CN102123026A - Chaos and hyperchaos based two-level video streaming media encryption method - Google Patents

Abstract

A two-level video streaming encryption method based on chaos and hyperchaos. The design method provided by the invention includes: proposing a two-level video streaming media encryption algorithm based on chaos and hyperchaos. This method is combined with the video compression standard H.263 under network transmission. The first level is: based on the Logistic chaotic map, the video stream data is encrypted with a selective encryption algorithm based on the discrete cosine transform (DCT) during the video compression process. ; After video compression, the second level of encryption is carried out, which includes two parts: video stream data scrambling based on discrete Baker mapping and data confusion for scrambled data based on hyperchaotic mapping; due to the use of multi-level encryption , and the hyperchaotic system has strong encryption performance, so the video stream encryption system using this method can effectively resist malicious attacks and professional decryption. This encryption method has important application value in the field of network streaming media transmission information security.

Description

A Two-Level Video Streaming Media Encryption Method Based on Chaos and Hyperchaos

【Technical field】：

The invention belongs to the fields of information security technology and network communication technology, relates to an algorithm for encrypting video streaming media transmitted in real time in a network, and provides a safe and effective encryption solution for safe and confidential transmission of network video information and protection of related intellectual property rights.

【Background technique】:

Today's society is a highly informationized society, and information security issues are becoming more and more important to the national economy, social development and people's lives. In information security issues, the security technology of network multimedia occupies a very important position. With the rapid development and wide application of computer technology, network technology and multimedia technology, it provides great convenience for people to obtain and exchange information, and also contains huge commercial benefits. How to protect the security of multimedia information has become an international research hotspot.

With the rapid development of computer network technology and hardware equipment, online video streaming media is also becoming more and more popular, such as our common online video service websites, confidential video conferencing, pay TV, online movies, and specialized commercial or military video transmission systems. The mutual penetration and mutual promotion of computer technology, network communication technology and multimedia technology has made the development of network streaming media technology reach a critical moment from quantitative change to qualitative change. The information technology transmitted by the network is getting higher and higher, from simple text information to the current text, image, sound, video, animation and almost all kinds of information. The continuous improvement of network bandwidth, the continuous development of data transmission coding technology and server technology for publishing media are all used to meet the transmission technical requirements of network streaming media data. However, due to the openness, sharing, and dynamics of the Internet, the security of network information has been seriously threatened and disturbed, and there are security problems such as illegal attacks, tampering, downloading, and piracy. Therefore, the security of network information, especially the confidentiality of multimedia data, has become a major issue related to economic development and social security, which has attracted people's attention. Due to the complex characteristics of video data, the high real-time requirements of video files, and the huge amount of video data, the traditional encryption algorithm for text files designed based on algebraic cryptography theory system is no longer suitable for video encryption. Therefore, it is necessary to use new theories and tools to design an information encryption system that can be applied to streaming media.

"Chaos" is recognized as one of the most important scientific discoveries of the last century. As a cutting-edge interdisciplinary subject with broad application prospects, it has been successfully applied in many practical fields of engineering technology and e-commerce, and has been applied in biological and medical engineering, mechanical engineering, electronic engineering, chemical engineering, information engineering, computer engineering, Laser technology, applied physics and many other fields have great application potential. At present, the most active application field of chaos theory is the field of communication, involving secure communication and synchronization of data, image encryption, digital watermark hiding, streaming media encryption and many other aspects. Due to some inherent characteristics of the chaotic system, such as the complex randomness of the chaotic sequence, difficulty in analysis and prediction, and extreme sensitivity to the initial value, it is very suitable for the encryption of information. However, the existing chaotic encryption algorithms and systems mainly have the following problems:

(1) Most existing chaotic encryption algorithms are designed for static objects such as images and watermarks, and are not suitable for online encryption of streaming media information.

(2) Existing chaotic video encryption algorithms mostly use selective encryption algorithm, the core idea of which is to encrypt only part of the data information. After the video stream information is processed, some important parts of the information are selected for encryption. Although the real-time performance is guaranteed, this kind of selective encryption algorithm has many defects from the perspective of security and is easy to be attacked and deciphered.

(3) The design of chaotic encryption algorithm is mostly based on simple chaotic system, and simple chaotic system is easier to be deciphered by mature nonlinear data processing technology.

