Research On Data Encryption Standard Based On AES Algorithm in Internet of Things Environment
Research On Data Encryption Standard Based On AES Algorithm in Internet of Things Environment
Abstract- The Internet of Things makes it possible to packet length to 128 bits, and the key length can only be
exchange information between people, things and things, one of 128, 192, and 256 bits [4]. The number of rounds of
people and things, and makes information security more the AES algorithm is related to the key length. The
and more important. However, there are few standards in dependencies are shown in Table ĉ.
the field of information security of the Internet of Things.
Therefore, based on the characteristics and application
requirements of the Internet of Things environment, this III. TABLE ĉ RELATIONSHIP BETWEEN KEY LENGTH AND
NUMBER OF ROUNDS IN AES
paper optimizes the Advanced Encryption Standard (AES)
and builds the data encryption standard DESI(Data Key length (bits) Number of rounds
Encryption Standard in IoT) in the Internet of Things 128 10
environment. . The analysis shows that DESI has better 192 12
security and is therefore suitable for encrypting data in the
IoT environment.Keywords: Internet of Things, encryption 256 14
standard, AES, DESI, information security. Figure 1 illustrates the encryption process for the
10-round AES algorithm
Keywords- Internet of Things, encryption standard, AES,
information security
I. INTRODUCTION
The Internet of Things is widely used in China's
logistics, power, medical, environmental protection and
transportation [1]. The application of the Internet of
Things has effectively reduced operating costs and
improved economic efficiency. At the moment, security
issues are another huge challenge facing the rapid
development of the Internet of Things [2]. Encryption
technology can provide effective protection for data
security of the Internet of Things.
II. RESEARCH ON AES ENCRYPTION ALGORITHM OF
INTERNET OF THINGS
Due to the shortness of DES's key space and its
relatively small packet size, the National Institute of
Standards and Technology began to publicly collect
Advanced Encryption Standard (AES) from the world in
1997. After more than three years, the Rijndael algorithm
designed by two Belgian cryptographers Daemen and
Rijmen eventually won from many candidates and became
the advanced encryption standard. AES itself has the
advantages of high speed, high security, and easy
hardware and software implementation [3].
The AES algorithm is basically identical to the
Rijndael algorithm. The difference between the two is that Figure 1 AES encryption process
the packet length and key length are different. The
Rijndael algorithm has a variable packet length and key The internal structure of the AES algorithm consists of
length, which can be set to an integer multiple of 32 bits, four basic steps [5], namely SubBytes, ShiftRows,
ranging from 128 to 256 bits. The AES algorithm fixes the MixColumns, and AddRoundKey. Each of these links
2072
the precondition is that the known subkey w i 1 is the key word is XORed with the round constant, which is
actually the XOR of the first byte to the left of the round
required. In addition, the basic unit of key spreading by constant. It should be noted that the round constants of
the AES algorithm is a 32-bit word. The key expansion each round are different, specifically defined as:
algorithm takes 4 words as input, and the output of the
Rcon[j] (Rc[j],0,0,0) , where Rc[1] 1 , Rc[ j] Rc[ j 1]
2
algorithm is a one-dimensional array of 44 words.
(this multiplication is defined on GF(28 ) ). Table Ċ is a
First, the initial key of AES is used as the first 4 key
words of the extended key array. Thereafter, the newly
hexadecimal representation of the ten round constants
used in the AES algorithm.
generated four key words are added to the remaining
portion of the extended key array each time. Within the
TABLE Ċ ROUND CONSTANT
extended key array, the generation of each new key word
w i depends on w i 1 and w i 4 . At the same time, the
j 1 2 3 4 5 6 7 8 9 10
algorithm uses a more complex function g to calculate the
key word whose array index is a multiple of 4. Figure 5
illustrates how the extended key is calculated. Rc[ 1
01 02 04 08 10 20 40 80 36
j] B
k0 k4 k8 k12
k1 k5 k9 k13 IV. AES-BASED ENCRYPTION STANDARD DESIGN IN
THE INTERNET OF THINGS
k2 k6 k10 k14
k3 k7 k11 k15
When determining the number of rounds of the
encryption algorithm, it is usually necessary to consider
the resistance of the algorithm to the shortcut attack. The
w0 w1 w2 w3 g reason is very simple. The shortcut attack is more efficient
than the exhaustive search key. However, in an IoT
environment where computing resources and storage
resources are relatively short, the number of rounds of the
encryption algorithm is not as good as possible. Therefore,
when designing the encryption standard DESI in the
Internet of Things, this paper considers to properly reduce
w4 w5 w6 w7 the number of rounds of the algorithm, and improve the
execution efficiency of the algorithm under the premise of
...
2073
initial set of keys, there is a different key in the 128th bit, Key 1 b1 a0 43 57 d9 c5 66 f3 d4 e7 31 59 d8 a6 7f 2e
where the key is expressed in hexadecimal.The first set of
random initial keys is shown in Table ċ. In the initial set Key 2 b1 a0 43 57 d9 c5 66 f3 d4 e7 31 59 d8 a6 7f 2f
of keys, there is a different key in the 128th bit, where the The results of the aggregation of the first group initial
key is expressed in hexadecimal. key after 10 rounds of key expansion are shown in Table
Č.
