Cell-free massive multiple-input multiple-output (MIMO) systems have emerged as a transformative architecture for sixth generation (6G) communication networks, where distributed access points (APs) collaborate to simultaneously serve all user equipments (UEs). However, in time division duplex (TDD) systems, the reciprocity of uplink channel and downlink channel is disrupted by hardware imperfections in radio frequency (RF) chains, leading to significant degradation in system performance. This paper begins with a theoretical analysis of the downlink performance under a conjugate beamforming scheme, considering scenarios with and without channel calibration. A key theoretical insight highlights the limitation of conventional least squares (LS) calibration method, which fails to achieve high calibration accuracy even with an unlimited number of pilot observations. To overcome this limitation, we propose a novel channel calibration approach based on a diffusion model, designed to successively refine the calibration vector obtained from the LS calibration method. Furthermore, to address the shortcomings of conventional denoising diffusion probabilistic model (DDPM) training architectures, we introduce an innovative bridge-based diffusion model that maps the distribution of LS calibration vectors to their perfect counterparts. The proposed diffusion neural network architecture employs a conditional generative process, integrating a message passing neural network (MPNN) to incorporate domain-specific calibration insights. Numerical results demonstrate the superior performance of our proposed calibration method compared to existing methods, with supplementary experiments and in-depth analyses confirming the efficacy of the proposed successive refinement design.

The quality of channel state information (CSI) feedback is critical for maximizing the spectral efficiency of massive multiple-input multiple-output systems. With multiple antenna arrays, the overhead of direct CSI feedback in frequency division duplex mode is usually large, and many CSI compression techniques have been proposed to alleviate this problem. Deep learning (DL) has achieved tremendous strides in CSI feedback. However, most current DL-based CSI compression methods utilize fully connected layers to achieve dimensionality reduction, which may be suboptimal for network optimization and result in noteworthy information loss and reduced CSI reconstruction accuracy. In this paper, we propose a novel recursive discretization compression framework with a selective state space model for CSI feedback, namely CsiMamba-RDC. The framework employs improved residual vector quantization to recursively refine CSI representation, reducing information loss and storage overhead. Additionally, we present an encoder-decoder model leveraging a selective state space model to extract diverse channel features.

Deep learning (DL) techniques have been extensively utilized to tackle challenges in the field of wireless communication, overcoming the limitations of traditional methods. However, training DL algorithms often requires large amounts of data, which is difficult to obtain in increasingly complex communication environments. Reducing the amount of data required for DL training is therefore an urgent problem to be solved. In this work, we develop a denoising diffusion probabilistic model (DDPM)-based digital twin (DT) framework of integrated sensing and communication (ISAC) multiple-input-multiple-output (MIMO) channel to address the data scarcity issue commonly found in DL-based scenarios. By sampling a small amount of data, our framework captures and simulates the data distribution, building a virtual data repository that can continuously provide samples to assist in executing control instructions to physical entities, even as the user equipment (UE) and target positions change. Specifically, we formulate the data generation problem as a distribution approximation task guided by the Kullback-Leibler (KL) divergence criterion and optimize it by meticulously designing a DDPM network composed of U-Net structure, time-embedding modules, and attention mechanisms. Moreover, we enhance the framework by formulating a task-driven objective function for two applications: 1) sensing channel estimation and 2) target detection. Numerical results demonstrate the superiority of our proposed DDPM-based DT framework compared with other data augmentation techniques in improving the performance of data-driven DL-based tasks, showcasing its robustness across diverse scenarios.

When large language models are trained on private data, it can be a \textit{significant} privacy risk for them to memorize and regurgitate sensitive information. In this work, we propose a new \emph{practical} data extraction attack that we call ``neural phishing''. This attack enables an adversary to target and extract sensitive or personally identifiable information (PII), e.g., credit card numbers, from a model trained on user data with upwards of $10\%$ attack success rates, at times, as high as $50\%$. Our attack assumes only that an adversary can insert as few as $10$s of benign-appearing sentences into the training dataset using only vague priors on the structure of the user data.

