This PDF file contains the front matter associated with SPIE Proceedings Volume 12641, including the Title Page, Copyright information, Table of Contents, and Conference Committee information.

Network traffic classification is an important part of network management. For dark net traffic, which is encrypted, The dark net is often used by criminals to transmit illegal information, and the dark net traffic is encrypted. How to detect the traffic of the dark network is of great significance for combating the criminal activities of the dark net. A good network traffic classification algorithm is inseparable from the efficient network traffic attributes. This paper uses a novel network flow-based property, packet interval time entropy, to classify dark web traffic. Because the dark net data transmission needs to forward through multiple routers, and each dark network router needs to resolve the next transmission of the dark network router address, so the dark network packet direct time interval will be different from the ordinary Internet packet time interval, packet interval time entropy can efficiently distinguish the dark web traffic and conventional Internet traffic. Experiments show that the packet interval time entropy property can effectively improve the classification accuracy under the KNN algorithm.

Logistic regression has the threat of privacy disclosure during data analysis. Differential privacy can prevent this security threat. However, differential privacy mechanism needs to add noise to the protected algorithm, which will affect the quality of data analysis. To improve the quality of data analysis, we propose dpLogic algorithm based on differential privacy and functional mechanism. This algorithm constructs an auxiliary function, which can help us get the noisy model as close as possible to the original model. Besides, dpLogic optimizes privacy budget allocation, reducing the disturbance of noise on the model. Experimental results illustrate that this algorithm can effectively improve the accuracy of data analysis at a high level of privacy protection.

Since the first half of 2020, the COVID-19 epidemic has continued to spread across the country. Based on this background, with the continuous promotion of a new round of technological innovation and industrial transformation, and the combined impact of the epidemic factors, the digital economy has become a new pillar of the steady development of China's macro economy. Emerging industries have provided good opportunities for the digital economy. Cyber security has risen to the height of national sovereignty, which is the direct embodiment of national competitiveness and the foundation for the healthy development of the digital economy. However, with the concentration of massive data and the progress of information technology, the data is easily and conveniently used, personal privacy security, corporate business secrets and even national security suffered serious damage, and network security protection has also become the bottleneck of the digital economy to a new level. It is particularly important to strengthen the network security governance capacity, improve the network security laws and regulations, and implement the hierarchical protection system.

Honeypot is an active defense mechanism in network security protection. Its main purpose is to collect various intrusion information. However, due to the complexity of attacker penetration attacks and the diversity of host vulnerabilities, a simple honeypot can’t to meet the defense requirements, so a honeynet system formed by a combination of various honeypots has emerged. But what kind of honeypots are included in the honeynet? How are honeypots connected? What is the reasonable size of the honeynet? These problems are restricting the development of honeynet. Therefore, the research of this paper is to analyze the vulnerability of an nodes in the network by establishing an attack graph, so as to formulate a honeynet deployment strategy. However, due to the difference in the network structure and the size of the honeynet, the effect of security protection will be very different. At the same time, considering the maintenance cost and protection benefits of the honeynet, when adopting a dynamic honeynet deployment method, we need to adjust the type and number of honeypots in the honeynet at any time. This paper uses ant colony algorithm and wolf swarm algorithm to optimize the honeynet, so as to realize the simplification of the attack graph and the dynamic configuration of the honeynet.

In recent years, China's digital industrialisation and digitisation of industries have continued to develop in depth, and online transactions have become the shopping choice of many internet users. The strong endogenous dynamics have put forward requirements for high efficiency, low cost, high security and high privacy for online mobile payments. Blockchain is a technology that has emerged with the popularity of digital cryptocurrencies such as Bitcoin, which is characterised by "decentralisation" and "immutability", but blockchain transactions have privacy protection flaws and problems that seriously threaten the privacy and security of transaction users. This paper proposes a clear solution to this problem by combining homomorphic encryption and ethereum's smart contract technology to design a blockchain online transaction privacy protection system that is compatible with heterogeneous trading platforms. This effectively makes up for the shortcomings of blockchain technology in terms of privacy protection. The security of this scheme has been proven to be unforgeable and the privacy data is safe. Through performance simulation experiments and theoretical derivation, it is proved that the privacy data is distributed, shared and computed in a ciphertext state in a more efficient way, and the proposed scheme can protect customer privacy more effectively than the traditional model.

