This PDF file contains the front matter associated with SPIE Proceedings Volume 12245, including the Title Page, Copyright information, and Table of Contents.

In telemetry, tracking, and control (TT&C) systems, noises can impact the frequency stability and therefore cause errors in velocity measurement. Allan variance is used to characterize the frequency stability in the time domain. In the frequency domain, spectral density of phase fluctuation is utilized in theoretical analysis, whereas phase noise is used in practical engineering. Therefore, when calculating the velocity measurement errors caused by phase noise, a conversion between the time domain expression and the frequency domain expression is needed. This work presents the corresponding relationship between the time domain and the frequency domain representation of frequency stability, and then derives a convenient way to estimate the impact of phase noise on velocity measurement errors. At last, with common TT&C equipment being taken as examples, simulations are given to show how phase noise impacts the velocity measurement precision in practical engineering.

The Internet of Things(IoT) is a communication technology which will have a deep influence on data sharing. However, IoT requires security access and data traceability against evil attacks. Blockchain, as the infrastructure of Bitcoin, perfectly meets the need. In this paper, we give out a generic data sharing framework based on consortium blockchain and IoT(CBIoT). Then a formal description and two security characteristics are purposed. Finally, an implementation is presented combing with block.

The virtual network function carried by the container has the advantages of strong orchestration ability and high flexibility. However, due to the poor natural isolation of the container, the security of the containerized virtual network function has attracted much attention. After trusted measurement of containerized virtual network functions based on trusted computing, how to reduce the privacy exposure of containerized virtual network functions under the premise of effective response to challengers has become an urgent problem to be solved. This paper proposes a trusted remote attestation protocol based on attribute certificates and ciphertext policy attribute-based encryption for containerized virtual network function, and uses Ban logic to perform inference analysis on this protocol. The results prove that the proposed is correct and can meet the expected goals.

Aiming at the limited energy load of LEO satellites, this paper researches on the digital precoding design of LEO satellite communication system based on energy efficiency (EE) maximization, combining with massive multiple-input multiple-output (Multiple-Input Multiple-Output, MIMO) precoding technology. At the same time ,we consider the joint constraints of quality of service (QoS) and total transmission power. First, the system model and channel model are analyzed. Then, for the fractional programming problem of EE maximization modeling, the Dinkelbach algorithm and the zero forcing (ZF ) precoding algorithm are combined to complete the precoding design. The simulation results show that the algorithm design can show good performance while ensuring QoS.

With the wide application of information and internet technology in power system, distribution secondary system is facing more complex security threats while being more flexible, open and efficient. As a key information infrastructure, intelligent distribution terminal has become an object requiring key security protection; In addition, with the development of distribution Internet of things, the number of acquisition terminals is huge, the terminal access data presents the characteristics of high concurrency, and the pressure of terminal security access is increasing. This paper proposes a service security isolation method based on subcarrier isolation, which divides all subcarriers of broadband power line channel into several subcarrier sets and assigns them to different services. The services monopolize their own subcarrier sets, so as to achieve the purpose of safe isolation of communication channels.

Point-to-Point Tunneling Protocol (PPTP) is a mature way to implement Virtual Private Network (VPN), which has been widely used by many companies and enterprises. The security of VPN mainly depends on the security of the protocol, so the analysis of the protocol has always been the top priority of network security. This paper conducts an analysis of PPTP implemented by Microsoft. In this paper, I first give an introduction to the structure of Microsoft's PPTP, followed by an in-depth analysis of the Microsoft version of the Challenge-Handshake Authentication Protocol (MS-CHAP) and Microsoft Point-to-Point Encryption (MPPE) subprotocols. And in the second half of this paper, I analyze and implement a man-in-the-middle attack on Microsoft's PPTP that does not require the challenge packet.

In order to improve the data processing capability of the power acquisition system, the electric information acquisition system can be better adapted to different communication environments, in this paper, a multi-mode communication mode with strong adaptability is proposed. In the upstream channel includes GPRS, optical fiber communication, 3G wireless private network, 3G wireless public network, 4G wireless public network, 4G wireless private network and other existing communication mode, and in the downstream channel contains a low-voltage narrow-band PLC, low-voltage broadband PLC, wireless sensor networks and RS485 and other communication modes. Through the analytic hierarchy process to achieve the switch between the various communication modes to adapt to different environments.