Network slicing-based communication systems can dynamically and efficiently allocate resources for diversified services. However, due to the limitation of the network interface on channel access and the complexity of the resource allocation, it is challenging to achieve an acceptable solution in the practical system without precise prior knowledge of the dynamics probability model of the service requests. Existing work attempts to solve this problem using deep reinforcement learning (DRL). However, such methods usually require a lot of interaction with the real environment to achieve good results. In this paper, a framework consisting of a digital twin and reinforcement learning agents is present to handle the issue. Specifically, we propose to use the historical data and the neural networks to build a digital twin model to simulate the state variation law of the real environment. Then, we use the data generated by the network slicing environment to calibrate the digital twin so that it is in sync with the real environment. Finally, DRL for slice optimization optimizes its performance in this virtual pre-verification environment. We conducted an exhaustive verification of the proposed digital twin framework to confirm its scalability. Specifically, we propose to use loss landscapes to visualize the generalization of DRL solutions. We explore a distillation-based optimization scheme for lightweight slicing strategies. In addition, we also extend the framework to offline reinforcement learning, where solutions can be used to obtain intelligent decisions based solely on historical data. Numerical simulation experiments show that the proposed digital twin can significantly improve the performance of the slice optimization strategy.

Ideally, the bi-directional channel in time division duplex (TDD) millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) systems exhibits reciprocity. However, the involvement of low-cost and non-ideal radio frequency (RF) chains disrupts this reciprocity. Consequently, prior to fully leveraging the advantage of channel reciprocity, it is essential to implement channel calibration. Despite numerous over-the-air calibration methods, such as Argos, the typical least square (LS) are proposed in the literature, none of their criteria directly focus on the calibration performance. To address this gap, we propose a novel deep learning based approach that aims to optimize the calibration performance and introduce device-level intelligence towards 6G networks. To be specific, two cascaded modules are designed in a model-assisted end-to-end manner. Firstly, we propose the double-CNN-based channel denoising module for joint bi-directional channel estimation by exploiting the characteristics of mmWave channel. Secondly, the deep calibration learning module is meticulously designed to obtain the calibration coefficients with the aid of assisted model. This traceable assisted model is established by leveraging the expert knowledge of calibration process, based on which the MetrNet and the CaliNet are designed. Numerical results demonstrate the superior performance of our proposed method compared to existing calibration methods. Particularly, additional simulations and analysis are conducted to verify the effectiveness of the two properly designed modules.

In this paper, the cross-layer optimization problem of access point selection (APS) and beamforming (BF) in cell-free network (CFN) with local CSI is studied, where constraints of per AP power and the number of active APs are considered. Such a joint APS&BF optimization problem is modeled as a mixed-integer nonlinear programming (MINP) problem aiming at maximizing the sum rate of the whole system. Fractional programming (FP)-based and alternating optimization (AO)-based algorithms with weightedl1-norm approximation are proposed to solve this MINP problem. However, the latter performs better than the former, with higher complexity. A lightweight multi-head single-body graph neural network (MHSB-GNN) algorithm is proposed, where the nodes and structures are innovatively designed. The MHSB-GNN benefits from the different node updating modules for different user equipment (UE), which introduce extra prior information into the graph and mine specific information of different UEs. Moreover, the equivalence between GNN and FP-based algorithm is proved to provide interpretability and theoretical guarantees for MHSB-GNN. The analysis of convergence and complexity validates the accuracy and effectiveness of the FP and AO-based algorithms. Leveraging the existing APS and BF solver, it is shown that the three proposed algorithms guarantee comparable performance as the exhaustive search algorithm in performance and complexity.

In federated learning (FL), local workers learn a global model collaboratively using their local data by communicating trained models to a central server for privacy concerns. Due to its local nature, FL is typically subject to various heterogeneities, including system and statistical heterogeneity. To address these concerns, Federated Proximal (FedProx) has been considered a promising FL paradigm to provide more stable learning convergence in the presence of computation stragglers and statistical heterogeneity. However, in wireless networks with unreliable communication channels, the errors of packet transmissions should be considered, introducing additional heterogeneity. For the first time, we rigorously prove the convergence of FedProx in the presence of transmission packet errors in heterogeneous networks. In addition, we propose a joint client selection and resource allocation strategy that maximizes the number of effective participating users for convergence acceleration. The method is combined with a random weight mechanism to reduce the statistical bias caused by the client selection strategy. An efficient low-complexity algorithm for solving the optimization problem is developed. The proposed method achieves faster convergence and requires fewer communication rounds to attain accuracy than existing state-of-the-art client selection methods.