Static analysis is a popular approach for finding vulnerability in software. However, suffered from its imprecision, it is quite difficult and time-consuming to confirm static warnings. It also leads to a vulnerability verification problem. In this paper, we regard the verification problem as line reachability problem and utilize directed symbolic execution to verify specific bug. Existing techniques mostly relied on static heuristic metric but neglected some special dynamic metrics. For this reason, we designed a new search strategy named as DCCM which modifies others forward strategy via collecting critical constraint during the process of dynamic symbolic execution. By conduct contrast experiments on example program and five real-world C projects, we find out that DCCM performs well than KLEE RP, KLEE CoverNew and SDSE search strategy on many testing scenarios.

As a core hardware in the development of satellite cryptographic devices, FPGAs are required to go through extensive functional correctness validation before finally bound to the channel equipment. However, the diversity of satellite communication systems and the bad habits of some developers often lead to the ineffective reuse of verification module. To improve the reuse rate of verification module and the verification efficiency of cryptographic devices, we propose an FPGA partial reconfiguration method based on minimum functional unit division (MFD). First, we present a formal description of the minimum functional unit (MFU), which is a collection of modules that are highly cohesive and atomized with respect to the system's entire functional design. Then a reconfiguration method based on MFU global search is given as an implementation of MFD. Next, the performance of MFD is analyzed in terms of reconfiguration granularity, implementation complexity, extensibility and module reuse rate, and the analysis results show a good performance in optimizing FPGA partial reconfiguration design. Finally, we design a MFD based satellite cryptographic device verification platform, in which the hardware interface, communication protocol, cryptographic algorithm and other functional modules are selected as MFU to complete the partial reconfiguration design. The design of the platform based on MFD shows the practicability of MFD, which can provide theoretical guidance for further applying FPGA partial reconfiguration technology in satellite cryptographic device verification platform.

In the electric power communication network architecture, the Internet of Things business requirements of mutual perception and coordination of each link, and the traditional security protection structure based on boundary isolation, it is a difficult point to achieve a high degree of compatibility and coupling. Therefore, when PCE performs path calculation in the multi-domain FlexE network, it is necessary to consider the delay of the transmission path. The low-delay transmission of the service is ensured. This paper analyzes the network-slicing requirements of typical business scenarios in the power grid. In view of the hierarchical architecture of the power Internet of Things, this paper analyzes its requirements for 5G technology and network slicing, and the overall architecture of the integrated energy Internet. Finally, according to the current business needs, this paper proposes a power slicing scheme that meets the needs of the power grid’s various business scenarios. On this basis, an end-to-end network architecture based on power slicing is designed. The results verify that the safe and efficient slicing strategy based on deep reinforcement learning has better performance than the traditional machine learning model, which can enhance the monitoring effect of the terminal network security of the power Internet of Things. It can also improve the terminal security of the power Internet of Things at this stage and can detect the potential risks of terminal security monitoring.

With the rapid progress of the distribution network construction, a large number of intelligent terminals need to be connected to the distribution network, which brings hidden dangers to the power grid security. Intelligent distribution master stations and terminals using the existing SM2 algorithm can obtain shared keys on insecure communication channels. However, due to the high concurrency of the main station and the limited computing power and storage capacity of the terminal, the simple use of the original SM2 algorithm cannot guarantee the fast and secure bidirectional identity authentication of the smart distribution network. Considering the high concurrency of the main station and the limited computing power and storage capacity, this paper proposes the distribution terminal security protection method, security chip and system deployed in the intelligent distribution terminal, add encryption authentication unit in the terminal security chip, adopt improved authentication key negotiation based on SM2, combine SM1 data based on decryption technology, realize secure bidirectional communication between the distribution main station and the distribution terminal.

In recent years, with the vigorous development of new businesses such as smart electricity and power trading in the power industry, the open interconnection capability of smart power grid is becoming stronger, and the number and types of terminal access to electric Internet of Things are also increasing. With the continuous development of Internet of Things technology and business, it also brings new challenges to the security of Internet of Things. As the Internet of Things has a wide range of access points, in the remote transmission of massive terminal data and information, there are bound to be many security problems in the process of cross-regional access authorization on heterogeneous networks. This paper mainly analyzes the status quo and risks faced by the electric Internet of Things, and discusses how to effectively use security technology to protect the electric Internet of Things. A security protection system of electric Internet of Things based on depth protection is proposed to ensure the security of electric Internet of Things.