TABLE ċ INITIAL KEY
Numb
Bit
er of
Key 1 Key 2 differe
round
nce
s
initial b1 a0 43 57 d9 c5 66 f3 d4 e7 31 59 d8 a6 7f 2e b1 a0 43 57 d9 c5 66 f3 d4 e7 31 59 d8 a6 7f 2f 1
1 94 72 72 36 4d b7 14 c5 99 50 25 9c 41 f6 5a b2 94 72 56 36 4d b7 30 c5 99 50 01 9c 41 f6 7e b3 9
2 d4 cc 45 b5 99 7b 51 70 00 2b 74 ec 41 dd 2e 5e d4 81 3b b5 99 36 0b 70 00 66 0a ec 41 90 74 5f 19
3 11 fd 1d 36 88 86 4c 46 88 ad 38 aa c9 70 16 f4 b0 13 f4 36 29 25 ff 46 29 43 f5 aa 68 d3 81 f5 52
4 48 ba a2 eb c0 3c ee ad 48 91 d6 07 81 e1 c0 f3 de 1f 12 73 f7 3a ed 35 de 79 18 9f b6 aa 99 6a 59
5 a0 00 af e7 60 3c 41 1a 28 ad 97 4d a9 4c 57 be 62 f1 10 3d 95 cb fd 08 4b b2 e5 97 fd 18 7c fd 71
6 a9 5b 01 34 c9 67 40 7e e1 ca d7 33 48 86 80 8d ef e1 44 69 7a 2a b9 61 31 98 5c f6 cc 80 20 0b 59
7 ad 96 5c 66 64 f1 1c 18 85 3b cb 2b cd bd 4b a6 62 56 6f 22 18 7c d6 43 29 e4 8a b5 e5 64 aa be 63
8 57 25 78 db 33 d4 64 c3 b6 ef af e8 7b 52 e4 4e a1 fa c1 fb b9 86 17 b8 90 62 9d 0d 75 06 37 b3 69
9 4c 4c 57 fa 7f 98 33 39 c9 77 9c d1 b2 25 78 9f d5 60 ac 66 6c e6 bb de fc 84 26 d3 89 82 11 60 73
10 45 f0 8c cd 3a 68 bf f4 f3 1f 23 25 41 3a 5b ba f0 e2 7c c1 9c 04 c7 1f 60 80 e1 cc e9 02 f0 ac 63
2074
V. EFFICIENCY ANALYSIS OF ENCRYPTION STANDARDS REFERENCES
[1] [Chunling Sun. Application of RFID Technology for Logistics on
Efficiency is a measure of the strengths and weaknesses of an Internet of Things[J]. AASRI Procedia,2012(1):106 – 111.
algorithm. In addition to the limited performance of the perceived [2] Almudena D´ıaz-Zayas, Cesar A. Garc´ıa-Pe´rez, A´ lvaro M.
Recio-Pe´rez. 3GPP standards to deliver LTE connectivity for
nodes in the IoT environment mentioned above, we value the IoT[C]. 2016 IEEE First International Conference on
operational efficiency of the designed encryption standard. Of Internet-of-Things Design and Implementation
(IoTDI),2016,283-288.
course, the pursuit of high efficiency will inevitably affect the [3] Wang Ying. Improvement of MixColumn() function in advanced
encryption standard AES [D]. Shaanxi Normal University, 2011.
security of the algorithm, so this paper hopes to make the best
[4] Christof Paar, Jan Pelzl. Exploring cryptography in depth -
balance between efficiency and security. Principles and applications of commonly used encryption techniques
[M]. Beijing: Tsinghua University Press, 2012.51-111.
When testing the efficiency of the built encryption standard, [5] Mary James, Deepa S Kumar. An Implementation of Modified
Lightweight Advanced Encryption Standard in FPGA[J]. Procedia
this paper performs encryption and decryption tests on data of Technology, 2016(25):582 – 589.
size 20KB, 40KB, 60KB, 80KB, and 100KB. The AES algorithm [6] Xiao Xiaocao, Li Shuguo. An Expression Method of S-Box and
Inverse S-Box Replacement in AES Algorithm[J].Microelectronics
and the encryption standard DESI constructed in this paper can & Computer,2014,31(1):112-115.
[7] Mo Jianhua. Research on an encryption algorithm for WSN data
encrypt 128-bit data each time, and encrypt 20KB data as an security [D]. Zhejiang University of Technology, 2010.
example: 20KB is 10
211 B, that is, 10
214 bit. Each time 128-bit [8] Mojtaba Alizadeh, Wan Haslina Hassan, Mazdak Zamani, et al.
Implementation and Evaluation of Lightweight Encryption
data is processed, 20KB of data needs to be cyclically encrypted. Algorithms Suitable for RFID[J]. Journal of Next Generation
Information Technology,2013,4(1):65-77
10
27 times. For the same data, AES and DESI are used for
encryption and decryption respectively. For each algorithm, the
average value is obtained by multiple tests. The experimental
results are shown in Table č.
DESI runtime
Data AES runtime (ms)
(ms)
size
encrypt Decryp encrypt Decryp
(KB)
ion t ion t
20 99 109 62 78
40 208 218.7 140 156
60 317.3 332.7 203 223.7
80 426.3 442 281 301.7
100 525.3 541 353.7 379.3
VI. CONCLUSION
Based on several commonly used encryption algorithms in the
Internet of Things, this paper optimizes the AES algorithm and
combines the characteristics of IoT computing resources and
storage resources to construct the data encryption standard DESI
in the Internet of Things. The core work of this paper is to build
the data encryption standard DESI in the Internet of Things based
on the AES algorithm. The results show that DESI has higher
efficiency than the AES algorithm. Combined with the security
analysis of DESI, it can be concluded that DESI combines
efficiency and security, and is suitable for providing encryption
protection for data in the IoT environment.
2075