後量子混合式密碼系統中H.264選擇性加密技術;A Hybrid Cryptosystem Based on Post-Quantum Cryptography Selective Encryption for H.264

NCU Institutional Repository > 資訊電機學院 > 資訊工程研究所 > 博碩士論文 > Item 987654321/95816

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/95816

题名:	後量子混合式密碼系統中H.264選擇性加密技術;A Hybrid Cryptosystem Based on Post-Quantum Cryptography Selective Encryption for H.264
作者:	陳彥文;Chen, Yan-Wen
贡献者:	資訊工程學系
关键词:	進階視訊編碼;混合式加密系統;後量子密碼學;進階加密標準;Advanced Video Coding;Hybrid Cryptosystem;Post-Quantum Cryptography;Advanced Encryption Standard
日期:	2024-08-15
上传时间:	2024-10-09 17:18:23 (UTC+8)
出版者:	國立中央大學
摘要:	在影片傳輸或視訊軟體上，經常使用各種加密技術來確保資料安全性。由於傳輸的資料量過於龐大，加密所耗費的時間隨之成長，因此當網路環境或硬體設備不佳時，會造成嚴重的卡頓與延遲，使得影片與視訊的安全性增加但卻造成使用者上許多困擾。本論文提出了4個用於選擇性加密影片與視訊的演算法，能應對各種不同的情境，也能最佳化影片與視訊的安全性與資源使用，同時不會造成過多的資源耗費。為了最佳化進階視訊編碼（Advanced Video Coding, H.264/AVC）的影片或視訊的安全性與資源使用，本論文提出了4種Encryption Frames by Security Selection（EnFrames-SS）選擇性加密演算法，皆使用了混合式加密系統（Hybrid Cryptosystem），並透過後量子密碼學（Post-Quantum Cryptography, PQC）中的CRYSTAL-KYBER來產生共享金鑰，此金鑰被使用於進階加密標準（Advanced Encryption Standard, AES），並且對AES進行了優化來降低加解密所需的耗時。在影片與視訊選擇性加密上使用了一種新穎的方式，根據參考幀（Reference Frame）的數量以及有無被加密來判斷當前幀的安全性，並根據被參考的次數來決定當前幀的重要性，使用幀的安全性與重要性來進行選擇性加密。透過這種方式能減少許多資源的耗費，同時確保影片與視訊能保有較高的安全性。本論文使用了3個判斷加密成效與效能的指標，與非選擇性加密演算法比較。透過這些指標可以得知，4種EnFrames-SS演算法可以獲得較高的效能，並且當參考幀的數量越多時，採用這4種EnFrames-SS演算法會更接近非選擇性加密的安全性，而整個加密系統所花費的時間差距會隨著解析度（Resolution）、Frames Per Second（FPS）、長度等因素漸漸變大。由實驗結果得知，當使用640x360的測試影片時，EnFrames-SS4演算法最高可透過加密1幀來提高14.60幀的安全性，而非選擇性加密演算法僅提高了11.25幀的安全性，而4種EnFrames-SS演算法進行資料重傳的成本也低於非選擇性加密演算法。 ;In video transmission or video software, various encryption techniques are often used to ensure data security. However, due to the large amount of data being transmitted, the execution time for encryption increases accordingly. This can result in video freezing and delays, especially when the network environment or hardware is suboptimal. While increasing the security of the video, it also causes many inconveniences for users. Therefore, this paper proposes four algorithms for selectively encrypting video. These algorithms can handle various scenarios and optimize the security and resource usage of video without causing excessive resource consumption. To optimize the security and resource usage of Advanced Video Coding (H.264/AVC) video, this paper proposes four Encryption Frames by Security Selection (EnFrames-SS) algorithms. The four EnFrames-SS algorithms all use a Hybrid Cryptosystem. The shared key is generated using CRYSTAL-KYBER from Post-Quantum Cryptography (PQC), which is used with the Advanced Encryption Standard (AES). AES has been optimized to reduce the execution time for encryption and decryption. A new method for selectively encrypting video is introduced, which determines the security of the current frame based on the number of referenced frames and whether they are encrypted, and the importance of the current frame is determined by the number of references from other frames. Selective encryption is performed based on the frame security and importance. In this way, it can reduce resource usage while ensuring high security for video. This paper uses three metrics to evaluate encryption effectiveness and performance. Compared to non-selective encryption algorithm, these metrics show that the four EnFrames-SS algorithms achieve higher performance. As the number of referenced frames increases, the four EnFrames-SS algorithms approach the security level of general encryption, and the execution time difference for the encryption system grows with factors such as resolution, frames per second (FPS), and length. From the experiment results, it is known that when using a test video with a resolution of 640x360, the four EnFrames-SS algorithms can increase security by encrypting up to 14.60 frames, whereas the non-selective encryption algorithm only increased security by 11.25 frames. The cost of data retransmission for the four EnFrames-SS algorithms is also lower than that of non-selective encryption algorithm.
显示于类别:	[資訊工程研究所] 博碩士論文

文件中的档案:

档案	描述	大小	格式	浏览次数
index.html		0Kb	HTML	47	检视/开启

在NCUIR中所有的数据项都受到原著作权保护.

社群 sharing

数据加载中.....