Stronger anonymity password authentication key exchange protocols based on traditional cryptography are not resistant to quantum computing attacks, and there is no lattice-only anonymous password authentication key exchange protocol protocol that satisfies the strong anonymity requirement. Therefore, this paper constructs stronger anonymity PAKE protocol based on the ring learning with errors (RLWE) and smooth projective hash function (SPHF) based on ideal lattice. The proposed protocol can achieve stronger anonymity and resists to desynchronization attacks and quantum attack. The security of the protocol is proved under the BPR model. The security and performance analysis shows that the proposed PAKE protocol has better advantages in security and efficiency.

In view of the current problems of long property rights validation period, transaction difficulties and data sharing, we propose a blockchain-based Intellectual property transaction sharing scheme by combining the characteristics of open and transparent, tamper-evident, traceable and decentralized data of blockchain with the proxy re-encryption technology and off-chain storage mode. We also improve the PBFT master selection strategy and design the VCS-PBFT consensus mechanism, which effectively reduces the network bandwidth consumption and improves the consensus efficiency. By comparing and analyzing with other schemes, it shows that this scheme has certain advantages and feasibility in function.

This paper presents a bypass attack on the Field Programmable gate array (FPGA) cryptographic chip Advanced encryption standard (AES) encryption algorithm. Since the accuracy of half-byte classification is low and the model is slow to converge and easy to overfit when using deep learning, this experiment proposes for the first time to introduce the deep learning network EfficientNet model into the field, which only targets the electromagnetic leakage information of the hardware during encryption for any half-byte, when the plaintext and ciphertext are unknown. This is followed by a "divide and conquer" approach to the entire key, where each half-byte is attacked to obtain the entire key. The accuracy of the model used in this study was found to be about 17% higher than that of common networks such as ResNet and DenseNet, and the model converged better.

With the development of science and technology, all kinds of data are generated in our daily life. While data analysis brings convenience, it also poses threat to users' privacy. This paper proposes a data security privacy protection model with blockchain in IoT. First, encrypt data before it is stored in the cloud. Secondly, in order to verify user identity and ciphertext safely, an efficient certificateless signature scheme is proposed. The signature scheme is security under random oracle model and time performance is efficient.

In this paper, we propose the concept of out-degree betweenness based on betweenness. Out-degree betweenness can reflect the minimum cost attack path of a network and also the ability of nodes to provide the number of attack paths. It is studied by combining honeypot, vulnerability attack graph, and minimum cost attack path. The experimental results show that out-degree betweenness can better evaluate the centrality of network nodes.

Information hiding technology can embed specific information in the digital host. Because the host information changes little, it will not arouse the attention of the monitor. In order to improve the information hiding effect and anti-violence extraction ability, logistic chaotic mapping and ECC encryption are introduced, and random hiding of information after encryption is proposed. Before the data is hidden, the information is divided and combined, and the control information is also embedded in the BMP image to realize the improved LSB algorithm.

With the increasingly prominent problem of information security in the network era, the establishment of an effective secure communication system has become the primary problem for enterprises. How to ensure the secure communication of data will be a long-term issue in the information and data age. Therefore, based on the relevant standards of CCSA technical requirements for security certificate management system of Internet of vehicles wireless communication technology based on LTE and the personalized requirements of the vehicle manufacturer, this paper studies the security certificate management and key management of intelligent Internet connected vehicles. The certificate management content covers the whole life cycle management process of certificates, including the overall business processes of application, distribution, cancellation, and update, and direct communication security design. The main goal of key security is to ensure the security of the key used in the identity authentication system in its generation, storage, use, update, abolition, archiving, destruction, backup, and recovery.

With the rapid development of computers, there are more and more network security problems. While the network brings convenience to people's life, it also brings many security risks. Many users have suffered losses due to vulnerabilities in the network. In order to maintain the security of cyberspace, it is urgent to build a network security ecosystem. Web fingerprint identification technology is a network security technology, which is widely used in vulnerability detection. However, traditional Web fingerprint identification technology. This paper will optimize the matching between Web fingerprint identification and fingerprint database according to the improved PSO particle swarm optimization algorithm to improve efficiency and accuracy.

