel variations.

Joint computational and optimized resource allocation in edge computing networks

Fog computing has several undeniable benefits, such as enhancing near-real-time response, reducing transmission costs, and facilitating IoT analysis. This technology is poised to have a significant impact on businesses, organizations, and our daily lives. However, mobile user equipment struggles to handle the complex computing tasks associated with modern applications due to its limited processing power and battery life. Edge computing has emerged as a solution to this problem by relocating processing to nodes at the network’s periphery, which have more computational capacity. With the rapid evolution of wireless technologies and infrastructure, edge computing has become increasingly popular. Nevertheless, managing fog computing resources remains challenging due to resource constraints, heterogeneity, and distant nodes. For delay-sensitive intelligent IoT applications within the fog computing architecture, cooperation and communication processing resources in 6 G and future networks are essential. This study proposes a joint computational and optimized resource allocation (JCORA) technique to accelerate the processing of data from intelligent IoT sensors in a cell association environment. The proposed technique utilizes an uplink and downlink power allocation factor and the shortest job first (SJF) task scheduling system to optimize user fairness and decrease data processing time. This is a complex assignment due to several non-convex limitations. The suggested JCORA-SJF model simultaneously optimizes time partitioning, computing task processing mode selection, and target sensing location selection to maximize the weighted total of task processing and communication performance. The simulation results demonstrate the effectiveness of the proposed JCORA-SJF algorithms, and the system’s scalability is also examined.

Efficient power allocation for downlink MIMO-NOMA-based visible light communication systems

Efficient power allocation for downlink MIMO-NOMA-based visible light communication systems

How to Design Location Development and Power Allocation for Hybrid RIS Downlink

How to Design Location Development and Power Allocation for Hybrid RIS Downlink

Downlink Power Allocation for CR-NOMA-Based Femtocell D2D Using Greedy Asynchronous Distributed Interference Avoidance Algorithm

This paper focuses on downlink power allocation for a cognitive radio-based non-orthogonal multiple access (CR-NOMA) system in a femtocell environment involving device-to-device (D2D) communication. The proposed power allocation scheme employs the greedy asynchronous distributed interference avoidance (GADIA) algorithm. This research aims to optimize the power allocation in the downlink transmission, considering the unique characteristics of the CR-NOMA-based femtocell D2D system. The GADIA algorithm is utilized to mitigate interference and effectively optimize power allocation across the network. This research uses a fairness index to present a novel fairness-constrained power allocation algorithm for a downlink non-orthogonal multiple access (NOMA) system. Through extensive simulations, the maximum rate under fairness (MRF) algorithm is shown to optimize system performance while maintaining fairness among users effectively. The fairness index is demonstrated to be adaptable to various user counts, offering a specified range with excellent responsiveness. The implementation of the GADIA algorithm exhibits promising results for sub-optimal frequency band distribution within the network. Mathematical models evaluated in MATLAB further confirm the superiority of CR-NOMA over optimum power allocation NOMA (OPA) and fixed power allocation NOMA (FPA) techniques.

Deep neural network design with SLNR and SINR criterions for downlink power allocation in multi-cell multi-user massive MIMO systems

In this paper, we propose a deep learning approach for solving power allocation problems in massive MIMO networks. We use signal-to-interference-plus-noise-ratio (SINR) and signal-to-leak-plus-noise ratio (SLNR) criteria for linear precoder design to define the max–min and max-prod power allocation challenges. The power allocation process to each user equipment in the base station coverage takes a long time and is inefficient, hence numerous base stations are deployed to serve multiple user equipments. As a result, we develop a deep neural network (DNN) framework in which the user’s equipment position is utilized to train the deep model, which is then used to forecast the ideal power distribution depending on the user’s location. Compared to the traditional optimization approach, the DNN design helps to obtain the optimal solution of the power allocation problem within a short time via a quick-inference process. Simulation results show that the SINR criterion outperforms the SLNR one. Meanwhile, deep learning achieves excellent results in forecasting power allocation with an accuracy of 85% for the max–min strategy and 99% for the max-product approach.

