最新刊期
卷 6 ， 期 4 ， 2022
- 摘要：Service platforms centered on the Internet of things and mobile Internet are in accelerating process.Hundreds of millions of end-users communicate through network real-time video services, which have become an irreplaceable core tool in human’s digital life.However, the Internet is becoming dynamic, and heterogeneous, which imposes stringent requirements on real-time video streaming control technology.Moreover, the QoE of real-time video is not satisfactory.An adaptive reinforcement learning-based video intelligent transmission algorithm was designed, which can deal with heterogeneous network environment.And then, an effective end-to-end on-device training framework was designed to decrease server overhead, and a detailed evaluation and analysis on the neural network design and structure was provided.Experimental results show that the proposed algorithm can effectively predict heterogeneous network bandwidth, and reduces the bandwidth prediction error by 48.48%, comparing with the representative streaming control algorithm.The effective bandwidth prediction can further improve the user QoE, such as improving the video fluency by 60.65%, and improving the video quality by 16.52%.Besides, the analysis can provide empirical insights for further study, and holds potential to push the development of intelligent video applications.关键词：real-time video;adaptive streaming control;quality-of-experience;reinforcement learning;on-device training764|594|0更新时间：2024-08-16
- 摘要：To address the low spectrum utilization issue of the distributed network due to multi-user collision and channel time-varying nature, the distributed channel access problem for multicast communication was investigated.Based on the optimal stopping theory, a statistical model of distributed channel access under wireless multicasts was established, and an optimal distributed wireless channel access method under reliable multicast communication was proposed.Each source competes for the channel in a distributed manner, the winner source determines whether to access the shared channel by comparing the reliable multicast access rate with a pure threshold to complete the reliable multicast communication from the winner source to all the sinks.Theoretic optimality of the method was proved rigorously.A corresponding low-complexity algorithm was designed, which has a pure threshold structure and good engineering practicability.Numerical results show that the proposed channel access method can effectively improve the average throughput of the system.关键词：reliable multicast communication;optimal stopping theory;distributed wireless channel access363|592|0更新时间：2024-08-16
- 摘要：Sensor nodes in linear sensor network (LSN) are deployed linearly due to the linear topology of the monitoring area.Most of the state-of-the-art medium access control (MAC) protocols designed for LSN adopted the duty-cycling and pipelined-forwarding (DCPF) schemes.They can reduce data propagation delay and network energy consumption but may still cause issues such as transmission competition, interference, and energy hole.To address these issues, an improved DCPF MAC protocol was proposed based on multiple channels and redundant-node deployment for LSN.The extensive simulations based on OPNET demonstrate that, compared with existing protocols, the proposed protocol shows a better performance in terms of energy efficiency, packet delivery ratio, throughput, and packet transmission latency.关键词：linear sensor network;redundant node;transmission interference;energy-hole;duty-cycling;multi-channel444|351|0更新时间：2024-08-16
- 摘要：In order to deal with the increasingly scarce computing resources, a cooperative edge computing scheme was proposed, which makes full use of the idle resources among users to improve the overall data processing performance.To maximize the user utility, the target problem was formulated as an MINLP (mixed integer non-linear programming), and a learning-optimization-integrated method was proposed to jointly optimize the resource allocation and user offloading decisions.Simulation results show that the proposed scheme can produce a near-optimal solution in sub-second and effectively improve the system utility at least 85.4% compared to the considered benchmark methods.关键词：mobile edge computing;utility maximization;convex optimization;reinforcement learning599|661|0更新时间：2024-08-16
- 摘要：In recent years, China has participated in the global climate governance with a positive stance.The accurate, rapid, and reliable monitoring is fundamental for the situation awareness and layout optimization of the carbon emission.However, the state-of-the-art review with systematic studies of this topic has hardly been identified.The development process and the existing standards of carbon emission monitoring were investigated.The terminal equipment-based methods and the internet of things (IoT)-based applications for carbon emission monitoring were explored, and the IoT-based methods were concluded which hold the advantages including clear structure, flexible design, and reliable transmission.Based on the IoT, an industrial carbon emission monitoring method using the “cloud-tube-edge-end” IoT structure was established.The enabling technologies that can implement the functions of the “cloud-tube-edge-end” were summarized, which are expected to provide guidance for future carbon emission monitoring.关键词：IoT;carbon emission;cloud-tube-edge-end;real-time monitoring849|1579|0更新时间：2024-08-16
