一种融合多维信息的电力物联网设备指纹识别技术
网络安全与数据治理
卢列文1,刘佳烨2,杨盛明1
1.工业和信息化部电子第五研究所; 2.中国人民解放军93131部队
摘要: 针对电力物联网无算力设备面临的恶意篡改、非法接入等安全问题,以及现有设备指纹技术在复杂环境下适应性不足,提出一种融合多维信息的设备指纹识别技术,提升设备识别安全性与准确性。方法上,融合硬件标识、通信协议特征及运行状态数据,通过自适应哈希编码、注意力增强编码、窗口算法等构建动态指纹模型,并与主流算法对比验证。研究结果表明,该技术在正常及异常网络环境和电力运行状态下,准确率、召回率和F1值均较高,适配无算力设备,且能适应一定程度的环境变化。该技术为电力物联网无算力设备的精准识别和安全管理提供了有效方案,未来可结合新技术进一步优化,以适应更多新场景。
中圖分類號:TP309.2文獻標識碼:ADOI:10.19358/j.issn.2097-1788.2025.09.001
引用格式:盧列文,劉佳燁,楊盛明. 一種融合多維信息的電力物聯網設備指紋識別技術[J].網絡安全與數據治理,2025,44(9):1-7.
引用格式:盧列文,劉佳燁,楊盛明. 一種融合多維信息的電力物聯網設備指紋識別技術[J].網絡安全與數據治理,2025,44(9):1-7.
A fingerprint recognition technology for power IoT devices integrating multidimensional information
Lu Liewen1,Liu Jiaye2,Yang Shengming1
1. China Electronic Product Reliability and Environmental Testing Research Institute; 2. 93131 Troops of the Chinese People′s Liberation Army
Abstract: To address security issues such as malicious tampering and illegal access faced by computing-power-free devices in the power Internet of Things (IoT), and existing device fingerprint technology lacks adaptability in complex environments, this paper proposes a device fingerprint recognition technology that integrates multi-dimensional information to enhance the security and accuracy of device recognition. Methodologically, it combines hardware identifiers, communication protocol characteristics, and operational status data to construct a dynamic fingerprint model through adaptive hashing, attention-enhanced encoding, window algorithms, and other techniques, and verifies its performance through comparison with mainstream algorithms. The research results show that this technology achieves high accuracy, recall rate, and F1 score in both normal and abnormal network environments and power operation states. It is suitable for computing-power-free devices and can adapt to a certain degree of environmental changes. This technology provides an effective solution for accurate identification and security management of computing-power-free devices in the power IoT. In the future, it can be further optimized by combining new technologies to adapt to more new scenarios.
Key words : power Internet of Things; computing-power-free device; device fingerprint; identification technology; modeling; experimental simulation
引言
電力物聯網作為智能電網發展的關鍵支撐,正逐漸改變著傳統電力系統的運營模式[1]。在電力物聯網架構中,無算力設備(通常是指不具備或僅具有極弱數據處理、計算能力的設備),如電表、數據終端單元(Data Terminal Unit,DTU)、饋線終端單元(Feeder Terminal Unit,FTU)、配電變壓器監測終端(Transformer Terminal Unit,TTU)等,這類設備數量龐大且分布廣泛,它們負責采集、傳輸電力數據,是電力物聯網實現智能化的基礎。然而,隨著電力物聯網的規模不斷擴大,這些設備面臨著嚴峻的安全挑戰,如設備被惡意篡改、非法接入等問題,嚴重威脅電力系統的安全穩定運行。
設備指紋識別技術作為一種有效的設備管理和安全防護手段,能夠為每臺設備生成獨一無二的身份信息,實現對設備的精準識別和追蹤。通過準確識別無算力設備指紋,對于及時發現非法設備接入,提高設備管理的精細化程度,構建電力物聯網安全防御體系具有重要意義。
本文詳細內容請下載:
http://www.tom3567.com/resource/share/2000006700
作者信息:
盧列文1,劉佳燁2,楊盛明1
(1.工業和信息化部電子第五研究所,廣東廣州511300;
2.中國人民解放軍93131部隊,北京100068)

此內容為AET網站原創,未經授權禁止轉載。