Algebraic coding theory has so many important applications in information sciences, cryptogram and communication technology, etc. In this paper, we study Z₂Z₄[u]-additive codes in three aspects: Gray map, MacWilliams identity and one-weight codes. We first defined a Gray map and then prove that it is a weight preserving and a distance preserving map. A MacWilliams identity which connects the weight enumerator of an additive code and its dual code is deduced. Finally, we discuss one-weight Z₂Z₄[u]-additive codes.

The cryptography has a history of thousands of years, and its development can be divided into three stages, classic cryptography, neoteric cryptography and modern cryptography. Nowadays, it has been further invigorated by 20th-century technology. Previous studies lack a systematic review, so the public has a relatively shallow understanding of the issue. Thus, to increase public awareness, the encryption and decryption are introduced in a easy to understand way. Through extensive reading, generalizing and combing, this paper introduced the number theory applied in some examples of classic, neoteric, and modern cryptography. It is found out that the knowledge of congruences, arithmetic, factorization in number theory are widely utilized in cryptographic algorithms, among them, congruences are the most secure and common type. It is also found out that the number theory can help create the algorithm to encrypt and decrypt messages.

KPI anomaly detection has always been a very important task in AIOps (Artificial Intelligence for IT Operations). Operations Engineers judge whether the service is stable by monitoring various KPIs. However, there are many challenges to build a KPI anomaly detection system, and the existing anomaly detection models are not strong. Given the effectiveness of the Transformer model in a series of fields such as NLP and CV, it has attracted great interest from researchers. Therefore, in this work, we build an end-to-end system named KPI_TransformerE for KPI anomaly detection. In particular, for the Class Imbalance problem, we use Focal Loss to solve it. Finally, we test our method on public datasets. The experimental results show that the method can achieve an F1 score of 0.963 in the test set which proves the power of our model.

Message clustering is the first and most crucial step in protocol reverse engineering, and the accuracy of protocol clustering directly affects protocol format extraction and state machine inference. The existing text protocol clustering methods do not apply to industrial control protocols, and the number of clusters and cluster centers cannot be determined automatically. The n-gram model is used to extract the protocol message features and construct the message feature vector for the clustering algorithm, which preserves the feature information in the protocol messages; density peak clustering is used to avoid the problem that the initial number of clusters and cluster centers cannot be determined automatically, and the automatic clustering of industrial control protocols is realized. It is tested on three commonly used industrial control protocol datasets, Modbus, DNP3, and S7common, and achieves good results on two metrics, purity and F1 score.

Aiming at the problem that traditional deep learning methods have low detection rate of hidden attacks when processing high-dimensional data, an intrusion detection model based on recurrent neural network is proposed. Use the Borderline-Smote algorithm to oversample the data set, On the basis of LSTM, double-layer attention mechanism is introduced to extract the main feature information through byte and sequence dimensions to generate a more accurate feature representation. Finally, the softmax classifier is used to obtain the classification results. Use UNSW-NB15 data set as the experimental data, the experimental results show that this model is compared with the similar deep learning detection model has higher accuracy and lower the rate of false positives.

Traffic security is one of the important topic in cyber security of smart city. With the widespread use of encrypted traffic, more and more malware prefers to use encrypted traffic to transmit malicious information. Since the transmission content is not visible, the traditional detection method based on deep packet inspection is not effective anymore. In this paper, by analyzing the protocol and the sessions of malicious encrypted traffic and benign traffic, a weight naive Bayes-based method for detecting malware encrypted traffic and classifying malware family is proposed. The method construct a hybrid fingerprint for each malware family traffic and benign traffic. First a hybrid fingerprint-based identification is performed to distinguish between malware families and benign applications. Second, a feature generalization method is adopted to improve the robustness of the fingerprint. Finally, for indistinguishable fingerprints, the target host information characteristic, combined weighted Bayesian is used to distinguish different benign applications and malicious families.