In view of the above problems, it is necessary to propose a new encryption algorithm for multimedia video stream data encryption based on chaos theory. The video encryption system established based on the new algorithm is applied to media video transmission. It must not only have good transmission performance, such as real-time, Compression ratio and image transmission quality, and high security. The usual selective encryption algorithm only encrypts part of the information, and the defense measures are relatively weak, which is not enough to resist the attacks of malicious attackers. Therefore, it is necessary to establish multiple levels of encryption measures. The chaotic encryption algorithm only uses simple chaos, so a more complex chaotic system-hyperchaos is needed. The complex time series produced by the hyperchaotic system greatly improves the encryption quality and is very difficult to decipher. Therefore, applying hyperchaos theory to information encryption has an important application prospect.

【Invention content】：

The purpose of the present invention is to overcome the above-mentioned problems existing in the prior art, provide an innovative two-level video streaming media encryption method based on chaos and hyperchaos, and provide a feasible solution for realizing the security transmission encryption system of video streaming media.

The two-level video streaming media encryption method based on chaos and hyperchaos provided by the present invention, its main contents include:

1. The structure of the two-level video stream data encryption scheme based on chaos and hyperchaos

A two-level video stream data encryption scheme based on chaos and hyperchaos is proposed. The design of the encryption system is combined with the video compression standard H.263 under network transmission, and the encoded H.263 is processed by byte-by-byte processing mode. stream, the first level is to encrypt the video stream data based on the selective stream encryption based on the chaotic map during the video compression process; Scrambling and data confusion, where block scrambling adopts two-dimensional discrete Baker mapping, and then uses hyper-chaotic system for data confusion on the scrambled data;

2. The first level video stream data encryption scheme based on chaotic map

The encryption scheme of the first level is: in the video compression process, the video stream data is selectively encrypted based on the chaotic map, that is, a part of the video data is stream-encrypted based on the chaotic map, and the selective encryption algorithm based on the chaotic map In , the coefficients of the discrete cosine transform DCT of all intra-frame blocks are stream encrypted;

Stream encryption: Stream encryption is also called sequence encryption. The idea of stream cipher encryption and decryption is to first treat the video stream data as a plaintext stream m composed of basic coding units 0 and 1, and then use a key stream k to match the plaintext stream m Encrypt bit by bit to get a ciphertext stream c. When decrypting, the same key stream k generated synchronously and this ciphertext stream c are decrypted bit by bit to recover the plaintext stream m; this key stream k is generated by a chaotic system sequence, the chaotic system uses a discrete one-dimensional Logist chaotic map;

The encryption strength of stream encryption depends entirely on the randomness and unpredictability of the key stream, and the chaotic system has pseudo-randomness and unpredictability, and because of its easy realization when generating the key stream, it is generated by the chaotic system. The sequence of can be used as key stream, and the present invention adopts discrete one-dimensional Logistic chaotic map in stream encryption;

Chaos-based selective encryption algorithm: the core idea is to encrypt only part of the data, that is, to encrypt a part of the video data based on the chaotic map;

Discrete cosine transform (DCT) coefficient: DCT is a data processing method, its essence is to convert a space domain function into a frequency domain function, and the 8*8 frequency system matrix obtained according to DCT transformation is called DCT coefficient, which further highlights the Main image information components (DC and low-frequency components) that have a large visual impact, while weakening or omitting secondary details (high-frequency components) that have little visual impact;

The energy of the compressed image is mainly concentrated in the low-frequency part and DC part of the 64 DCT coefficients, so in order to obtain sufficient security against possible attacks, in our selective encryption algorithm based on chaos, all intra-frame blocks The DCT coefficients (1 DC and 63 AC) are all stream encrypted;

Third, the second level video stream data encryption scheme based on discrete Baker map and hyperchaotic map

For the H.263 video stream data after the first level of encryption and coding, the second level of encryption is performed. The second level of encryption scheme consists of two parts, that is, firstly, the two-dimensional discrete Baker map is used to scramble the data, and then based on the super Chaos mapping performs data obfuscation on the scrambled data;

Section 3.1, data scrambling

The data scrambling of the encoded H.263 video stream data is realized by using two-dimensional discrete Baker mapping; in the corresponding cryptographic system structure, the H.263 video stream data encoded by the first level of encryption is used as a plaintext stream, and the following will give The specific steps of data scrambling:

Step 1: Take N×N bytes of data from the first-level encrypted H.263 video stream data for two-dimensional discrete Baker mapping scrambling iterations. The scrambling key k _B is composed of l, n ₁ , n ₂ ,..., n _k constitute, l represents the number of iterations of data scrambling, (n _i , i=1, 2,..., k) is the component value of the key, N, l, n _i are set to be greater than 0 Integers, the selection principle of N and l is to achieve a better balance between security and processing speed, the larger N and l, the stronger the security, and the slower the processing speed, (n ₁ , n ₂ , L n _k ) must satisfy n _i |N, i=1, L, k, and n ₁ +L+n _k =N;