From practical considerations in massive multiple-input multiple-output (MIMO) systems, with the involvement of radio frequency (RF) chains, the channel reciprocity no longer holds even under time division duplex (TDD) operation. To fully leverage the advantage brought by TDD systems, channel reciprocity calibration needs to be necessarily investigated. In this paper, we propose the CNN-enhanced calibration method, which is composed of the channel estimation task and the calibration coefficient calculation task. Different from previous works, our method is based on our proposed double-CNN-based bi-directional channel estimator, which is designed specifically for the calibration problem to exploit the bi-directional channel correlation, the spatial correlation, and the angular correlation in millimeter wave (mmWave) channel. Based on this, a formulated LS calibration problem is solved. Numerical results manifest that our proposed method outperforms the existing calibration methods in the literatures.

Accurate channel estimation is an indispensable prerequisite for intelligent reflecting surface (IRS) aided communication systems to achieve huge system performance gains. Current works show that deep neural network-based channel estimation is a promising solution to achieve competitive performance compared with the conventional methods. However, neural network-based approaches generally realize the channel estimation by manually designing network architectures in a trial-and-error manner which need complex neural network domain knowledge and tremendous computation resource. This paper automatically constructs a high-performance neural network architecture to obtain dedicated channel estimation schemes intelligently. Specifically, we propose a channel estimation neural network architecture search (CENAS) method based on gradient alternatively search strategy to search a channel estimation neural network. With the search space designed meticulously for the channel estimation task, the network searched by the proposed method outperforms the conventional and deep learning-based channel estimation algorithms.

Large antenna arrays and beamforming are necessary for the mmWave communication system, resulting in heavy time and energy consumption in the beam training stage. Therefore, dual-band operations are expected to be deployed in future communication systems, where low-frequency channels are used to meet basic communication needs, and millimeter wave (mmWave) channels are exploited when the high-rate transmission is required. Existing works utilize deep learning methods to extract low-frequency channel state information (CSI) to reduce the mmWave beam training overheads. However, an important limitation of deep learning approaches is that the model is usually trained in a given environment. When employed in an unseen environment, it usually requires a large amount of data to retrain. In this paper, a model-agnostic optimization algorithm based on meta-learning is proposed to provide a general mmWave beam prediction model. This model can be deployed to edge base stations and effectively adapted to the environment without the need for a heavy collection of data. Simulation results demonstrate that the proposed approach could reduce the model adaptation overheads. The meta-learning-based beam prediction model is robust and achieves high prediction accuracy and spectral efficiency in different signal-to-noise ratio (SNR) regimes.

Deep neural network-based learning methods have been considered promising techniques used in beam selection problems. However, existing research ignores the peculiar vulnerabilities of neural networks. The adversaries can use data poisoning to embed predefined triggers into a model during training time such that the neural network-based beam model may make an incorrect output decision of a test example when patched with the trigger. Data poisoning offers attackers the possibility to build backdoors. The goal of backdoors is often unethical, such as giving users a poor experience by manipulating infected models to output inappropriate beams. In this paper, first, we introduce a simple backdoor attack method by using data poisoning in a mmWave beam selection system. By numerical simulations, we verify that this poisoning attack is effective for neural networks with different structures. In addition, we explore the effect of poisoned data volume on the effect of backdoor attacks. The results show that the backdoor can be successfully implanted into the beam selection neural network. Besides, we fine-tune the trained model for a new wireless communication environment, and the results show that backdoors still exist even when the model is tuned with data from new scenarios. Then, we propose a machine unlearning solution to mitigate the backdoor of the trained beam selection model. The problem of eliminating backdoors is modeled as a minimax optimization problem. We propose a novel adversarial unlearning method along with label smoothing to solve the backdoor removal problem. We compared the proposed backdoor elimination method with the classical fine-tuning elimination method and the neural network pruning method through numerical simulations. The results show that the fine-tuning and the pruning methods cannot effectively remove the backdoor. The proposed machine unlearning method can make the trained model forget about the backdoor under the condition that the performance of the benign task (beam selection tasks when the trigger does not appear) is guaranteed to be slightly degraded. In summary, our work illustrates that data poisoning-based backdoor attacks may exist in wireless networks, and we propose a scheme to eliminate backdoors.

