Keywords

1 Introduction

With the rapid development of information technology and the in-depth improvement of new military reform, the era of information war has entered. Information war has the following main characteristics: the dominant element of combat power changes from material energy to information energy; the winning idea of war has changed from entity destruction to system attack; the release mode of combat effectiveness has changed from quantity accumulation to system integration; the range of battlefield space becomes full dimensional. Compared with the traditional technology, the new generation information and communication technology has lower delay, Edge computing solves the problem of data volume and time delay. It is the platform integrating the key capabilities of application, storage and network.

Edge computing and Quantum technology can greatly improve the intelligence capacity. First, it greatly improves the efficiency of intelligence information process. In modern war, the amount of battlefield datalake is largely huge unstructured data. If we use conventional methods to process these massive information. Using big data to process intelligence information, the theoretical time-consuming can reach the second level and the processing speed jumps exponentially, which can greatly improve the intelligence information acquisition and processing ability. Second, more valuable information can be founded. Under the constraints of investigation means, battlefield environment and other factors, the technology can quickly and automatically classify, sort, analyze and feed back the information from multiple channels. It separates the high-value military intelligence of the target object from a large number of relevant or seemingly unrelated, secret or public information to effectively solve the problem of intelligence Insufficient surveillance and reconnaissance system. Third, it can improve command and decision-making ability. The use of big data analysis technology can provide intelligent and automatic auxiliary means for the decision analysis, it improve the intelligent degree of the system and effectiveness of decision-making, so as to greatly improve the command efficiency and overall combat ability.

2 Characteristics of Edge Calculation

Edge Calculation defines three domains including device domain, data domain and application domain. The layers are the calculation objects of edge calculation. Device domain establishes TPM (trusted platform modules), which integrates the encryption key in the chip into the chip that can be used for device authentication in the software layer. If encode/decode of non shared key path occurs in TPM, the problems can be easily solved. Data domain e communicates with more edge gatewayswhich provide access to the authentic network. Application domain realizes interworking through Data domain or centralized layer. Edge computing is nearby the data source, it can firstly analyze and intelligently process the data in real time, which is efficient and secure. Both edge computing and cloud computing are actually a processing method for computing and running big data. Connectivity and location in Edge computing is based on connectivity. Because of the various connected data and application scenarios, edge computing is required to have rich connection functions.

When the network edge is a part of the network, little information can be used to determine the location of each connected device. It realizes a complete set of business use cases. In the interconnection scenario, edge gateways provide security which constraints and support the digital diversity scene of the industry.

High bandwidth and low delay of edge computing is nearby the datalake, simple data processing can be carried out locally. Since the edge service runs close to the terminal device, the delay is greatly reduced. Edge computing is often associated with the Internet of things which participate in a large amount of data generated network.

Distribution and proximity in Edge computing. Because edge computing is close to the data receiving source, it can obtain data in real time, analyze and process, In addition, edge computing can directly access devices, so it is easy to directly derive specific commercial applications. Integration and efficiency in edge computing distance is close, and the data filtering and analysis can be realized. With the real-time data, edge computing can process value data. On the other hand, edge computing having challenges including real-time data and collaboration data.

3 Information System Architecture Design Based on Edge Computing

According to the operational needs, the system dynamically connects various warning radar, reconnaissance satellite, aerial reconnaissance and message, image, video, electromagnetic. Depending on the supportive requirements, the information products are sent to the authorized users at different levels such as the command post according to the subscription and elationship formulated by the useras shown in Fig. 1.

Fig. 1.
figure 1

Overall architecture of military intelligence analysis and service system

Full size image

The support layer is the basic layer of the overall architecture providing a platform and business support environment for intelligence big data analysis and processing and service-oriented applications. It includes platform support and application support. The platform support part provides a platform environment for system construction and operation, including service-oriented support environment, data storage, distributed infrastructure, cluster computing environment and storm stream processing environment. The service-oriented support environment supports system development with a service-oriented architecture. The data storage module is used to support the storage and management of massive intelligence data resources. Storm big data processing frameworks provide a distributed parallel processing environment for massive big data. The application support part provides basic business support for the construction and operation of the system, and it provides common function module support for the service layer and application layer, including basic services such as data preprocess, image analysis, message analysis, audio analysis, video analysis, electromagnetic analysis, association mining, timing analysis, semantic analysis, knowledge reasoning and so on.