-
Influence of the optical switch’s extinction ratio on the crosstalk in fiber-optic seismometer array
摘要：Optical switch is one of the key components in the time-division multiplexing fiber-optic seismometer array and has significant influence on the crosstalk of the array.The source of crosstalk in the array was analyzed based on differential delay heterodyne demodulation scheme.The expression of crosstalk appearing in the seismometer’s output was deduced with the limited extinction ratio.Via simulation, the relationship between the extinction ratio of optical switch and the seismometer’s crosstalk was presented under different array scales, i.e.basically the crosstalk decreases linearly with the increase of optical switch’s extinction ratio, and increases linearly with the increase of the array scale.Finally, an 8-element fiber-optic seismometer array was set up in the experiment, on which different optical switches (acousto-optic modulator and electro-optic modulator) owning distinct extinction ratios could be applied to generate pulse light.Experimental results verified the theoretical analysis as well as the simulation results.关键词：fiber-optic seismometer array;time-division multiplexing;extinction ratio;crosstalk252|415|0更新时间：2024-08-16 - 摘要：With the popularity of intelligent chip in the application of edge terminal devices, a large number of AI applications will be deployed on the edge of networks closer to data sources in the future.The method based on DNN partition can realize deep learning model training and deployment on resource-constrained terminal devices, and solve the bottleneck problem of edge AI computing ability.Thekernel based partition method (KPM) was proposed as a new scheme on the basis of traditional workload based partition method (WPM).The quantitative analysis of inference performance was carried out from three aspects of computation FLOPS, memory consumption and communication cost respectively, and the qualitative analysis of the above two schemes was carried out from the perspective of flexibility, robustness and privacy of inference process.Finally, a software and hardware experimental platform was built, and AlexNet and VGG11 networks were implemented using PyTorch to further verify the performance advantages of the proposed scheme in terms of delay and energy consumption.It was concluded that, compared with the WPM scheme, the KPM scheme had better DNN reasoning acceleration effect in large-scale computing scenarios.And it has lower memory usage and energy consumption.关键词：edge intelligence;deep neural network partition;cooperative computation;parallel partition399|407|0更新时间：2024-08-16
- 摘要：Federated learning (FL) is an emerging machine learning paradigm that can make full use of crowd sourced mobile resources for training on decentralized data.However, it is challenging to deploy FL over a wireless network because of the limited bandwidth and clients’ selfishness.To address these challenges, an incentive mechanism with bandwidth allocation (IMBA) was proposed.Considering the difference between clients' data quality and computing power, IMBA designs a payment scheme to incentivize high-quality clients to contribute their computing resources, thus improving the training accuracy of the model.By minimizing the weight sum of training time and payment cost, the optimal payment and bandwidth allocation scheme was determined, and the training delay was reduced by optimizing bandwidth allocation.Experiments show that IMBA effectively improves training accuracy, reduces the training delay and helps the server flexibly balance training delay and hiring payment.关键词：federated learning;incentive mechanism;bandwidth allocation;Stackelberg game;training quality430|430|0更新时间：2024-08-16
- 摘要：For industrial applications, the time-sensitive services (TSS) and non TSS need to share the radio resources and the same buffer queue in 5G radio access network (RAN).This incurs the complex coupling and uncertainty of the delays.Considering the scenario where ON-OFF TSS and non TSS coexist, the calculation methods for the random delay (including queueing delay) characteristics were established based on the two-stage tandem queueing model, to obtain the mean value, standard deviation, bound violation probability and probability distribution of the delay of TSS.On this basis, the influence of non TSS on the random delay characteristics of TSS was analyzed.It is shown that the influence of non TSS on the random delay characteristics of ON-OFF TSS is related to the rotation frequency of ON-OFF stages.Moreover, the delay probability distribution of ON-OFF TSS has approximately the same right tail length as Gamma distribution.关键词：5G;time-sensitive service;random delay characteristic;tandem queueing model417|418|0更新时间：2024-08-16