Deep learning (DL) is an emerging paradigm for accurate channel estimation for intelligent reflecting surface (IRS)-aided wireless communication systems. It has been proven to be a promising way to achieve better channel estimation performance for the IRS-aided wireless communication system than traditional methods (e.g., least-square algorithm). However, existing DL-based methods rely on ground truth (labels of the true channels) which is difficult to obtain in real networks. In this paper, we propose a self-supervised learning (SSL) method for the IRS channel estimation problem. No ground truth channel is needed in the training, while a simple and novel self-supervised denoising formula without a clean reference signal is presented. Particularly, in the training phase, the self-supervised signal and the input are the received signal vector and its noisy version, respectively. While in the inference phase the input is the estimated channel by using the least-square method and the output is the refined channel estimation. That is, our neural network-based channel estimation algorithm is not reciprocal for training and testing. We demonstrate that the proposed SSL solution has good convergence performance and generalization ability through numerical simulations. Interestingly, we find a “double descent” phenomenon in the learning curve during the test phase, i.e., when we gradually increase the number of training epochs, the performance first gets better, then becomes worse, and further gets better again. Besides, we propose to analyze SSL using the loss landscape and centered kernel alignment method. The results show that the self-supervised model has a similar loss landscape and representational similarity to the supervised model. We explored the effects of different signal-to-noise ratios (SNRs), different neural network sizes, and different training data volumes on our algorithm through numerical simulations. Extensive numerical simulation results show that our SSL algorithm is still competitive without ground truth. We also show that the developed scheme exhibits robustness to SNR ratio mismatch.

Federated learning (FL) systems have an inherent vulnerability to adversarial backdoor attacks during training due to their decentralized nature. The goal of the attacker is to implant backdoors in the learned model with poisoned updates such that at test time, the model’s outputs can be fixed to a given target for certain inputs (e.g., if a user types “people from New York” into a mobile keyboard app that uses a backdoored next word prediction model, the model will autocomplete their sentence to “people in New York are rude”). Prior work has shown that backdoors can be inserted in FL, but these backdoors are not durable: they do not remain in the model after the attacker stops uploading poisoned updates because training continues, and in production FL systems an inserted backdoor may not survive until deployment. We propose Neurotoxin, a simple one-line backdoor attack that functions by attacking parameters that are changed less in magnitude during training. We conduct an exhaustive evaluation across ten natural language processing and computer vision tasks and find that we can double the durability of state of the art backdoors by adding a single line with Neurotoxin.

Mobile edge computing (MEC) based heterogeneous vehicular networks (HetVNets) can interwork between IEEE 802.11p-based vehicular networks and cellular-assisted vehicular networks for vehicle-to-everything (V2X) communications. It is an attractive technology for supporting low latency applications for vehicles. However, in the practical system without precise prior knowledge of the dynamic wireless environment, solving joint vehicle-to-edge (V2E) association and resource allocation problem is a challenge. In this paper, first, we use stochastic geometry to model a real scenario. Specifically, the intersection area is modeled as two perpendicular streets, the spatial distribution of vehicle nodes on each street is modeled as an independent one-dimensional (1D) homogeneous Poisson Point Process (PPP), the spatial distribution of different types of edge nodes is modeled as different and independent PPPs. We consider the service time during which a vehicle node with different types of network interfaces gets a service from an edge node. Then, a deep reinforcement learning (DRL) based method is proposed to solve the uplink-and-downlink V2E association problem minimizing the service time while ensuring the computation resource allocation constraints. Simulation results illustrate the better performance of our solution than that of other traditional methods.

Learning-based resource allocation (LRA) is envisioned as an integral element of 6G. This article proposes a novel learning-based paradigm to address resource allocation problems in heterogeneous ultradense networks (HUDNs). Our paradigm is a highly efficient realization for utilizing the inherence properties in HUDN, which comprise the local validity and correlation attenuation. Concretely, we formulate the HUDNs as a heterogeneous bipartite graph model and propose the corresponding heterogeneous bipartite graph neural network (HBGNN). The local sampling characteristic of HBGNN matches the inherence properties. Meanwhile, our paradigm combines data-driven and model-driven learnings and employs online and offline trainings. Hence, two of LRA’s obstacles: 1) the overreliance on the perfect data set and 2) the low calculation efficiency are mitigated, and the realizability of our paradigm is improved. Besides, entropy regularization is utilized to guarantee the effectiveness of exploration in the configuration space. We apply our approach to a representative resource allocation problem, the jointly user association (UA) and power allocation (JUAPA) problem. We formulate JUAPA as a combination classification and regression problem and adopt a dynamical hyperparameter output layer to address the discrete variable of UA. Simulation results demonstrate that the proposed method has better performance and higher computational efficiency than traditional optimization algorithms.