The encryption and decryption time of semantically secure encryption techniques have a great influence on the total end-to-end delay of an ORAM scheme. However, to the best of our knowledge, there is no work to study the impacts of combinations of block cipher and common model of operation on implementing semantically secure encryption techniques by comparing and analysing their execution time. This paper examines the running time of commonly used combinations of block cipher and model of operation to find the most matchable one for an ORAM scheme. We first selected the candidates of symmetric block encryption algorithms based on the existing block ciphers included in the LibTomCrypt, a popular cryptography toolkit. We second designed a comprehensive experiment in terms of prominent application scenarios of ORAM. We finally analysed the experimental results and found that the encryption algorithm of RC5-CBC outperforms others. In the end, we conclude that the RC5-CBC is the most matchable combination of block cipher and model of operation for implementing an ORAM scheme.

Based on the problem that existing antivirus software cannot effectively respond to emerging viruses and lacks real-time detection and effective interception of viruses, a class of detection propagation models based on log feature detection of viruses is established based on the classical susceptible-exposed-infected-immune (SEIR) propagation model with the introduction of containment nodes. The experimental results show that the drop rate of infected nodes of this model is increased by 26.45%, 40.68%, and 54.24% when the detection rate is taken as 0.5, 0.7, and 0.9, respectively, compared with the previous SEIR model. Therefore, the detection propagation model based on the detection of log feature information of viruses can effectively contain and remove the spread of new viruses in the network, and the higher the detection rate, the better the effect.

Differential protection of distribution network has high communication requirements for reliability, time delay and delay jitter. Theoretically 5G can meet these requirements. However, real network test results show that there are problems such as burst large delay and burst packet loss for differential protection based on the current 5G network, and deterministic capability of the current network is poor. This paper first gives the real network test results of differential protection based on 5G, and then gives a general description of deterministic technologies and the possible solution to meet the requirements of differential protection.

This paper presents a method of UAV formation trajectory tracking control based on fuzzy control. Aiming at the trajectory tracking problem of UAV formation, starting from the UAV formation control, the fuzzy control method is adopted to obtain the revised PD parameters through fuzzification, determination of membership, fuzzy rule reasoning, and clarification, so as to realize the real-time adjustment of UAV formation control parameters, and finally make the UAV formation reach the docking position quickly and stably. Through simulation, the effectiveness of the fuzzy algorithm is verified, the control logic of UAV formation is simplified, and the flight stability of UAV formation is improved.

As the main infrastructure of the power grid, the electrical power communication network has an urgent need for digitalization, and the digital twin technology can support the digital operation of various services of the electrical power communication network. Based on the actual production needs of electricity, this paper divides the digital twin system into five layers: the physical layer, data perception layer, data governance layer, functional model layer, and model application layer. In order to achieve practical application, the five-layers model is studied based on actual cases, and then the key technologies required to build a digital twin model are introduced. Finally, two examples of optical cable resource management and intelligent analysis of communication power failure and remote management are shown.

Deep steganalysis networks (DSNs) has made a great progress in detection performance. However, most of the deep image steganalysis networks are not complete end-to-end models. In their approach, traditional hand-crafted features are employed to pre-process the images, which can obtain the high-frequency noise residuals to alleviate the interference of the image content. To avoid relying on domain knowledge of deep learning-based methods, we design an end-to-end deep image steganalysis neural model that combines multi-scale feature extraction and residual fusion modules. Firstly, we use the standard convolution kernels of different sizes to extract the features of different scale receptive fields in the input image. Then, depth-wise convolution is used to independently models the inner-channel correlations of multi-scale features and retains the discriminative statistical characteristics of each channel. The residual fusion technique is introduced to aggregate hierarchical feature and strengthen information representation in the network. Experimental results show that, compared with the existing classical deep image steganalysis networks, the proposed steganalysis scheme has a great improvement in steganalysis error rates.

In this project, a novel Internet of Things (IoT) ultra-narrowband (UNB) network architecture, including multiple(various) multiplexed frequency bands and channels was introduced for the uplink spread. Internet of Things (IoT) devices is able to select frequency band and transmit their data packets randomly. However, a base station (BS) is only able to obtain data from one multiplexing band due to hardware limitations. The complexity of conducting a Fast Fourier Transform (FFT) at a very limited sampling interval on multiplexed bands to deduce the influence of IoT device uncertainty and synchronize onto transmissions is the main reason for hardware constraint. This paper has looked into some previous work and experiments such as UNB LPWAN networks, as well as several concepts mentioned in other research. More importantly, the paper has also introduced various technologies of IoT networks to help obtain a more thorough view of current work. The purpose of this paper is to figure out a way of assigning base stations to multiplexing bands, therefore maximizing the packet decoding probability (PDP). This work comes up with an algorithm aiming to better assign the band in order to reach a maximization in PDP, importantly, transfer it into finding the maximum SINR. With the use of our algorithm, three results reveal to meet the requirements. All the methods that have been looked into and the one that is introduced in this paper significantly exceed the performance of the original random allocation. In addition, this paper also seeks to improve the algorithm by figuring out how the interference signals are assigned and obtain a maximum PDP by calculating the transmission with the minimum interference.

