摘要:Deterministic networking was proposed to meet the demands of high bandwidth, low delay and high reliability for audio and video, industrial control, power protection and other applications.With the development of new services such as virtual/augmented reality and cloud games, the application scenarios of deterministic networking have been enriched, and the scope of deterministic networking technology has also been extended.In view of the overall consideration of current application scenarios, technical standards and industrial development, the strong and weak deterministic requirements and characteristics for different scenarios were analyzed, as well as different kinds of deterministic networking technologies, including scientific and engineering.The diversified development path of deterministic networking was proposed and the development of combination technologies of deterministic networking for 6G and computing force network was looked forward.
关键词:deterministic networking;scientific and engineering;combination technology
摘要:As the Internet evolves from a consumer network to a productive network, emerging industry applications necessitate differentiated and deterministic service quality assurance and communication capabilities that span all dimensions of time and space.How to realize the heterogeneous compatibility of ubiquitous multi-domain networks and support deterministic transmission between nodes in space-air-ground integrated network (SAGIN) has become a significant research problem.Firstly, the emerging full-domain 6G scenarios along with their requirements for deterministic services were analyzed.Then, the hierarchical space-air-ground integrated deterministic networking architecture, including global cooperative network management and control layer, multi-domain dynamic deterministic fusion layer, and ubiquitous heterogeneous deterministic networking layer, was proposed, and key technologies for fixed-mobile-satellite fusion networks in relevant scenarios were studied.Finally, the challenges and development trend of the space-air-ground integrated deterministic networking were presented.
摘要:The efficient allocation of data elements is particularly important for achieving digital transformation in traditional industries.Network supply capabilities not only require low latency and high bandwidth for data packets, but also require low jitter and certainty in data packets.Deterministic networking technology is the key to build lossless, reliable, and deterministic wide-area communication capabilities.The key control technologies and design concepts for wide-area deterministic networking were proposed, and the specific plans for achieving deterministic networking perception, service planning, and traffic scheduling were elaborated on.Through experimental verification and application testing, the effectiveness of the system structure is proven and it meets the management and scheduling needs of deterministic networking.
关键词:deterministic networking;networking control;software define;cycle queuing and forwarding
摘要:With the rapid development of AIGC large models and intelligent computing applications, deterministic networking technology for wide-area RDMA has become a recent research hotspot.The development path, implementation solutions and technical challenges of RDMA-WAN were analyzed, by setting up a laboratory and an outfield testing environment, with a series of testing and verifications.The deterministic demand characteristics and boundary performance indicators of RDMA-WAN were proposed.Finally, the deterministic networking technologies for RDMA-WAN were summarized and prospected.
摘要:Many network applications (e.g., industrial control) require networks to provide deterministic transmission services, and traditional statistical multiplexing networks can not meet the needs of this type of applications.IEEE 802.1 TSN standard and related research can provide deterministic networking services over Ethernet.The basic principles of TSN were described briefly, the key queuing mechanisms of TSN by using network calculus were analyzed quantitatively, and an auto-learning TSN mechanism was proposed to adaptively support non-TSN device connections by network intelligence.Together with delay guarantee algorithms, a E2E TSN solution can be formed and used for the further promotion and application of TSN technology.
摘要:TSN and DetNet are major components of the deterministic networking solution, detailed research has been conducted by industry and academia.However, the study seldom considers the joint of these two technologies under large-scale networks.To facilitate the analysis of the joint networks, TSN and DetNet specifific extensions to the popular OMNeT++/INET framework for network simulations was presented, in particular, the time slot scheduling, the TSN node/host/interface and DetNet node/host/interface, the framework for collaboration between the controller and the simulation system.A proof of concept implementation and exemplary evaluation of the joint of these two technologies under large-scale networks was presented.The results show that in large-scale cross-domain scenarios, the time slot scheduling mechanism can provide end-to-end deterministic service guarantee.
关键词:OMNeT++;large-scale DetNet;forwarding and control collaboration;slot scheduling;cross-domain
摘要:To address the problems of poor robustness and high complexity of current physical layer user authentication algorithms, a lightweight convolutional neural network (CNN) channel feature extraction algorithm was proposed to reduce the channel state response required for training by changing the form of network input, and a multi-user physical layer channel feature-assisted authentication mechanism was established based on this algorithm to design a detailed process from user registration to authentication, and multi-user authentication and network parameter update online were completed.Simulation results show that the proposed algorithm can complete multi-user authentication, obtain good detection performance with smaller training rounds, and require fewer training samples than existing multi-user authentication algorithms.
摘要:A phase tracking algorithm was proposed for massivemultiple-input multiple-output(MIMO) systems based on the modified Newton (MN) method, which effectively reduced the high computational complexity in traditional high-performance hybrid precoding schemes.The algorithm optimized the analog precoding matrix from the perspective ofsub-dimensional vector recovery.In each sub-dimension optimization, the phase tracking method was used to transform the recovery of the analog precoding vectors into an unconstrained nonlinear optimization problem, which was then solved using the MN method.Concurrently, this strategy led to a marked reduction in the computational intricacy pertaining to both the computation of correction factors and the inversion of the Hessian matrix within the framework of the MN method.This was achieved through the insightful incorporation of Gerschgorin’s Disk theorem and the Hermitian matrix block-inverse lemma.Simulation results show that the proposed algorithm has higher spectral efficiency and lower computational complexity than several conventional high-performance hybrid precoding schemes.