Federated learning (FL) is an emerging paradigm for distributed machine learning that uses the data and the computational power of user devices while maintaining user privacy (e.g., position and motion track). It has been proved a promising way to help the learning-based millimeter wave (mmWave) system achieve efficient link configuration. However, FL systems have an inherent vulnerability to backdoor attacks during training, and this has not received attention in current FL-based beam selection research. The goal of a backdoor attacker is to implant a backdoor in the model such that at test time, the model will mispredict a certain family of inputs, and corrupt the performance of the trained model on specific sub-tasks. We study backdoor attacks in an FL-based beam selection system based on a deep neural network that utilizes user location information. Specifically, we propose a backdoor attack scheme that can be configured in the real world. The attacker’s trigger is an obstacle placed in certain locations. When the model encounters an input with these obstacles, the backdoor will be triggered, and the model will output the beam specified by the attacker. Through experiments, we show that the proposed attack can achieve a high attack success rate in a system without a defense mechanism. Moreover, we show that the traditional norm-clipping defense method cannot effectively defend against our attack. Furthermore, we propose a new backdoor attack defense method and verify the effectiveness of this scheme through experiments. In addition, we propose a backdoor detection method: the federated noise titration method, which can diagnose whether the model has a backdoor. Overall, our work explored backdoor attacks, defenses, and detection of the FL-based mmWave beam selection system.

Federated learning (FL) is a promising way to use the computing power of mobile devices while maintaining the privacy of users. Current work in FL, however, makes the unrealistic assumption that the users have ground-truth labels on their devices, while also assuming that the server has neither data nor labels. In this work, we consider the more realistic scenario where the users have only unlabeled data, while the server has some labeled data, and where the amount of labeled data is smaller than the amount of unlabeled data. We call this learning problem semi-supervised federated learning (SSFL). For SSFL, we demonstrate that a critical issue that affects the test accuracy is the large gradient diversity of the models from different users. Based on this, we investigate several design choices. First, we find that the so-called consistency regularization loss (CRL), which is widely used in semi-supervised learning, performs reasonably well but has large gradient diversity. Second, we find that Batch Normalization (BN) increases gradient diversity. Replacing BN with the recently-proposed Group Normalization (GN) can reduce gradient diversity and improve test accuracy. Third, we show that CRL combined with GN still has a large gradient diversity when the number of users is large. Based on these results, we propose a novel grouping-based model averaging method to replace the FedAvg averaging method. Overall, our grouping-based averaging, combined with GN and CRL, achieves better test accuracy than not just a contemporary paper on SSFL in the same settings (>10%), but also four supervised FL algorithms.

Image-level weakly supervised semantic segmentation is a challenging task. As classification networks tend to capture notable object features and are insensitive to over-activation, class activation map (CAM) is too sparse and rough to guide segmentation network training. Inspired by the fact that erasing distinguishing features force networks to collect new ones from non-discriminative object regions, we using relationships between CAMs to propose a novel weakly supervised method. In this work, we apply these features, learned from erased images, as segmentation super-vision, driving network to study robust representation. In specifically, object regions obtained by CAM techniques are erased on images firstly. To provide other regions with seg-mentation supervision, Erased CAM Supervision Net (ECS-Net) generates pixel-level labels by predicting segmentation results of those processed images. We also design the rule of suppressing noise to select reliable labels. Our experiments on PASCAL VOC 2012 dataset show that without data annotations except for ground truth image-level labels, our ECS-Net achieves 67.6% mIoU on test set and 66.6% mIoU on val set, outperforming previous state-of-the-art methods.