Step 2: Scramble the iteratively selected data (b ₁ , b ₂ , L L b _N×N ) for one time to obtain the scrambled data (bp ₁ , bp ₂ , L L bp _NN );

Step 3: Store the scrambled data (bp ₁ , bp ₂ , L L bp _NN ) in the original location of the video stream data;

Section 3.2, data confusion

The sequence generated by the hyperchaotic equation is selected to achieve data confusion. The state sequence (x ₁ , x ₂ , L L x _NN ) generated by the hyperchaotic equation will be discretized to obtain the bit sequence (c ₁ , c ₂ , L L c _NN ) and the set The scrambled H.263 video stream data sequence (bp ₁ , bp ₂ , L L bp _NN ) is used for obfuscation operation, and the obfuscation operation is realized by XOR operation; the obfuscated H.263 video stream data and the unchanged H.263 video Streaming data are used together as input data for video network transmission.

Advantages and positive effects of the present invention

The invention proposes a two-level video streaming media encryption method based on chaos and hyperchaos, and realizes the purpose of secure transmission of video streaming media information.

The present invention provides an advanced and effective method for the design and analysis of video stream media encryption system, greatly improves the security degree of video data information transmission, and provides an innovative technology for information security in networked society. It has good economic and social benefits in the fields of finance and military, and can promote the enrichment and development of chaotic cryptography theory. Therefore, this achievement has important theoretical value and application prospects.

The present invention has the following advantages:

(1) The design idea of a two-level chaotic video encryption system is proposed, and combined with the video compression standard H-263 under network transmission, video stream chaotic encryption based on discrete cosine transform (DC) is performed in the video compression process, After video compression, block scrambling and data confusion are performed on the video stream. This encryption scheme is completely suitable for application in network video streaming media transmission, and has good feasibility, versatility and security.

(2) The encryption scheme of the first level is to perform selective stream encryption on the video stream data based on the Logistic chaotic map in the process of video compression, that is, to perform stream encryption on a part of the video stream data based on the chaotic map. An auxiliary encryption link for the security of the encryption system. In the selective encryption algorithm based on chaotic maps, the coefficients of the discrete cosine transform DCT of all intra-frame blocks are encrypted by the flow, so sufficient security can be obtained to resist possible The attack that appears. After video compression, the two-dimensional discrete Baker map is used to scramble the video stream data and the scrambled data is scrambled based on the hyper-chaotic map; it is an organic fusion of data encryption measures with different attributes, which makes the encryption system more efficient. High security.

(3) Existing chaos-based encryption schemes use scrambling methods to realize key scrambling of digital images or videos. And our cryptographic system here uses two-dimensional discrete Baker mapping to realize the data scrambling of the encoded H.263 video stream.

(3) Not only a simple chaotic system but also a super chaotic system are used in the encryption algorithm. The simple chaotic system only contains a positive Lyapunov exponent, has a one-dimensional complex dynamic expansion, and has the possibility of being easily deciphered by mature nonlinear data processing techniques. Hyperchaos is a chaotic system with complex dynamic expansion in high-dimensional regions, and it is a chaotic system with more than two positive Lyapunov exponents. The complex time series generated by the hyperchaotic system has greater randomness and unpredictability. Using this signal encryption greatly improves the encryption quality and is very difficult to decipher.

(4) The two-level hyperchaotic video encryption system based on this encryption algorithm has been tested and analyzed in a network environment, and its security and various performance indicators such as processing speed, compression ratio, and image quality after reconstruction have reached ideal performance indicator requirements, showing the good practicability of the method.

[Description of drawings]:

Figure 1 is a chaotic video encryption network communication system.

FIG. 2 is a sequence diagram of socket (socket) invocation of the UDP protocol.

Figure 3 is a structural diagram of the chaos-based video encryption system model.

Figure 4 is a block diagram of a two-level video streaming encryption algorithm based on chaos and hyperchaos.

Figure 5 is the trajectory diagram of the Logistic chaotic sequence.

Fig. 6 is a working flowchart of the first level encryption and decryption system.

Fig. 7 is a working flowchart of the second layer encryption system.

Fig. 8 is a two-dimensional continuous Baker map.

Fig. 9 is a two-dimensional discrete Baker map.

Figure 10 is a hyperchaotic sequence trajectory diagram.