In the network function virtualization environment, a low-cost deployment algorithm is proposed for the deployment cost of the service function chain that guarantees high availability. The algorithm makes full use of VNF reuse technology to reduce the total cost of deployment, and divides the deployment process into two stages. In the first phase, a meta-heuristic algorithm was designed to initially deploy the SFC, and in the second phase, a VNF backup algorithm based on the Availability Critical Importance Parameter (ACIP) was proposed to ensure high service availability. Simulation results show that the proposed algorithm can effectively reduce the total deployment cost and ensure high service availability.

In order to solve the problem that the traditional method of communication transmitter individual identification ignores the nonlinearity, shallow features and timing characteristics of the signal, this paper proposes a method of communication transmitter individual identification based on GAF-ResNet. This method uses the Gramian Angular Field to convert timeseries signals into two-dimensional images of time-domain signals, which takes better advantages of the dependence and correlation of time series. The residual block of the residual network is used to realize the fusion between shallow features and deep features. The experimental results show that the best network model trained by this method has a recognition rate of 92.7%, which proves the effectiveness of the method.

In recent years, in the field of internal network threat analysis, some scholars have proposed a user behavior evaluation method based on analytic hierarchy process (AHP)^[1]. The method is mainly to give weight to the user behavior evidence after standardization, and realize the evaluation of user behavior according to the weight. In the early stage, some scholars assigned the weight of evidence by expert evaluation method, but it is difficult to avoid the problems of excessive subjectivity and unable to adapt to data changes. This paper combines the reliefF feature selection algorithm with the weight assignment in user behavior evaluation to ensure that the weight is matched with the data and obtain the evidence weight that meets the actual requirements.

Aiming at the problems of distortion, low quality, and multiple transmissions of generative stego-image based on generative adversarial networks, a new generative image steganography method combing variational autoencoders (VAE) and generative adversarial networks (GAN) is proposed. This method combines the advantages of VAE and GAN. First, the binary secret information is grouped, and flag bits are added, which is converted into noise according to the mapping relationship. The noise is input into the VAE-GAN generator to obtain a high-quality stego-image group and randomly combine a new image. When extracting information, the stego-image is input to the extractor to recover the noise and converted into a binary message according to the mapping relationship. Finally, the secret information is recovered according to the flag bit. Experimental results show that the method not only improves the quality of stego-image but also has a high steganography capacity.

The traditional polling system in Internet of Things (IoT) communication suffers from low fairness under high load, and difficulty in adjusting the scheduling strategy according to the network state, and it is difficult to achieve the desired control effect by the traditional polling method. To overcome these problems, we propose a polling system based on Deep Q-Network (DQN) adaptive network load by combining deep reinforcement learning. In addition, the deep reinforcement learning approach is used to solve the high-dimensional complex problems, which are difficult to solve in the polling system model. The system is simulated for experiments, the queue length at polling epoch and mean waiting time of site information of the polling system were counted, to verify the accuracy of the method in learning to control the service strategy of each site. And a comparison experiment was conducted, demonstrates the system's superiority in controlling queue message latency and service fairness.

In order to achieve efficient identification of unknown protocol message types without a priori knowledge, this paper proposes a hierarchical density-based clustering method for unknown protocol messages. The method effectively combines the characteristics of the hierarchical density clustering algorithm, which reduces the average time complexity while improving the accuracy of the message clustering results. We conduct experiments using public data sets and compares them with other previous methods. The results show that the proposed method can reduce the time spent on clustering while ensuring the quality of the clustering results, which verifies its effectiveness.

With the development of the new generation mobile communication technology, the application of network function virtualization (NFV) technology has become a boom. With NFV, virtual network functions (VNFs) are deployed in the form of service function chain (SFC), so how to place VNFs in the best position has become a challenge. This paper studies this challenge to maximize resource utilization and minimize energy cost. Specifically, a two-stage deployment strategy (TSBA) is designed. Firstly, a matching algorithm (MA) based on weighted graph matching is proposed to improve resource utilization. Then the idle server is shut down by VNF migration method to reduce the cost of server energy consumption. The experimental results show that the two-stage deployment strategy proposed in this paper effectively improves the resource utilization and reduces the energy consumption compared with the existing methods.