Proposed by Maddah-Ali and Niesen, a coded caching scheme has been verified to alleviate the load of networks efficiently. Recently, a new technique called placement delivery array (PDA) was proposed to characterize the coded caching scheme. In this paper, we consider a caching system in the scope of ultra dense networks (UDNs). Each base station (BS) has a finite cache and stores some contents. We propose an efficient coded content caching scheme called double coded caching to make the transmission robust to in-and-out wireless network quality. Then the dynamic caching and multicast scheduling are considered to jointly minimize the average delay and power of the content-centric wireless networks. This stochastic optimization problem can be formulated as a Markov decision process (MDP) with unknown transition probabilities and large state space. We propose a deep reinforcement learning approach to deal with the decision problem. Our algorithm uses a variational auto-encoder (VAE) neural network to approximate the state sufficiently, and uses a weighted double Q-learning scheme to reduce variance and overestimation of the Q function. Numerical results demonstrate that the proposed double coded caching scheme increases the probability of the successful transmission, and the caching and scheduling policy can effectively reduce the delay and the power consumption.

Coded caching is a promising approache to support low latency transmission over broadcast wireless networks. The process of selecting the nodes that forward coded messages can be considered as a set cover problem. However, existing research efforts don't focuse on solving the set cover problem. This paper investigates the problem of the cover problem for coded caching in wireless networks. First, we propose a novel coded caching method using deep neural networks. Then, we establish a mathematical model for cover problem of the coded caching system. Then, we propose a deep learning approach to solve it. Different from previous works, our proposed deep neural architecture uses sequence-to- sequence model to learn the solutions. Finally, numerical results are given to demonstrate the proposed coded caching method have lower load than traditional coded caching method, and that proposed method for solving the cover problem can effectively implement coded caching with lower computational complexity.

This study investigates a novel unmanned aerial vehicle (UAV)‐based wireless sensor network, where the UAV acts as a flying base station to serve multiple wireless sensor nodes (SNs). The authors goal is to maximise the system energy efficiency of the UAV while satisfying the fairness among SNs by jointly optimising the UAV trajectory and UAV time allocation. The formulated problem is shown to be a non‐convex fractional optimisation problem, which is hard to tackle. To this end, they decompose the original problem into two sub‐problems, and the block coordinate descent method and successive convex optimisation technique are employed to solve these two sub‐problems iteratively. Specifically, in the first sub‐problem, the optimal UAV time allocation is obtained by maximising the minimum achievable rate of SNs with given UAV trajectory constraints. In the second sub‐problem, the UAV trajectory is achieved by minimising the energy consumption of the UAV with the given UAV time allocation. Subsequently, an iterative algorithm is proposed to optimise the time allocation and UAV trajectory alternately. Furthermore, the convergence and complexity of their proposed algorithm are provided. Numerical results show that the proposed scheme outperforms the existing benchmark strategies in terms of energy efficiency.

This paper investigates the problem of proactive caching for multi-view 3D videos in the fifth generation (5G) networks. We establish a mathematical model for this problem, and point out that it is difficult to solve the problem with traditional dynamic programming, then we propose a deep reinforcement learning approach to solve it. First, we model the proactive caching system for multi-view 3D videos as a Markov decision process jointing views selection and local memory allocation. Then, we present an actor-critic, model-free algorithm based on the deep deterministic policy gradient to find effective proactive caching policy. Since the action space is affected by the system state, we embed dynamic k-Nearest Neighbor algorithm into actor-critic algorithm to implement the deep reinforcement learning algorithm working in an action space of variable size. Finally, the numerical results are given to demonstrate that the proposed solution can effectively maintain high-quality user experience for high-mobility 5G users moving among small cells. We also investigate the impact of configuration of critical parameters on the performance of the algorithm.

In this paper, we employ the unmanned aerial vehicle (UAV) as a flying base station (BS) to offload the data computing tasks from mobile terminal (MT) for saving mobile energy consumption. Our goal is to minimize consumption of the computational tasks at MT by jointly designing the resource partitioning scheme and bit allocation strategy. Specifically, the portion of total bits for local computation at MT is optimized, and the other portion of bits is computed by jointly optimizing the number of bits transmitted in the uplink, the number of bits computed locally at UAV and the number of bits transmitted in the downlink. The formulated problem has been shown in a convex form, which has optimal solutions. Instead of solving original problem using standard convex optimization techniques, we propose a resource partitioning scheme and bit allocation strategy based on dual decomposition, which has been shown in a low computational complexity. Furthermore, the numerical results are provided to demonstrate the superiority of our proposed scheme over the compared benchmarks.