Figure 11 is an example effect of two-level video stream data encryption based on chaos and hyperchaos, (a) is the original image, (b) is the coded image after the first layer of encryption, and (c) is the second layer of encryption Encoded image, (d) is the decrypted image.

【Detailed ways】:

The video streaming media encryption system is mainly composed of the following parts: related hardware equipment, video streaming media encryption software, video streaming media decryption software, where the encryption software is placed at the streaming media publishing end, the decryption software is placed at the streaming media receiving terminal, the encryption software and The core part of the decryption software is the encryption and decryption algorithm. What the present invention mainly describes are the design steps and technical principles of the video stream encryption algorithm scheme based on chaos and hyperchaos. The connection principle and system function design are introduced, and the following main part describes the specific implementation steps of the encryption method in the invention in combination with video sequence examples.

1. Network connection principle and system function design of chaos-based video encryption system

1.1 Principle of video network connection

For online streaming media publishing, video conferencing, digital TV, etc. that require encryption, the encrypted video data needs to be transmitted to the client through the network, so there are related network transmission control problems, we only give the relevant information here The chaotic video encryption system in the invention is closely related to concepts such as network transmission protocols, and the chaotic video encryption network communication system is shown in Figure 1:

In the network environment, the communication between different computers is distributed process communication, which adopts the client/server (Client/Server) mode. Both the client and the server are application programs for communication. In this mode, the client requests services from the server. , and the server responds and serves the client. The TCP/IP protocol, that is, the Transmission Control Protocol/Internet Protocol, is the basis for realizing network interconnection. It is composed of the underlying IP protocol and the TCP protocol; the IP protocol is the Internet Protocol. All communication parties need to communicate on the Internet, and all connections to the Internet All computers must abide by the IP protocol. In order to complete the communication between application processes of different computers, the TCP/IP protocol uniquely identifies a process within the entire network. At this time, it is necessary to use the IP address of the network layer and the port number of the transport layer. Together, they are called a socket. Socket address. Distributed process communication in the network environment needs to use Socket programming. Each computer in the network environment has its own operating system, and the operating system should provide an interface for network applications, that is, the network application programming interface (Application Program Interface, API). Socket shields the difference between the underlying communication software and the specific operating system, making it possible to communicate between any two computers that install TCP protocol software and implement the socket specification.

The real-time video system requires high data transmission rate. Among the TCP/IP protocols, the User Datagram Protocol (UDP) protocol is mainly used for application layer protocols that require high transmission efficiency. In the UDP client-server pattern, the client-server represents two application processes communicating with each other. UDP servers provide services to clients through well-known port numbers. A UDP client refers to an application program that uses a certain network by requesting a service from a server through a temporarily applied port number. Both the UDP server and the UDP client will create an output queue and an input queue, which are used to send and receive UDP packets respectively. The UDP server handles concurrent services in a duplicate server manner. The repeating server uses a request queue to store incoming service requests, and processes server requests sequentially on a first-come, first-served basis. The number of client requests processed by a duplicate server is limited by the length of the request queue, but it can effectively control the processing time of server requests. The timing diagram of the Socket call of the UDP protocol is shown in Figure 2. The actual operation has also verified that the UDP protocol used in the transport layer can sufficiently guarantee the real-time transmission of video data in the chaotic video encryption system.

1.2 System function design

The chaotic video encryption system is established in the TCP/IP network protocol. The interaction mode between the two processes is the client/server mode (C/S mode), that is, the server side of the network video real-time acquisition and transmission encryption system waits for the client to make a request and Respond, and the network video real-time acquisition and transmission decryption system client will apply to the server when needed, and the server will always run as a daemon process, monitoring the network port, once a customer requests, it will start a service process to respond to the user. Software function realization mainly consists of four parts: real-time video capture and playback, compression encoding and decoding; video network transmission, video encryption and decryption. For each video input packet, in order to realize the end-to-end communication in the local area network, the IP and port number of the server must be known in advance, and the network connection dialogue and data transmission must be carried out. Here it is emphasized that the encryption system realized by the present invention does not adopt the transmission control protocol RTP to ensure the QoS (Quality of Service) of video data transmission, which can be further improved in future practical applications to enhance the applicability under the wide area network.