Hierarchical cluster architecture is usually adopted in large scale unmanned aerial vehicle (UAV) network. Due to mission scheduling scheme, the assigned mission determines the number of UAVs in the cluster. Aiming at the problem of rapid changes of cluster head caused by high mobility in cluster, we propose a stability improved weighted cluster head selection algorithm (SI-WCSA) for mission-oriented flying ad hoc networks based on speed similarity, energy, distance and node degree. Firstly, it improves the mobility metric of speed similarity, which can accurately reflect relative movement trend. Then, the scheme proposes three energy consumption rates and introduces weight threshold indicator for the maintenance of cluster head. Finally, we adopt combination weight coefficient calculation methodology based on analytic hierarchy process and the entropy weight method. Simulation results indicate that the algorithm can significantly reduce the frequency of cluster head replacement and balance energy consumption within clusters compared to the benchmark approaches.

In order to provide Internet services, the Domain Name System (DNS) plays a crucial role. Muti-Domain Cooperative Alliance is proposed to promote DNS globalization and maintain the permanent sustainability of the Internet. Data synchronization transactions in Muti-Domain Cooperative Alliance must ensure mutual trust. Blockchain technology is one of the effective ways to solve the problem. However, mutual trust transactions between blockchains are still problematic in synchronizing data cross-chain. This paper summarizes and analyzes the properties of various cross-chain technology and proposes a cross-chain solution. A cross-chain solution is designed as a main-sub chain mode, with the main chain constructed using the Hyperledger Fabric of semi-centralized attributes. The channel mechanism of Fabric ensures effective mutual trust transactions of cross-chain information transfer in the cross-chain system and improves the efficiency of the cross-chain system. Finally, an experiment evaluated the performance of the cross-chain relay chain by testing its carrying capacity, and the results illustrate that the solution can hold cross-chain information interaction.

In the context of decoupling and clipping by using message middleware in the network management alerting subsystem, the requirement of the timing of data in distributed systems is difficult to achieve. To address a series of problems caused by message disorderliness in the network management system, the article introduces RocketMQ sequential messages and the solution to message duplication. Firstly, it introduces the message requirements of the network management system, then discusses the advantages of RocketMQ in comparison, and by analyzing its implementation method and principle of sequential messages, it is applied to the alerting system and tested, and the results show that the method can better guarantee the problem of data timing, and finally, it solves the problems of message loss and message duplication that may occur with sequential messages.

The functions of navigation satellite include positioning and timing, which are based on high-precision time comparison. Most of the existing studies ignore the relativistic effect of the satellite-earth time comparison, or study some delay based on general relativity alone. There is no complete calculation model of the satellite-earth time comparison based on special relativity. In this paper, the satellite-earth time alignment model based on special relativity is derived and the space-time diagram of special relativity is established. In order to improve the accuracy of the satellite-earth time alignment, the uplink Sagnac effect delay and the special relativity effect delay are derived based on the relative motion of the satellite-earth. By calibrating the model of the satellite-earth alignment of the special relativity theory with the above two dealy terms, the alignment accuracy is improved. The experimental results show that the satellite-earth time comparison model derived in this paper is correct. After correction and optimization of the uplink Sagnac effect delay and special relativity delay, the maximum accuracy improvement and the average accuracy improvement are 0.022% and 0.014% to GPS. The highest accuracy improvement was 0.079%, and the average improvement was 0.012% to GLONASS . It can be seen that the accuracy of the satellite-earth time comparison model under the relativity theory is higher than that of the un-optimized satellite-earth time comparison model.