Federated Learning Over Fully-Decoupled RAN Architecture for Two-Tier Computing Acceleration

Two-tier computing paradigm that takes full advantage of both the end-user and the cloud computation capabilities has emerged as a promising way to deal with computationally-intensive tasks in the next generation wireless networks. For promoting the integration of the two-tier computing, federated learning (FL) provides an effective framework to enable the collaboration between the end-user and the cloud. However, the key performance metric, i.e., FL training latency, will be severely affected by the worst wireless link quality in both uplink and downlink. In this paper, aiming at accelerating the FL enabled end-cloud two-tier computing over the wireless networks, we introduce the uplink and downlink fully-decoupled radio access network (FD-RAN) architecture to enhance the minimum wireless link rate via multiple base stations (BSs) access collaboration and power management solution. First, the Lagrange dual decomposition and the binary variable relaxation methods are leveraged to obtain an optimal multiple BS access scheme for the enhancement of minimum uplink and downlink SINR. Subsequently, we exploit the successive convex approximation (SCA) algorithm to deal with the uplink power control and downlink power allocation with a proved data rate lower bound. Furthermore, considering the dynamic channel realizations, a stochastic optimization technique with a convex surrogate function is utilized to find the best end-cloud two-tier computing scheme for FL applications. Simulation results have demonstrated the effectiveness of our proposed joint multiple access collaboration and power management solution over FD-RAN for achieving a faster FL enabled two-tier computing task.

Datasets associated with the article "Learning-Based Downlink Power Allocation in Cell-Free Massive MIMO Systems" published in IEEE Transactions on Wireless Communications

Datasets associated with the article "Learning-Based Downlink Power Allocation in Cell-Free Massive MIMO Systems" published in IEEE Transactions on Wireless Communications

Learning-Based Downlink Power Allocation in Cell-Free Massive MIMO Systems

This paper considers a cell-free massive multiple-input multiple-output (MIMO) system that consists of a large number of geographically distributed access points (APs) serving multiple users via coherent joint transmission. The downlink performance of the system is evaluated, with maximum ratio and regularized zero-forcing precoding, under two optimization objectives for power allocation: sum spectral efficiency (SE) maximization and proportional fairness. We present iterative centralized algorithms for solving these problems. Aiming at a less computationally complex and also distributed scalable solution, we train a deep neural network (DNN) to approximate the same network-wide power allocation. Instead of training our DNN to mimic the actual optimization procedure, we use a heuristic power allocation, based on large-scale fading (LSF) parameters, as the pre-processed input to the DNN. We train the DNN to refine the heuristic scheme, thereby providing higher SE, using only local information at each AP. Another distributed DNN that exploits side information assumed to be available at the central processing unit is designed for improved performance. Further, we develop a clustered DNN model where the LSF parameters of a small number of APs, forming a cluster within a relatively large network, are used to jointly approximate the power coefficients of the cluster.

Power Allocation in Ultra-Dense Networks Through Deep Deterministic Policy Gradient

The existing reinforcement learning-based downlink power allocation (PA) schemes mostly consider the power optimization space as a discrete value space, however, their results will deviate from the optimal result in ultra-dense networks, and the deviation grows as the network size increases. This letter proposed a PA model based on deep deterministic policy gradient (DDPG), where policy-based power selection assisted with value-based evaluation is leveraged to explore the optimal result from a continuous power space. Specifically, this model uses two CNNs named actors to formulate continuous deterministic PA strategy function instead of discrete power distribution sampling, and designs another two CNNs named critics for PA strategy evaluation and actor CNNs’ update supervision. Additionally, to reduce the interference, a tunable service base station set is designed for each user and is considered for the model training. Experiments demonstrate the proposed DDPG-based PA model respectively reaches 116.2% and 95.9% sum-rate relative to the iterative algorithm in small and large-scale networks.

Multi-agent reinforcement learning based joint uplink–downlink subcarrier assignment and power allocation for D2D underlay networks

Multi-agent reinforcement learning based joint uplink–downlink subcarrier assignment and power allocation for D2D underlay networks

Joint Computational and Wireless Resource Allocation in Multicell Collaborative Fog Computing Networks

In 6G and future networks, joint optimization of communication and computational resources lays the foundation for various delay-sensitive intelligent IoT services in the fog computing architecture. In this paper, we present a multi-device collaborative computing architecture in the cell association environment to accelerate the processing procedure of data generated by smart IoT devices. In this scenario, a two-tier task scheduling scheme and an uplink and downlink power allocation factor are jointly optimized to reduce the data processing delay and improve fairness among different users, which is in nature a hard problem due to a series of non-convex constraints. To make the problem tractable, the problem is transformed into a smooth non-convex problem with the introduction of auxiliary variables and then decoupled into two subproblems based on the data transmission and processing procedure. Thereafter, different methods such as Successive Convex Approximation (SCA) and Block Successive Upperbound Minimization (BSUM) are employed to reconstruct several upper-bound convex optimization subproblems. Besides, a fast 0–1 binary offloading scheme is proposed based on the original algorithm. Finally, the simulation results depict the effectiveness of the proposed algorithms in detail, and the scalability of the system is also examined.