1.3 Real-time video capture and playback function realization

Real-time acquisition and playback of video information is required in the chaotic video encryption system, so it is necessary to select easy-to-implement and mature video acquisition software and hardware. The video capture technology is applied in the video software. A digital video software package VFW (Video for Windows) SDK specially used for video capture can be used. VFW SDK provides a standard interface for video capture in Windows system, which greatly reduces the difficulty of program development. VFW enables applications to obtain digitized video clips from traditional analog video sources through digitizing devices. A key idea of VFW is that it does not require special hardware for playback. It introduces a file standard called AVI, which specifies how video and audio should be stored on the hard disk, and alternately stores video frames and corresponding frames in AVI files. Matching audio data. VFW enables programmers to capture, play and edit video clips by sending messages or setting properties. AVICap supports real-time video stream capture and video single frame capture.

The AVICap window class provides the following functions:

◆Separately control the collection of audio and video;

◆Use overlay (real-time superposition) or preview (preview) to display video images;

◆Work simultaneously with ICM and ACM to compress audio and video data directly into the application;

◆Compress audio and video streams directly into AVI files without requiring developers to understand the details of the AVI file format;

◆Know the video and audio input devices dynamically;

◆Create, save and load palettes;

◆Copy the image palette to the clipboard;

◆Control MCI equipment;

◆Capture a single frame image and save it in DIB format.

1.4 Realization of compression coding function

We adopt the H.263 standard to realize compression encoding and decoding functions. The H.263 standard is a video compression coding standard with a low data transmission rate. It uses motion compensation, inter-frame prediction, etc. to remove the redundancy of the image in the time domain, and then uses discrete cosine transform (DCT) to remove the redundancy of the image information space. Redundancy, and finally use variable word length statistical coding to remove the statistical redundancy contained in the quantized DCT coefficients, so as to achieve the purpose of data compression.

2. Model expression of chaos-based video encryption system

The video encryption system based on chaos belongs to the digital encryption system, that is, the chaotic sequence generated by solving the chaotic equation is designed as a key and video sequence encryption under finite precision. It also belongs to the symmetric key system. Video frames are called plaintext, and correspondingly encrypted video frames are called ciphertext. Here we give the model expression of the chaos-based video encryption system, including basic concepts and structural descriptions. The model structure is shown in Figure 3:

The chaos-based video encryption system consists of five parts, whose symbols are expressed as P, C, K, {E _k , k∈K}, {D _k , k∈K}, and the meaning of the mathematical expression is as follows:

1) P represents a finite set of original video frames.

2) C represents a finite set of encrypted video frames.

3) K represents a key, which is a set of keys generated by one or more chaotic or hyperchaotic systems.

4) For each k∈K, there is an encryption equation E _k ∈ε and a corresponding decryption equation D _k ∈D, here for each E _k : P→C and D _k : C→P, there is equation (1) as follows:

D _k (E _k (x)) = x (1)

is true for every video frame. where ε and D are the set of all encryption and decryption equations.

The first-level chaos-based stream encryption scheme in Section 3.2 and the scrambled H.263 video data encryption scheme based on hyperchaotic equations in Section 3.3.2 are designed under the above-mentioned architecture, except that The biggest difference lies in the difference between the key K and the encryption equation E _k : P→C and the decryption equation D _k : C→P, which is the focus of the work of the cipher designer.

3. Implementation steps and examples of a two-level video streaming encryption scheme based on chaos and hyperchaos

3.1 Structure of two-level video streaming encryption scheme based on chaos and hyperchaos

Figure 4 presents the block diagram of the proposed encryption scheme. It can be seen from Figure 4 that the two-level multi-chaotic video encryption system is also a symmetric encryption system. The client and server of both communication parties have the same key k _A , k _B , k _C to encrypt plaintext and decrypt ciphertext . In the software implementation, a series of video frames are captured, compressed and transmitted through the IP network. During this process, encryption algorithms are applied to resist security threats. In order to meet the requirements of real-time video transmission, the encryption system uses H. 263 codec, suitable for encryption of various video stream data meeting the H.263 standard.

In this example, the video stream data parameters used are as follows: the video frame video format used is Qcif format, the video signal brightness resolution is 176x144 pixels, the video signal chrominance resolution is 88x72 pixels, and the transmission bit rate is black and white video 6.1Mbit /s, color video 9.1Mbit/s, the maximum allowable bit of each frame is 64, and the video frame adopts the intra-frame coding mode, that is, all video frames are regarded as I frames. Because the encryption system we designed is a symmetric encryption system, the decryption process is the reverse process of the encryption process, so the encryption algorithm and encryption process will only be given with examples.

3.2 Chaos-based first-level stream encryption scheme

3.2.1 The structure of the first level chaotic encryption system

The encryption scheme of the first level is based on the Logistic chaotic map to perform selective stream encryption on the video data, that is, to encrypt all DCT coefficients. The specific design of the algorithm is as follows.