In the Multimedia Mobile Edge Computing (MMEC) environment, cryptography technology is used to ensure secure transmission of multimedia data when Terminal accesses to MMEC or Terminal accesses to the central cloud. However, the problem of the participants’ identity authentication and integrity measurement can not be solved. Therefore, in this paper, a novel remote attestation model and protocol is proposed, which introduces mutual anonymous identity authentication, mutual integrity measurement and verification between Terminal and MMEC as well as between MMEC and cloud. Moreover, the proposed protocol guarantees data security and identity security of the platform, as well as integrity state maintenance when MMEC interacts with Terminal. We make a thorough security analysis and performance simulation over the proposed protocol. Security analysis shows that the proposed protocol is superior to the other two similar schemes in terms of privacy protection ability and anti-attack ability. Performance simulation results demonstrate that the remote attestation protocol with non-intensive requests state achieve much higher efficiency.

This paper analyses the research status and progress of location privacy protection technology in sensor networks. Firstly, the attack model and performance evaluation model are introduced. Then, the existing research results are classified according to four strategies: path camouflage, trap guidance, network anonymity and communication control, and the core technologies of representative protocols are described. Finally, the performance of each protocol is compared and summarized, and the future research direction is prospected.

The optical structure suitable for the on-line detection of high-voltage equipment is designed, the on-line analysis of trace decomposition gas composition of insulating gas SF6 of the high-voltage equipment is realized, and the prototype is developed. The optical noise suppression technology suitable for the field is studied, the optical path system is designed to withstand the gas pressure of high-pressure equipment, and the influence of field vibration and temperature change of the high-pressure equipment on the optical path structure is solved. The judgment threshold of expert diagnosis system is formed to predict the operation state and fault type of GIS equipment according to the changes of types and contents of typical decomposition products of the sulfur hexafluoride. The above achievements lay the theoretical foundation for realizing the on-line monitoring of the insulation status of SF6 Electrical equipment, greatly promoted the development and application of ultraviolet spectrum and non dispersive infrared optical live sensing detection technology, provided effective guarantee for the safe and stable operation of the sulfur hexafluoride electrical equipment, further improved the reliability of equipment, and ensured the ability of power grid to safely and continuously provide clean energy, ensure reliable power supply and good social benefits.

In the network security test environment, due to the destructive nature of network adversarial behavior and uncontrollable boundaries and other characteristics, it is easy to lead to security risks in the event capture process. A real-time and secure event capture method for network security tests is an effective guarantee of the authority of the exercise results, and is also a pressing challenge for the key applications of network security tests. To this end, the article investigates the event capture technology for network security tests in a trusted execution environment, and designs a secure and efficient event capture method based on SGX technology, Elastic Stack technology stack and virtualization technology tools to ensure the authenticity and confidentiality of the original data in the network security test environment, which can realize the secure and The method enables secure and efficient capture of virtual host behavior logs in the test environment.

With the advancement of digital transformation, digital relay protection devices have replaced traditional protection devices and become an important part of the stable operation of the power system. The IEC 61850 standard is the basic protocol for real-time communication and data exchange between key substation equipment and is widely used in substation automation equipment and protection devices. However, IEC 61850 has a man-in-the-middle attack security risk. This paper studies the network attack of IEC 61850 Generic Object-Oriented Substation Event (GOOSE) protocol. By exploiting protocol loopholes to achieve penetration attacks, forged GOOSE data frames are injected into the interval layer of the substation communication network, and sent to multiple relay protection devices at the same time to carry out coordinated network attacks, which can cause multiple relay protection devices to incorrectly issue trip commands , resulting in cascading failures in the grid, which eventually lead to power outages. Finally, simulation experiments are carried out through hardware loop simulation and real-time digital simulator (RTDS) to verify the effectiveness of the attack simulation.

The typical structure of SF6 gas insulation is cylindrical. The research on the performance of its typical insulation structure has a good guiding significance for the design and optimization of insulation structure. In this paper, the electro-thermal model considering the temperature non-linearity of SF6 gas insulation structure and its thermal stability are analyzed in detail. Including the nonlinear relationship between material parameters and temperature of high voltage insulation structure. Analytical calculation of current distribution in the tubular conductor. Analysis of transient temperature distribution of cylindrical insulation structure. Analysis of thermal stability and analysis of current carrying capacity of cylindrical insulation structure. This paper mainly deals with the test and numerical fitting analysis of the nonlinear relationship of the performance parameters of insulation materials, as well as the formula deduction process in the theoretical level, and analyses the relationship between the electric field and the thermal field more deeply, and provides a certain theoretical basis and guidance for the insulation structure design and the accident analysis in the power system.

