Full-duplex (FD) integrated access and backhaul (IAB) system requires effective self-interference (SI) suppression, relying on accurate channel estimation with high pilot overheads. This paper proposes a cost-effective end-to-end (E2E) hybrid beamforming framework for FD enabled millimeter-wave (mmWave) IAB systems with implicit channel state information (CSI), which unifies the pilot training and data communication to avoid accurate channel estimation. We introduce a Sensing Transformer block to sense spatial channels of the network, where sensing vectors are designed for pilot observation. Specifically, the Sensing Transformer is designed to implement an attention mechanism to effectively leverage implicit CSI from sequential pilot observations with an arbitrary length. Thereafter, we combine graph neural networks and Transformer mechanisms as a novel Graphformer algorithm to design hybrid beamformers with high spectrum efficiency, which can efficiently mitigate multiuser interference and SI via capturing both local and global spatial structure features of the whole network. Numerical results verify that: 1) The proposed E2E framework can significantly improve the spectrum efficiency with lower pilot overheads compared with the conventional method with explicit channel estimation. 2) The proposed E2E framework exhibits superior channel sensing and interference cancellation performance over traditional learning-based schemes across various network configurations, even with modest pilot overheads.
IEEE 802.11ax introduces orthogonal fre-quency division multiple access(OFDMA)technology to achieve parallel access for multiple users.However,when a large number of users access,the probability of collision will be greatly increased,resulting in greatly reduced throughput.In high-density wireless networks,in order to reduce collisions between users,a hybrid contention/designation-based uplink OFDMA random access(HCDRA)scheme is proposed.The proposed scheme adopts a hybrid of contention and transmission of designation,where the access point(AP)arranges one of the collided stations(STA)on the resource unit(RU)for transmission of designation in the next trans-mission,and other STAs still follow the traditional con-tention mechanism.The addition of designated trans-mission scheme can effectively reduce the intensity of contention and improve the utilization rate of RU without knowing the prior STA and without increas-ing the overhead of the sensing time.The proposed scheme can greatly improve throughput and latency performance without introducing additional overhead.Moreover,our proposed hybrid scheme has almost no changes to the IEEE 802.11ax standard protocol,and is suitable for IEEE 802.11ax and beyond.The feasibility of the proposed scheme has been demonstrated through numerical simulation results.
Based on current research, there is a lack of comprehensive review articles that systematically address the security issues of the internet of vehicles (IoV) using blockchain technology. In this study, we thoroughly analyze the threats and challenges within IoV systems and provide a systematic review of blockchain-based IoV security solutions. This paper summarizes blockchain solutions for addressing key challenges in IoV, including network security, communication efficiency, resource optimization, data management, and transaction operational efficiency. Our findings indicate that blockchain technology can enhance V2X communication security, optimize resource utilization, ensure data integrity, and improve transaction security. Additionally, this paper explores future research directions, including advanced security mechanisms, efficient consensus algorithms, the integration of edge computing, and intelligent applications, demonstrating the potential of blockchain in tackling IoV system challenges.
A covert communication approach in time-frequency domain is presented. By deliberately embedding covert signal into the primary transmitted signal, the covert signal is completely hidden, rendering conventional signal detection methods ineffective. However, the synchrosqueezing transform (SST) maps the signal into the time-frequency domain, resulting in improved time-frequency localization and more effective separation of its components. Based on the proposed instantaneous frequency (IF) curve extraction and filtered reconstruction scheme, the waveform of the covert signal can be automatically extracted. Experimental results from synthetic signal testing validate the efficiency of the proposed method, demonstrating its potential for covert communication applications.
In the process of minimizing the training loss of machine learning model pursuit,it is very easy to inadvertently remember sensitive private data,which leads to data reconstruction,member reasoning attacks and other security problems.In order to mitigate these risks,differentiated privacy has become a key standard for privacy preserving machine learning.The classical differential privacy depth learning algorithm,differen-tial privacy stochastic gradient descent(DP-SGD),that adapts the standard stochastic gradient descent(SGD)algorithm to incorporate differential privacy,ensuring that the trained model doesn't reveal sensitive infor-mation about individual training data points.However,DP-SGD has the problems of slow convergence speed and large utility loss.We propose an effective solution that is the cooperative combination of selective updat-ing and early vertical gradient disturbance.Selective updating ensures that the training track of the model is aligned with the optimal direction,significantly ac-celerating convergence.Subsequently,the application of vertical gradient perturbation ensures that the model with significantly improved accuracy can be achieved even under strict privacy constraints(small privacy bud-get).Through theoretical analysis and a large number of experiments,this paper proves that DP-VGPSU has superior performance in convergence speed and accu-racy.