Application Layer is a cost-effective edge computing gateway launched by inhand for the field of industrial device. With a variety of broadband services deployed worldwide, the product provides uninterrupted interconnection and connection available everywhere. It supports many mainstream industrial protocols. At the same time, it can connect with many mainstream cloud platforms so that field devices can be easily put into the cloud; It has an open edge computing platform, supports user secondary development, and realizes data optimization, real-time response and intelligent analysis at the edge of the Internet of things. The excellent product features, easy deployment and perfect remote management function help enterprises with digital transformation. It is used to transmit equipment or environmental safety warning information. If not avoided, it may lead to equipment damage, data loss, equipment performance degradation or other unpredictable results. As shown in the Fig. 1, the upper layer is application deployment, which is mainly responsible for deploying edge applications and creating an edge ecosystem of APP/vnf. The middle layer is edge middleware and API, creating standard edge platforms and middleware, and unifying API and SDK interfaces. The bottom layer is the layer which interfaces with the open source edge stack. This is mainly to solve the problem of weak network and restart. Even with network tunneling, the fact that the network instability of edge nodes and the cloud cannot be changed, and there is still constant disconnection. The edge autonomy function meets two edge scenarios. The network is disconnected between the center and the edge, and the service of the edge node is not affected. The edge node is restarted. After the restart, the services on the edge node can still be restored.

4 Characteristics Analysis Performance

According to the principles of distributed organization management and unified resource sharing, the system adopts distributed operation management technology to uniformly control information analysis tasks, computing power and data resources, realize collaborative scheduling according to information support requirements, and jointly complete information analysis tasks. Using the service-oriented architecture, the core intelligence analysis function, image intelligence analysis service, message intelligence service, open source intelligence analysis service and intelligence data service carries out unified classification management based on the service registration mechanism to form service resource directory. Realize the sharing of intelligence analysis function among nodes in the system.

Real time aggregation of trajectory data. At present, the terminal perceives the real-time access of collected data and comprehensively obtains all kinds of travel data. Established a special analysis model, it masters the trajectory of key areas, and realizes the real-time analysis, research and judgment of intelligence information. The platform includes visual intelligent track analysis and query, research and judgment analysis of abnormal activities, intelligent statistical analysis, dynamic monitoring, analysis and early warning, intelligent information retrieval and other functions which can produce obvious results in a short time.

Closed loop operation of early warning information. Early warning information is synchronously pushed to the public security organs in the control and early warning places, realizing information sharing, breaking the information barrier, and realizing the closed-loop operation of early warning, research and judgment, verification, feedback and other links. Focused on gathering and integrating all kinds of social data,it can play an important role in operations, intelligence research and judgment, carefully study the conversion and processing of all kinds of data, gives full play to the cross secondary comparison of data, and improves the effective utilization of data.

Early warning synchronous mining analysis. Analyze and mine the key tracks and key personnel in the same category and region, and provide stability control suggestions for intelligence work at all levels. The platform has realized the downward extension of system construction and the upward aggregation of data resources, forming a four-level information platform linkage application system; At the same time, it provides platform support for joint operations and cooperation. It provides a strong guarantee for synthetic operations.

5 Summary

This paper proposes an information system architecture based on edge computing. It introduces the advantages of each layer of the system. The system can better complete the cloud edge end collaborative network computing and solve the flow control layer by layer. Because the node location and end-to-node delay are divided into different levels, the traffic volume to be carried by nodes at different levels is different. The capabilities and technical points to be provided are also different. Edge computing needs to solve the following key problems: Resource management and protocol analysis: 1. provide the connection and communication between local devices, realize the local exchange of massive data, provide the ability to adapt and normalize different devices, shield the differentiation of industrial protocols. Storage and forwarding device can provide relatively complete functions of data acquisition, processing, analysis and alarm when the real-time requirements are high, the amount of data transmission is too large or the network connected to the platform is unavailable. At the same time, the local provides a certain storage capacity, which can forward the data to the platform during network recovery. Platform integration realizes comprehensive collaboration with the platform end, flexible data acquisition and distributed computing functions for the decision center at the platform end. It can support seamless running of applications and can be uniformly configured rather than manual compiling and developed programs.