This research focuses on the topic of image super-resolution based on a deep network prior. This research uses a U-Net-like network structure and modifies the network's hyperparameters, such as the number of downsamples, to find a model with better performance using a control variable approach. Then this best model was applied to different classes of images, and the performance of the network model was investigated by measuring the effect of super-resolution on these classes of images. This research hopes to understand the network structure's role on image super-resolution through such exploratory experiments.

Handwritten numbers play a huge role in human production and life. At present, how to automatically recognize handwritten digits accurately and efficiently has become a very realistic research problem. This article takes the handwritten digits on the blackboard as the research object, and introduces the convolutional neural network model and the Resnet model. Furthermore, a kind of recognition of handwritten digits on blackboard based on Resnet model is proposed, which can improve the recognition rate of handwritten digits by adding residual module to realize residual learning. Based on the above theory, the Resnet model is constructed using the python programming language, and the model is trained through the public data set MNIST. Finally, the model is applied to recognize the handwritten digits data set of blackboard writing, and the corresponding digits are successfully recognized.

In recent years, wireless body area networks have developed ubiquitously and universally. The technology of using the human body as a communication medium to communicate through capacitive coupling has become a research hotspot. In order to explore the transmission characteristics of human body communication with different body mass indexes, this paper is based on the FDTD method, using homogeneous and inhomogeneous tissue human body models to simulate human body communication channels with different body mass indexes. It can be drawn from the simulation results that the loss of human body communication on the surface increases with the increase of the transmission distance. At the same time, in vivo experiments were carried out with reference to the simulation model. We used Maximum Likelihood Estimation (MLE) and Akaike Information Criterion (AIC) to analyse the normalized received power, and then used different fading models to perform statistical analysis on it. At 50MHz operating frequency, the channel loss increased with distance, and the increase in body index would reduce transmission loss. Comparing simulation and measurement results, we found that the attenuation trends of the two are not consistent. Through measurement and transmission of ASK data, the average BER of surface human body communication was ≤10^-8, which verified that the higher BMI, the better performance of the human body communication channels. Based on the above results, this paper initially verified the feasibility and individual differences of the surface human body communication channels to transmit information.

In this paper, the adjustment strategy and signal acquisition ability of the active reflection panel of 500m aperture spherical radio telescope are studied. Firstly, based on the fact that the ideal paraboloid is rotating, the geometric model is established, the geometric elements of the parabola are solved combined with the geometric properties of the parabola, and the ideal paraboloid equation is obtained. The focal distance is 280.053m, and the vertex coordinates are (0,0,-300.440). Then, rotate the original coordinate system and establish a new object coordinate system so that the measured object is located in a similar special position. The parabolic focal distance is 280.053m, and the vertex coordinates are (-49.327,-36.894,- 294.058). Finally, the programming model is established, the expansion and contraction of the relevant actuator are taken as the decision variable, and the sum of the absolute values of the deviation between the main cable node of the reflective panel and the ideal paraboloid is taken as the objective function. With the help of a genetic algorithm, the programming model can be solved to obtain the expansion and contraction of the relevant actuator, and then the coordinates of the main cable nodes of the paraboloid participating in the adjustment in the basic coordinate system can be solved according to the principle of coordinate system transformation.

This article first establish a directional network. And calculate the number of child nodes corresponding to each node, add other parameters, and calculate the music influence model of each artist through the entropy weight method (EWM). Through this evaluation model, we can quantitatively describe the influence of an artist. Then this paper uses the cosine similarity algorithm to model the music similarity using the theme model. Use the cosine similarity algorithm to calculate the cosine similarity in the space of any two songs, and then judge the similarity. Based on the data set, this paper calculate the influence of each characteristic of all music on the popularity of music, and get the influence of music characteristic on "infectivity". Through the radar chart of the musical characteristics of the artist Son House and his three typical influenced people, it is analyzed whether the artist can influence the creative tendency of his influenced people. With the help of the "elbow method" in the clustering determination k, the inflection point corresponding to the elbow is selected, and the inflection point represents the changer. Select the Electronic genre and divide the genre into three stages with the help of the total amount of music. Analyze the mean value of certain music indicators in these three stages, and use the change of the mean value of these music indicators to explore the evolution of the genre over time.