- 摘要：Dynamic transmission requirements and the limited cache space bring great challenges to wireless data transmission in the malicious jamming environment.Aiming at the above problems, a collaborative anti-jamming channel selection and data scheduling joint decision method for distributed internet of things was studied from the perspective of frequency domain and time domain.A data transmission model based on multi-user Markov decision process was constructed and a collaborativeanti-jamming joint-channel-and-data decision algorithm based on multi-agent deep reinforcement learning was proposed.Simulation results show that the proposed algorithm can effectively avoid the malicious jamming and the co-channel interference.Compared with the comparison algorithm, the network throughput is significantly improved, and the number of packet dropout is significantly reduced.关键词：collaborative anti-jamming;channel selection;data scheduling;multi-agent reinforcement learning;deep learning733|562|0更新时间：2024-08-16
- 摘要：Most of the existing studies on the scheduling of mixed traffic in time-sensitive network (TSN) focus on how to ensure low-latency of stream reservation (SR) traffic.However, SR traffic blocks best-effort (BE) traffic for a long time will lead to excessive delay in BE traffic, which is not conducive to the state maintenance of network and the management of system.In order to reduce the maximum end-to-end delay of BE messages, a traffic scheduling algorithm which combines with enqueue shaping was proposed.It reduced the blocking of BE traffic by reducing the reserved bandwidth of SR traffic.The algorithm first added enqueue buffers in front of SR queues.Then, the reserved bandwidth ratio among the same priority traffic was adjusted by setting the number of frames in each enqueue buffer entering the SR queue during enqueue shaping.Finally, it jointed egress shaping to reserve bandwidth resources for each SR flow to match its latency requirement.Simulation results showed that combining enqueue shaping can reduce the maximum end-to-end delay of BE messages by 9.66%～75.76%.关键词：enqueue shaping;traffic scheduling;reserved bandwidth;end-to-end delay;TSN437|501|0更新时间：2024-08-16
- 摘要：For the dynamic fading time-varying channel state information, a dynamic adaptive offloading (RLDO) method based on WPT-MEC was proposed to solve the task offloading and resource optimization problems for multiple users by combining wireless power transmission (WPT) technology and mobile edge computing (MEC).The wireless power transmission technology can provide energy to wireless end-user (WEU) and effectively alleviate the problem of limited energy supply from conventional batteries.To maximize the resource utilization, a wireless powered MEC network model was designed where the energy collected by the wireless end-user from the wireless access point (AP) was stored in a rechargeable battery, and then this energy was used for task computation or task offloading.The approach performed real-time offloading decisions through a fully connected deep neural networks (DNN) deployed in the MEC server.A fully binary offloading strategy was used for the offloading decision.Simulation results show that the computation rate of the method can still be maintained above 92% in a multi-user time-varying wireless channel-oriented environment.Compared with the basic method, it has great advantages in improving the calculation rate, reducing the delay and energy consumption，and effectively reduces computational complexity.关键词：channel state information;mobile edge computing;wireless power transmission;deep neural network;dynamic adaptive offloading342|561|0更新时间：2024-08-16
- 摘要：ASCON was the most promising algorithm to become an international standard in the 2021 NIST lightweight authenticated encryption call for proposals.The algorithm was designed to achieve the best performance in IoT resource-constrained environments, and there was no hardware IP core implementation based on this algorithm in the open literature.A software-hardware collaborative implementation method of ASCON was proposed, which improved the speed and reduced the memory footprint of ASCON in IoT security authentication applications through hardware means such as S-box optimization, prior calculation and advanced pipeline design.As a comparison, ASCON has been transplanted on the common IoT embedded processor platform.The results showed that the described method was more than 7.9 times faster, while the memory footprint was reduced by at least 90%.The schemes can be used for the design and implementation of IoT security application-specific integrated circuits or SoCs.关键词：IoT;ASCON;software and hardware co-design;hardware IP core;FPSoC1403|1119|0更新时间：2024-08-16