Path Design and Resource Management for NOMA Enhanced Indoor Intelligent Robots

A communication enabled indoor intelligent robots (IRs) service framework is proposed, where non-orthogonal multiple access (NOMA) technique is adopted to enable highly reliable communications. In cooperation with the ultramodern indoor channel model recently proposed by the International Telecommunication Union (ITU), the Lego modeling method is proposed, which can deterministically describe the indoor layout and channel state in order to construct the radio map. The investigated radio map is invoked as a virtual environment to train the reinforcement learning agent, which can save training time and hardware costs. Build on the proposed communication model, motions of IRs who need to reach designated mission destinations and their corresponding down-link power allocation policy are jointly optimized to maximize the mission efficiency and communication reliability of IRs. In an effort to solve this optimization problem, a novel reinforcement learning approach named deep transfer deterministic policy gradient (DT-DPG) algorithm is proposed. Our simulation results demonstrate in the following: 1) with the aid of NOMA techniques, the communication reliability of IRs is effectively improved; 2) radio map is qualified to be a virtual training environment, and its statistical channel state information improves training efficiency by about 30%; 3) proposed DT-DPG algorithm is superior to the conventional deep deterministic policy gradient (DDPG) algorithm in terms of optimization performance, training time, and anti-local optimum ability.

Downlink Power Allocation in Massive MIMO via Deep Learning: Adversarial Attacks and Training

The successful emergence of deep learning (DL) in wireless system applications has raised concerns about new security-related challenges. One such security challenge is adversarial attacks. Although there has been much work demonstrating the susceptibility of DL-based classification tasks to adversarial attacks, regression-based problems in the context of a wireless system have not been studied so far from an attack perspective. The aim of this paper is twofold: (i) we consider a regression problem in a wireless setting and show that adversarial attacks can break the DL-based approach and (ii) we analyze the effectiveness of adversarial training as a defensive technique in adversarial settings and show that the robustness of DL-based wireless system against attacks improves significantly. Specifically, the wireless application considered in this paper is the DL-based power allocation in the downlink of a multicell massive multi-input-multi-output system, where the goal of the attack is to yield an infeasible solution by the DL model. We extend the gradient-based adversarial attacks: fast gradient sign method (FGSM), momentum iterative FGSM, and projected gradient descent method to analyze the susceptibility of the considered wireless application with and without adversarial training. We analyze the deep neural network (DNN) models performance against these attacks, where the adversarial perturbations are crafted using both the white-box and black-box attacks.

Distributed DRL-Based Downlink Power Allocation for Hybrid RF/VLC Networks

Hybrid radio frequency (RF) and visible light communication (VLC) networks can provide high throughput and energy efficiency with VLC access points (APs) while ensuring ubiquitous coverage with RF APs. Due to dynamic channel conditions and limited resources, the hybrid RF/VLC networks’ resource allocation problem is complex and challenging. Conventional resource allocation techniques fail to overcome these challenges. Heuristic methods can solve high complexity problems; however, they are not robust against changes such as dynamic channel conditions or alternating user requirements. Heuristic methods require centralized control for stability which adds communication overhead between APs. Deep Reinforcement Learning (DRL) based solutions can solve high complexity, dynamic channel conditions, and alternating user requirements while not requiring centralized control. In this paper, we formulate a distributed downlink power allocation problem to optimize the transmit power for users to reach target data rates in hybrid RF/VLC networks. Then, we propose a distributed DRL-based algorithm Deep Deterministic Policy Gradient (DDPG), to solve the formulated computationally-intensive problem. We implement a simulation environment to benchmark the proposed distributed DRL-based method against other methods such as Q-Learning (QL) and Deep Q-Networks (DQN), and centralized heuristic power allocation algorithms. Our simulation results show that the distributed DDPG-based algorithm learns to adapt against changes in the channel or user requirements, while centralized Genetic Algorithm and Particle Swarm Optimization-based algorithms fail to endure against these changes even with coordination between APs. Additionally, we quantify the performance of the DDPG-based algorithm to prevail amid DRL-based algorithms at the expense of higher implementation complexity.

Dynamic User Clustering and Optimal Power Allocation in UAV-Assisted Full-Duplex Hybrid NOMA System

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&#x2019;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 <inline-formula> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula> 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.