This paper analyzes the self-service mode of "taking pictures to make money" and studies the most important and core problem in this mode-task pricing. Through qualitative and quantitative analysis, combined with the actual influence of geographical location, we give the scheme to achieve a higher task completion rate at less cost than the original pricing method. First of all, through the statistical analysis of GPS data, it is concluded that each task presents a multi-center aggregation trend. By referring to the form of the uniform distribution density function of two-dimensional random variables, the dynamic expectation function is constructed by imitating the definition of mathematical expectation in probability theory, and the correlation between each expectation and pricing is analyzed to verify the conjecture. Next, inspired by the idea of grey analysis, the calculation method of task expectation and member expectation is extended, and each task position's difficult expectation and quota expectation are calculated. Finally, according to the differential form of logistic model, the pricing strategy is determined by connecting with the microbial growth curve function. Compared with the original scheme, the task cost decreased by about 0.1%, and the completion rate increased by 12.5%.

In the stowage optimization problem of general cargo ships loading in one port and unloading in multiple ports, only thinking over the maximum total loading volume may lead to higher stevedoring complexity. In addition, the two-dimensional cargo ship stowage diagram is difficult to describe the loading rules and stevedoring complexity of stacked general cargoes. In this paper, by means of three-dimensional stowage, an algorithm for calculating the sum number of the reloading operations of the general cargo ship during unloading in multiple ports is proposed. The optimization objective is to maximize the total volume of cargo hold loading and minimize the sum number of the reloading operations of the cargo hold during unloading in each port. A three-dimensional stowage optimization model of a general cargo ship is established by combining the general cargo loading sequence with the loading optimization rules. A genetic algorithm improved on mutation operator solves the model by coordinating and optimizing the general cargo loading sequence. Through example verification, the model can obtain a general cargo ship stowage plan that satisfies the large total volume of loading and the insignificant sum number of the reloading operations during unloading, proving the algorithm model's effectiveness.

Predicting C4 olefin output and optimising operating variables in the production process are of great practical significance for China's petrochemical enterprises to reduce production costs and improve the quality of petrochemical products. Firstly, this paper uses the nonlinear least square method (PLS), combines the advantages of statistical regression and neural network, and uses nonlinear partial least square regression method based on a neural network to establish the nonparametric model of catalyst and temperature on ethanol conversion rate and C4 olefin selectivity. Based on the nonlinear partial least squares regression method based on neural network, the random forest model is constructed to construct decision trees for different catalysts and temperatures. Next, a chaotic particle swarm optimization algorithm is constructed to optimize the parameters. The discrete particle swarm optimization algorithm is creatively designed and initialized by the label propagation method, and the feeding scheme for optimizing product quality is successfully provided, and the optimization result is good.

This paper is research on the ordering and transhipment scheme of raw materials in production enterprises. The order and purchase data of 402 raw material suppliers and the transport loss rate data of 8 transporters were collected and analysed. First of all, from the three dimensions of hard power (based on the "Q.C.D.S" principle), soft power and potential, the relevant information of seven indexes such as just-in-time delivery rate, coefficient of variation of supply quantity and proportion of supply quantity is extracted to analyse its supply characteristics quantitatively. Then an improved TOPSIS model based on grey relational analysis is established to comprehensively evaluate the 402 suppliers to ensure the normal production of the enterprise, from which the list of the most important suppliers can be obtained. With the help of the K-means clustering algorithm, we screen the suppliers whose supply proportion and average delivery just-in-time rate are good at the same time and determine that the minimum number of suppliers is 23 (at this time, the two indicators are about 80%). Then, to dig deeply into the raw material supply of suppliers in the next 24 weeks, we build time series models (ARMA, ATLSTM) to forecast. Based on this, we establish the 01 planning model, which minimises capacity gap and cost as the objective function to determine the best ordering plan.