Considering the issues of lack of integrity, authentication and confidentiality of Modbus/TCP protocol broadly utilized in SCADA system, this study suggests a trusted Modbus/TCP protocol based on digital envelope technology on the basis of trusted cryptography module (TCM). To the basic Modbus/TCP protocol, the protocol adds identity authentication sub-protocol and communication sub-protocol. Utilizing the SM2, SM3 and SM4 algorithms provided by TCM, the security of sub-protocol communication procedure and communication information is guaranteed by digital envelope technology and hash calculation. The security of the sub-protocol is verified by the SPAN tool, as well as the time overhead determined by the sub-protocol is tested. Experiments illustrates that the proposed trusted Modbus/TCP protocol can meet the high real-time prerequisites of industrial control systems and various business performance criteria while ensuring integrity, authentication and confidentiality.

Coding theory has so many important applications in cryptogram, communication technology and network security, etc. The MacWilliams type of identities for linear codes over rings are very important research object, becauce they are related to the computation of decoding error probability and error probability of undetectable codes. In this paper, we study linear codes over the ring Z₄ + uZ₄ with u² = 0 . The Hermitian inner product over such ring is defined. New definitions for the complete weight enumerator, the symmetrical weight enumerator, the Lee weight enumerator and the Hamming weight enumerator of linear codes over the ring Z⁴ + uZ₄ for the Hermitian inner product are given. The MacWilliams identities for these weight enumerators are studied.

Searchable encryption is the best choice for secure retrieval solutions in Internet of Vehicles scenarios. However, the search time is positively correlated with the number of keywords among most of public-key searchable encryption schemes. Also, the traditional keyword search strategy cannot meet the specific search needs of users, that is, the needs of supporting any Boolean expression search strategy. In order to reduce the search time and improve the search flexibility, this paper proposes a Boolean search scheme that supports constant search overhead for Internet of Vehicles users. Boolean search expressions in the proposed scheme can support search strategy for arbitrary Boolean expressions. It uses vectorized search expressions and keyword values in the form of predicate vectors, and the cloud server only needs to pair once to complete the Boolean search operation. This scheme can effectively resist the chosen keyword attack, and experimental simulation shows that the proposed scheme can efficiently realize ciphertext retrieval in the Internet of Vehicles environment.

Amplify-and-forward (AF) relay protocol can decrease system’s transmission time slots and full-duplex (FD) technique can improve system’s spectral-efficiency (SE). Combing relay technique with FD technique can further improve system’s SE. In this paper, two transmission strategies with considering the existence of direct link in relay transmission are designed first. At the same time, the detailed bidirectional SE and outage probability analyses of the designed transmission strategies are given. In addition, the energy-efficient transmissions of the designed transmission strategies are achieved. Furthermore, the EE comparisons among the designed transmission strategies are given. Simulations reveal that even with the residual self-interference, the designed transmission strategies still can improve the system’s SE and EE.

By studying bandwidth allocation and puncture schemes, this paper intends to make a trade-off between QoS requirements of different services. For eMBB scheduling problem, we used proportional fairness algorithm (PF) as its resource scheduling algorithm, and then asked URLLC to perform puncture on the basis of eMBB scheduling results. Considering the case of URLLC overload, we set a rate loss threshold for eMBB users. When the eMBB rate loss reaches the threshold, the URLLC users in the current time slot will delay the transmission to the next mini-slot. In this paper, we propose a bandwidth allocation problem whose optimization objective is to minimize eMBB rate loss and URLLC transmission delay, and use matching theory to solve this problem.

In the quantum cheque schemes that have been proposed, a strong quantum capability of the user is required and the quantum resources consumption is relatively high. To solve this problem, we propose a quan tum cheque protocol based on quantum state computational distinguishability, in which a bank issues a blank cheque, a semi-classical user can sign the amount on the cheque and finally the bank verifies the validity of the cheque. Compared to existing scheme, our scheme requires less quantum capability and reduces the overhead of quantum resource. Since the scheme is based on a hard problem that can resist quantum attacks, the quantum cheque cannot be forged or altered.

Cloud computing is an emerging service computing model centered on web services and applications, which is used to manage, schedule, and integrate various resources distributed on the network. The rapid development of cloud computing has led to the emergence of a wide variety of cloud services. In the face of many cloud services, users generally select high-quality cloud services through user evaluations. However, the subjectivity of users' evaluation of services and false user evaluations make later users fail to select efficient services. To address this problem, the authors take the QoS (quality of service) of cloud services as the data base, obtain the subjective weights of cloud services by AHP hierarchical analysis, obtain the objective weights of cloud services by entropy weighting method, and obtain the recommendation degree of each service by weighting, and obtain the trustworthiness of cloud services by TOPSIS decision method at the same time, and finally propose to construct a cloud service trustworthiness metric model. This model reduces the influence of subjectivity of user evaluation on the trustworthiness assessment of cloud services by providing an objective and effective trustworthiness measure for cloud services. In this paper, the research focuses on processing QoS data, classifying the metrics, and using the entropy weight method to obtain the objective weights of the corresponding metrics of cloud services.