- 摘要：The development of the discrete manufacturing shows a trend of intelligence, openness and collaboration.As a result, many heterogeneous devices are connected to the industrial internet, which brings serious challenges to the security.Therefore, it is particularly important to introduce trust management and trusted access to devices for trusted measurement.In order to more timely and accurately evaluate the trustworthiness of the edge terminal initially accessing the system, a trustworthiness measurement method based on the device vulnerability database was innovatively proposed.This method adopted the architecture of cloud-edge collaboration, established a device information database and a vulnerability database in the central cloud, and then calculated the terminal risk factor at the edge.Finally, the trust initialization of the access terminal was completed.The simulation results show that the method can well balance the efficiency and security of the system.关键词：industrial internet;device access;security;trust management;trust measurement;vulnerability assessment416|353|0更新时间：2024-08-16
- 摘要：In recent years, to break down data barriers, federated learning (FL) has received extensive attention.In FL, clientscan complete the model training without uploading the raw data, which protects the user’s data privacy.For the issue of clients’ heterogeneity, the contribution of each client to accelerating convergence of the global model as well as the communication cost in the system was considered, aiming at maximizing the weight change of the client's local training model, a client selection optimization problem in FL under theconstraint ofthe delay foreach training round was solved.Subsequently, two federated learning protocols based on the knapsack model were proposed, namely OfflineKP-FL protocol and OnlineKP-FL protocol.OfflineKP-FL protocol was based on the offline knapsack model to select appropriate clients to participate in the aggregation and update of the global model.In order to reduce the complexity of the OfflineKP-FL protocol, OnlineKP-FL protocol based on the online knapsack model to select clients was proposed.Through simulations, it is found that OfflineKP-FL protocol converges faster than the previously proposed methods in certain cases.Furthermore, compared with OfflineKP-FL protocol and FedCS protocol, underthe proposed OnlineKP-FL protocol, not only does the system select fewer clients per round, but also it can complete the model training in 64.1% of the time required by FedCS protocol to achieve the same accuracy for the global model.关键词：federated learning;client selection;knapsack model745|506|0更新时间：2024-08-16
- 摘要：The traditional “centralized” access control technology can no longer guarantee the security of private data access control in the current internet of things environment.Based on the ABAC (attribute-based access control) model, a blockchain based access control framework for the internet of things and a private data access control model were proposed.Firstly, the basic framework and process of access control were described and analyzed in detail, and an auditable access control model was proposed to systematically manage the access control policies of private data through the requests, responses and access records stored in the blockchain network.Then, an auditable access control system based on blockchain technology was proposed, which can provide distributed, fine-grained and dynamic access control management in the internet of things, realize the effective management and auditable access to data, and adopt the access control method based on smart contract to realize the transparent, traceable and automatic access control over the internet of things resources.Finally, simulation experiments and performance tests verify the effectiveness and security of the access control model and system.关键词：blockchain;IoT;access control;smart contract;data security433|407|0更新时间：2024-08-16
- 摘要：With the in-depth development and continuous innovation of blockchain technology, blockchain networks suitable for different application scenarios and design requirements have emerged as the times require, the mutual independence of blockchains inevitably forms the value island effect of the blockchain.Cross-chain technology is an important technical means to realize industrial collaboration and value circulation between different blockchains and improve their interoperability and scalability.Firstly, the basic concept of cross-chain was introduced.And then, the technical difficulties of cross-chain, the technical characteristics of the main mechanism of cross-chain, and the security of cross-chain were analyzed in detail.Finally, the challenges faced by the current cross-chain technology were introduced, and the future development of cross-chain technology was prospected.关键词：blockchain;cross-chain;notary;side chain;Hash-locking1175|990|0更新时间：2024-08-16
Theory and Technology
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
- Total Visits：212817 Visits Today：640
0