关键词:massive MIMO;hybrid precoding;phase tracking;modified Newton method
摘要:Due to its powerful representation learning capabilities and efficient computing capabilities, deep learning-based hashing (deep hashing) methods are widely used in large-scale image retrieval.However, there are less studies on the security of deep hashing models.A dual-branch autoencoder network (DBAE) to study targeted attacks on such retrieval was proposed.The main goal of DBAE was to generate imperceptible adversarial samples as query images in order to make the images retrieved by the deep hashing model semantically irrelevant to the original image and relevant to the target image.Numerous experiments demonstrate that DBAE can successfully generate adversarial samples with small perturbations to mislead deep hashing models, and italso verifies the transferability of these perturbations under various settings.
摘要:In response to the low accuracy of synthetic speech spoofing detection based on traditional acoustic feature parameters, poor detection performance for unknown types of synthetic speech, and performance degradation in noisy environments, a method for detecting spoofing synthetic speech was proposed using constant Q modulation envelope (CQME) .The motivation of the method was from the fact that the temporal envelope of speech contained abundant information and there was a big difference in detail between the envelope of synthetic speech and genuine speech.The modulation envelope spectrum of speech was obtained by employing constant Q transform (CQT), and the root mean square of each frequency component was calculated to derive the CQME feature vector.And then the CQME feature vector was used to train the random forest classifier for discriminating genuine speech from spoofing synthetic speech.Experimental results demonstrate that the random forest trained with CQME features achieves high detection performance on the ASVspoof 2019 dataset and exhibites good detection efficacy for unknown types of synthetic speech.Furthermore, the proposed method shows high detection performance even under various noise conditions, having excellent noise robustness.
摘要:To address the problems of low retrieval efficiency of searchable encryption scheme under traditional public key cryptosystem, high storage overhead of keyword index, and index does not support dynamic update, based on the good space and time efficiency of counting Bloom filter to determine whether an element belongs to a certain set, the feature into searchable encryption was applied to establish keyword index, thus, the keyword set was mapped into a binary vector, thereby reducing the keyword storage cost.Simultaneously, the feature of counting Bloom filter can be deleted was used to realize the dynamic update of keyword index.In addition, to prevent adversaries from inferring keyword plaintext information from binary vectors through statistical analysis, the binary vector was blinded by permutation, and then the permutation was encrypted by attribute-based encryption for secret sharing.After legitimate users generated keyword traps correctly, the service provider helped users complete keyword matching queries based on the user’s keyword trapdoor.Finally, the security and efficiency of the scheme were proved through security analysis and simulation experiments.
摘要:Massive multiple-input multiple-output (MIMO) system can provide satisfying gain of spectrum efficiency for 5G and future wireless communication systems.In frequency-division duplex (FDD) mode, downlink channel state information (CSI) needs to be accurately fed back to the base station side to obtain this gain.To improve the feedback accuracy of downlink CSI eigenvector, a self-attention mechanism-based CSI feedback method named SA-CsiNet was proposed.SA-CsiNet respectively deployed self-attention modules at the encoder and the decoder to achieve feature extraction and reconstruction of CSI.Experimental results show that compared with codebook-based and conventional deep learning-based CSI feedback approaches, SA-CsiNet provides higher reconstruction accuracy of CSI.
摘要:Location service features are enabled in 5G to allow new and innovative location-based services to be developed for kinds of scenarios and the key performance requirements are identified by 3GPP.Firstly, the planning of location service feature in each standard release of 5G was introduced based on the standard process of 5G high-accuracy positioning in 3GPP.Secondly, 5G network-assisted BeiDou navigation satellite system was studied, including the positioning technologies and the standard progress of A-BDS in 3GPP.Thirdly, the network architecture, key technologies and the standards progress for sidelink positioning in 5G were introduced.Finally, the evolution of 6G positioning was analyzed.
摘要:To mitigate the impact of tropospheric radio-duct interference on 5G network, 3GPP has defined functionalities for remote interference management.Based on this functionality, an automated scheme for SetID and gNBID was developed, enabling the reuse of SetID in different tropospheric radio-duct interference regions, striking a balance between efficiency and uniqueness.By analyzing remote interference measurement data from base stations, an intelligent localization and identification method for tropospheric radio-duct interference was summarized.An intelligent network optimization process for identifying the aggressor base stations was created and successfully validated in the network.An effective operational plan was provided for management of 5G tropospheric radio-duct interference, alleviating the constraints on the development of the 5G industry caused by the interference.
摘要:MOS (mean opinion score) is usually used to evaluate voice quality in the industry.It can objectively and fairly reflect the user’s voice service perception.It is difficult and costly to obtain data by road test, so a trained supervised learning model is usually used to predict the MOS score.However, the operator voice data has the characteristics of low percentage of MOS low score data and time sequence change, which affects the accuracy and generalization of the model prediction.Based on the study of existing data acquisition systems and machine learning algorithms of operators, an adaptive algorithm for MOS evaluation of 5G speech quality was proposed.Firstly, POLQA algorithm test equipment based on full parameter evaluation obtained training data to ensure the accuracy of training samples.Secondly, by means of data enhancement, the difficulty of acquiring poor quality samples was solved.Finally, based on the adaptive algorithm selection, the optimal MOS prediction model could be selected periodically and dynamically according to the timing changes of data features, so as to achieve large-scale and intelligent evaluation of 5G voice quality.
摘要:In view of the common problems in power supply stations such as multiple systems, poor App experience, non linkage, and difficult data acquisition, zero code technology and robotic process automation (RPA) technology were researched, terminal service integration digital engine was built, providing zero code construction capability, order management capability, PC and mobile terminals collaboration capability, which was suitable for various disciplines such as electricity bills, metering, material, so as to realize the online, automated and mobile common service, reduce the burden and increase efficiency for the employees.
关键词:digital power supply station;terminal service integration;zero code technology;RPA technology;digital transformation