To solve the problems of low hidden capacity, weak concealment and low transmission efficiency in the existing blockchain covert communication, a covert communication model of Ethereum based on smart contracts is proposed. First, the sender processes the plaintext information and the key information into a binary sequence, to get the secret information to be embedded. The sender embeds the secret information into the smart contract source code by using the space method. Secondly, the sender compiles and deploys the smart contract containing the secret information to the Ethereum network. Finally, the receiver can use the pre-shared account information and key to restore the plaintext information to the acquired smart contract through the inverse process of the embedding algorithm. Through the deployment experiment of the Ethereum test network, the results show that the model realizes covert communication based on smart contract information hiding. By analysing and comparing the existing similar models, the model further improves the hidden capacity, reduces the communication cost, has better performance concealment and security.

Recently, Arikan has proposed a new scheme to transfer the error correction part to external codes for encoding, namely polarization-adjusted convolutional (PAC) coding. However, all existing decoding methods of current research for PAC codes focused on the classic sequential decoding-Fano decoding. The stack decoding which is another well-known sequential decoding, has not been reported. In this paper, we try to do a thorough study of the stack decoding for PAC codes. First, we find that the traditional stack decoding (SD) is not very effective for the PAC codes, especially in terms of complexity. Therefore, we propose a soft output stack decoding (SOSD) for PAC codes, which can reduce the computation very much. In the SOSD algorithm, the discarded paths are recorded and used in the final decision of decoding codeword. The SOSD algorithm can reduce the stack size (amount of computation) very much with almost no decoding performance loss. By carefully choosing the stack size and soft output threshold, in the case of a PAC code with code length 128 and rate 1/2, the SOSD algorithm can reduce 82% of the computation by comparing the conventional SD algorithm.

In order to improve the road traveling efficiency and vehicle driving safety, a lane selection artificial potential field method is proposed based on the dynamic topology of the vehicle cluster model with Internet of Vehicles. The real-time decisions of lane changing for collision avoidance are made by the reformed lane repulsion of artificial potential field. Then, a lane-changing trajectory is planned using 2-segment 3-order Bezier curves. Meanwhile the safety control points of the main lane and the safe distances of the target lane are proposed as the safe constraints for lane-changing process. Finally, based on the MATLAB platform, the simulation is performed for validation under the typical conditions of different obstacles under the high-speed background. The results show that the lane selection artificial potential field method can make the real-time lane-changing decision effectively, and the lane-changing trajectory planning under the set safe constraints can effectively avoid the occurrence of vehicle collision accidents.

With rapid economic, technological, and industrial development, the transport network is under increasing load pressure. Making the best use of existing road network resources and traffic management resources, improving traffic and travel management, and increasing the efficiency of road use are essential issues we need to face. In this paper, the traditional BPR model is modified to take into account complications such as intersections, traffic accidents, road maintenance under congestion, and delay time due to queueing under congestion. The queueing time of a congested traffic flow is estimated using a trajectory tracking method using the cell phone signal of the driver and the base station data in the traffic network. The traditional BPR model was creatively modified to develop a modified BPR functional model for estimating traffic delay, which can be applied in transportation network planning. The final experimental model simulates the effect of traffic delay during vehicle movement, reflecting road flow, and providing effect support for road network planning.

With the rapid development of computer communication technology, open source service framework system, represented by Dubbo and Spring Cloud in the Internet field, and internal private service framework system are applied to different business scenarios. How to share service registration and configuration information among multiple service frameworks in a simple, efficient and stable way has become an urgent problem to be solved. This paper proposes a service registration and configuration center supporting heterogeneous systems, which is responsible for providing unified service registration and configuration management, and supports clients between heterogeneous systems to obtain the latest service registration and configuration information in real time. Supports the service registration and configuration center of heterogeneous systems to store and take effect in real time service registration and configuration information between heterogeneous systems through the regularization processing and change information push functions. The experiment shows that the heterogeneous system service registration and configuration center can meet the needs of unified management of service registration and configuration between different business systems in the new generation control system, and ensure the reliable acquisition of service registration and configuration information under different architectures.