Network sum-rate maximization via joint power allocation and antenna selection for clustered downlink/uplink NOMA networks

This paper considers the problem of joint power allocation and antenna selection (J-PA-AS) for downlink (DL) and uplink (UL) clustered non-orthogonal multiple-access (NOMA) networks. In particular, the goal is to perform antenna selection for each user cluster and allocate transmit power to its users so as to maximize the network sum-rate in the DL and UL directions, while satisfying quality-of-service (QoS) requirements. The formulated problem happens to be non-convex and NP-hard, and thus, there is no systematic or computationally-efficient approach to solve it directly. In turn, a low-complexity two-stage algorithm is proposed. Specifically, the first stage optimally solves the sum-rate maximizing power allocation for each (antenna, user cluster) pair. After that, antenna selection is optimally solved in polynomial-time complexity via the Kuhn–Munkres with backtracking (KMB) algorithm. Extensive simulation results are provided to validate the proposed algorithm, which is shown to efficiently yield the optimal network sum-rate in each link direction, in comparison to the optimal J-PA-AS scheme (solved via a global optimization package), and superior to other benchmark schemes. Light is also shed on the impact of spatial-diversity on the network sum-rate, where it is shown that the greater the number of base-station antennas is, the higher the network sum-rate, and the lower the outage events. Additionally, the significance of decoupling antenna selection in each link direction on the network sum-rate is highlighted. Lastly, the cases of imperfect channel state information (CSI) and imperfect successive interference cancellation (SIC) have been investigated, where it is demonstrated that spatial-diversity gains reduce the adverse effects of imperfect CSI and SIC on the network sum-rate.

Optimal downlink power allocation schemes for OFDM-NOMA-based Internet of things

With the continuous development of fifth-generation technology, the number of mobile terminal Internet of Things devices has increased exponentially. How to effectively improve the throughput of fifth-generation systems has become a challenge. In the Internet of Things networks, ultra-dense networks and non-orthogonal multiple access technology have drawn extensive attention in recent years, because they can achieve multiplexing from the space domain and power domain. To improve the throughput of the system, this article combines non-orthogonal multiple access with ultra-dense networks technology and considers the orthogonal frequency division multiplexing non-orthogonal multiple access–based ultra-dense networks with multiple base stations and multiple Internet of Things devices. In particular, first, we build the network model and channel model. Second, we construct the downlink transmission rate maximizing problem subject to the total power. Then, to solve this problem, we divide it into three sub-problems: device grouping, inter-sub-band power allocation, and intra-sub-band power allocation problems. Solving these sub-problems, we obtain the optimal power allocation schemes by jointly employing channel-state sorting–pairing algorithm, water-filling algorithm, and convex optimization theory. Finally, numerical simulations are conducted to validate the performance of our proposed optimal downlink power allocation scheme. Experimental results show that the total throughput of the system has been significantly improved.

Universal Adversarial Attacks on Neural Networks for Power Allocation in a Massive MIMO System

Deep learning (DL) architectures have been successfully used in many applications including wireless systems. However, they have been shown to be susceptible to adversarial attacks. We analyze DL-based models for a regression problem in the context of downlink power allocation in massive multiple-input-multiple-output systems and propose universal adversarial perturbation (UAP)-crafting methods as white-box and black-box attacks. We benchmark the UAP performance of white-box and black-box attacks for the considered application and show that the adversarial success rate can achieve up to 60% and 40%, respectively. The proposed UAP-based attacks make a more practical and realistic approach as compared to classical white-box attacks.

Rate maximization in full‐duplex systems: Gains and complexity of employing adaptive power allocation in downlink and/or uplink

AbstractThe next generations of mobile communications should deal with a set of new use cases, leading to a higher demand for spectral efficiency. Full‐duplex (FD) arises as a promising technology since it has the potential to double the system capacity without additional spectrum. However, FD networks are hindered by self‐interference (SI) and co‐channel interference caused by the reuse of frequency resources. Therefore, radio resource allocation (RRA) in FD networks by means of user pairing, frequency assignment, and power allocation are of utmost importance in order to improve spectral/energy efficiency and ensure quality of service (QoS). Although previous works investigated RRA to improve spectral efficiency, the fundamental question of where adaptive power allocation should be employed in the uplink and/or downlink of FD networks in a cost‐effective way remains unanswered. This article addresses this question in the context of total data rate maximization with QoS constraints where we propose four optimization problems using different power allocation strategies. The branch and bound method is employed to optimally solve the four formulated problems, which were used as upper bound solutions for benchmarking purposes. Moreover, we propose a low computational complexity solution able to adaptively switch the power allocation strategy depending on the scenario. Detailed system‐level simulations indicate that the best node to employ adaptive power allocation (mobile terminal or base station) depends on the cell radius and SI level. Also, our results show that, compared to existing solutions, the proposed low‐complexity is a great and intelligent strategy in terms of system performance and computational complexity considering both perfect and imperfect channel state information.