The encryption system designs the structure of the encryption system with reference to Figure 3, and its five structural elements are as follows:

1) The key k is composed of μ and x ₀ ; P={a _i }={DC, AC ₁ , L L AC ₆₃ } consists of the DCT process and all 64 DCT coefficients after quantization;

2) C consists of the DCT coefficient set {b _i } of all streams encrypted;

3) E _k can be described as an Xor (exclusive OR) operation, which is obtained by the following formula (2):

${\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&CirclePlus;</span>{\mathrm{<span\; class="notranslate">\; dx</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

4) D _k can be described as an exclusive OR (Xor) operation, which is obtained by the following formula (3):

${\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&CirclePlus;</span>{\mathrm{<span\; class="notranslate">\; dx</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&CirclePlus;</span>{\mathrm{<span\; class="notranslate">\; dx</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&CirclePlus;</span>{\mathrm{<span\; class="notranslate">\; dx</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

3.2.2 The process of chaos-based video stream encryption system

The encryption process of this encryption system consists of the following six steps:

1) Generation of key stream. According to the given initial key, that is, the initial state x ₀ of the chaotic system and the control parameter μ, iteratively calculate the Logistic chaotic system N times to obtain the chaotic state x _n , and then discretize the chaotic state x _n to obtain the 16-bit discretization of x _n Expressed. One-dimensional discrete-time Logistic chaotic system is expressed by formula (4):

x _k+1 ＝μx _k (1-x _k ) (4)

In the formula, x _k ∈ X, k=0, 1, 2, 3L is a discrete time sequence, and the system maps the current state x _k of the sequence to the next state x _k+1 . For example, starting from the initial state x ₀ and iterating repeatedly, the sequence {x _k ∈X, k=0, 1, 2, 3L} is obtained. This sequence is called a trajectory of the discrete chaotic system. Among them, 0<μ≤4 is called branch parameter, x _k+1 ∈(0,1). When 3.5699456≤μ≤4, the sequence obtained by calculating the chaotic system is the chaotic sequence. In this example, μ=3.5699614 x ₀ =0.45678912345; Figure 5 shows the calculated Logistic chaotic sequence.

2) Take out a frame from the video sequence, and then obtain the video data in RGB format after video processing;

3) Perform DCT process and quantization process to obtain all intra-frame macroblocks;

4) Perform stream encryption operation on all DCT coefficients of all blocks in the frame, the specific equation is Ek (exclusive OR operation) introduced in the previous section;

5) Variable length coding (VLC) is carried out to the encrypted video stream;

If the processed frame is the last frame in the video sequence, exit the entire encryption process, otherwise go to step 2 to continue.

Since the encryption system is a symmetric encryption system, the decryption process corresponds to it. See Figure 6 for the specific structure diagram.

3.3 Second-level video encryption scheme based on Bakaer map and hyperchaos

The encryption scheme of the second layer consists of two independent parts, that is, data scrambling based on 2-D Baker mapping and data obfuscation of the scrambled data with a hyperchaotic system. In the corresponding cryptographic system structure, after encoding The H.263 video stream is used as plaintext. The design module of the encryption scheme of the second level is shown in Fig. 7 .

3.3.1 H.263 video data scrambling based on 2-D Baker mapping

Our cryptographic system here uses 2-dimensional discrete Baker mapping (2-D Baker mapping) to implement data scrambling of the encoded H.263 video stream. The calculation method of 2-D Baker mapping is as follows:

A two-dimensional Baker map is a chaotic map that can define a two-dimensional Baker map in the unit square. As shown in Figure 8, the unit square is divided into k rectangular blocks in the horizontal direction [F _i-1 , F _i )×[0, 1), i=1, L, k, F _i =p ₁ +Λ+p _i , F ₀ =0, where p _i is the width of the rectangle, and p ₁ +L p _k =1. For the i-th rectangular block [F _i-1 , F _i )×[0, 1), the Baker map actually stretches 1/p _i in the width direction, and compresses p _i in the length direction, and finally Turn a column of rectangular blocks arranged horizontally into a column of rectangular blocks stacked on top of each other in the vertical direction. The above process is expressed mathematically as:

$\mathrm{<span\; class="notranslate">\; B</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&Element;</span><span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; 0,1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 5</span>}<span\; class="notranslate">\; )</span>$

When implemented on a computer, it is necessary to discretize the continuous Baker map, that is, to establish a 2-D Baker map. The specific 2-DBaker mapping can be carried out according to the following steps:

1) Divide the N×N square into k rectangular blocks in the horizontal direction, each rectangular block has N×n _i pixels, where n _i can divide N evenly;

2) Each rectangular block is further divided into n _i sub-blocks (as shown in Figure 9), because each large rectangular block has N×n _i pixels, so after being divided into n _i sub-blocks, each sub-block has exactly N pixel;

代表离散化的一般Baker映射，3) In each rectangular block, rearrange the pixels into a row in order from bottom to top and from left to right. Let the square be N×N, use

Represents the discretized general Baker map,

N _i =n ₁ +Λ+n _i , n _i |N, i=1, A, k, and n ₁ +Λ+n _k =N (6)

For pixel (r, s),

${\mathrm{<span\; class="notranslate">\; B</span>}}_{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&Lambda;</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; N</span>}}{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}\mathrm{<span\; class="notranslate">\; mod</span>}\frac{\mathrm{<span\; class="notranslate">\; N</span>}}{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; ,</span>\frac{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{\mathrm{<span\; class="notranslate">\; N</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}\mathrm{<span\; class="notranslate">\; mod</span>}\frac{\mathrm{<span\; class="notranslate">\; N</span>}}{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 7</span>}<span\; class="notranslate">\; )</span>$

Where N _i ≤r<N _i +n _i , 0≤s<N.

After a certain number of iterations, the 2-D Baker map exhibits chaotic characteristics.

Take N×N bytes of data from the encoded H.263 video stream data for two-dimensional discrete Baker mapping scrambling iterations, and the scrambling key k _B consists of l, n ₁ , n ₂ ,...,n _k , l represents the number of iterations of data scrambling, (n _i , i=1, 2, ..., k) is the component value of the key, N, l, n _i are taken as integers greater than 0, the larger N and l are, The stronger the encryption security, but the greater the amount of calculation, the slower the processing speed, so the selection principle of N and l is to achieve a better balance between security and processing speed, (n ₁ , n ₂ , L n _k ) also satisfy n _i |N, i=1, L, k, and n ₁ +L+n _k =N; in this example, take N=32, l=9, k=4, {n ₁ , n ₂ , n ₃ , n ₄ }={4, 8, 16, 4}.

The specific steps of data scrambling are given below:

1) Take out N×N byte data from the encoded H.263 video stream (N=32 in this example). The first 8 bytes of the H.263 video stream are unchanged, mainly because it is easy for a cryptanalyst to guess the header structure of the encoded video stream.

2) Iterate the selected data (b ₁ , b ₂ , L L b _N×N ) scrambling operation l times.

3) Store the scrambled data (bp ₁ , bp ₂ , L L bp _NN ) in the original position of the video stream.

3.3.2 Confusion of H.263 video data after scrambling based on hyperchaotic equation

Data obfuscation is to change the character value of the H.263 video data based on the scrambling operation of 2-D Baker mapping. This link adopts the encryption system based on hyperchaos. In this example, the calculation of the hyperchaotic system sequence takes the following model:

The hyperchaotic system is represented by equations (8):

For this hyperchaotic equation, as long as the initial state X ₀ = (x ₀ , y ₀ , z ₀ , w ₀ ) is given, the time series {X _j = (x _j , y _j , z _j , w _j ), j=1,...,NN}, often using the classic Runge-Kutta (Runge-Kutta) fourth-order algorithm for calculation.

In this example, the model parameters a=35, b=3, c=12, d=7 and e=0.58 are taken, and the key k _c is obtained from the initial value state X ₀ =(x ₀ , y ₀ , z ₀ , w ₀ ), which is taken as X ₀ = (1.12345678912345, 2.12345678912345, 3.12345678912345, 4.12345678912345). At this time, the trajectory of the system appears a chaotic trajectory of exponential diffusion in more than two directions. From a security point of view, because the hyperchaotic system has a more complex phase space than the low-dimensional chaotic system, the encryption algorithm designed with it is not easy to be deciphered. Under the above parameters and keys, the sequence trajectory generated by this system is shown in Figure 10.

After obtaining {X _j , j=1,...,NN}, one of the components can be taken, for example, the hyperchaotic state sequence {x _j ,j=1,...,NN} formed by the first component is used for encryption. The hyperchaotic state (x ₁ , x ₂ , L L X _NN ) will be discretized to obtain a bit sequence (c ₁ , c ₁ , L L c _NN ) to experiment with the confusion operation. The specific equation (10) of the confusion operation (XOR operation) is as follows:

${\mathrm{<span\; class="notranslate">\; bm</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; bp</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&CirclePlus;</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; L\; L\; N</span>}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 10</span>}<span\; class="notranslate">\; )</span>$

The obfuscated H.263 video stream is used together with other unchanged H.263 video data as input data for video network transmission.

The above steps are the complete process of the two-level video encryption method based on chaos and hyperchaos.

3.4 The effect of the two-level video encryption method

Figure 11 shows the encrypted and decrypted image renderings of the last frame of the video data walrus video, Figure 11.(a) is the original image, and Figure 11.(b) is the code after the first layer of encryption Image, Figure 11.(c) is the coded image after the second layer of encryption, and Figure 11.(d) is the decrypted image. From Figure 11.(c), it can be seen that the two-level video encryption algorithm has a good encryption effect, because from the subjective perception of the human eye, almost no visible information appears in the image.

Claims (1)

1. A two-level video stream encryption method based on chaos and hyperchaos, characterized in that the method comprises: 1. The structure of the two-level video stream data encryption scheme based on chaos and hyperchaos A two-level video stream data encryption scheme based on chaos and hyperchaos is proposed. The design of the encryption system is combined with the video compression standard H.263 under network transmission, and the encoded H.263 is processed by byte-by-byte processing mode. stream, the first level is to encrypt the video stream data based on the selective stream encryption based on the chaotic map during the video compression process; Scrambling and data confusion, where block scrambling adopts two-dimensional discrete Baker mapping, and then uses hyper-chaotic system for data confusion on the scrambled data; 2. The first level video stream data encryption scheme based on chaotic map The encryption scheme of the first level is: in the video compression process, the video stream data is selectively encrypted based on the chaotic map, that is, a part of the video data is stream-encrypted based on the chaotic map, and the selective encryption algorithm based on the chaotic map In , the coefficients of the discrete cosine transform DCT of all intra-frame blocks are stream encrypted; Stream encryption: Stream encryption is also called sequence encryption. The idea of stream cipher encryption and decryption is to first treat the video stream data as a plaintext stream m composed of basic coding units 0 and 1, and then use a key stream k to match the plaintext stream m Encrypt bit by bit to get a ciphertext stream c. When decrypting, the same key stream k generated synchronously and this ciphertext stream c are decrypted bit by bit to recover the plaintext stream m; this key stream k is generated by a chaotic system sequence, the chaotic system uses a discrete one-dimensional Logist chaotic map; Third, the second level video stream data encryption scheme based on discrete Baker map and hyperchaotic map For the H.263 video stream data after the first level of encryption and coding, the second level of encryption is performed. The second level of encryption scheme consists of two parts, that is, firstly, the two-dimensional discrete Baker map is used to scramble the data, and then based on the super Chaos mapping performs data obfuscation on the scrambled data; Section 3.1, data scrambling The data scrambling of the encoded H.263 video stream data is realized by using two-dimensional discrete Baker mapping; in the corresponding cryptographic system structure, the H.263 video stream data encoded by the first level of encryption is used as a plaintext stream, and the following will give The specific steps of data scrambling: Step 1: Take N×N bytes of data from the first-level encrypted H.263 video stream data for two-dimensional discrete Baker mapping scrambling iterations. The scrambling key k _B is composed of l, n ₁ , n ₂ ,..., n _k constitute, l represents the number of iterations of data scrambling, n _i , i=1, 2,..., k is the component value of the key, N, l, n _i are taken as integers greater than 0, The selection principle of N and l is to achieve a better balance between security and processing speed. The larger N and l are, the stronger the security is, and the slower the processing speed is, n ₁ , n ₂ , L, n _k must satisfy n _i |N, i=1, L, k, and n ₁ +L+n _k =N; The second step: the iteratively selected data b ₁ , b ₂ , L L, b _N×N are scrambled for one time, and the scrambled data bp ₁ , bp ₂ , L L, bp _NN are obtained; Step 3: store the scrambled data bp ₁ , bp ₂ , LL, bp _NN in the original position of the video stream data; Section 3.2, data confusion Select the sequence generated by the hyperchaotic equation to achieve data confusion. The state sequence x ₁ , x ₂ , L L, x _NN generated by the hyperchaotic equation will be discretized to obtain the bit sequence c ₁ , c ₂ , L L, c _NN and the scrambled H.263 video stream data sequence bp ₁ , bp ₂ , L L , bp _NN are obfuscated, and the obfuscated operation is realized by XOR operation; the obfuscated H.263 video stream data is combined with the unchanged H.263 video stream data It is used as input data for video network transmission.

Images