Development of Security Rules and Mechanisms to Protect Data from Assaults

Abstract

Cloud cryptography is the art of converting plain text into an unreadable format, which protects data and prevents the data from being misused by the attacker. Different researchers designed various Caesar cipher algorithms for data security. With the help of these algorithms, the data can be converted into a nonreadable format, but the data cannot be completely secured. In this paper, data security is provided in different phases. Firstly, data are secured through a bit-reversing mechanism in which those replace the actual values with no relation to the original data. Then the four-bit values are added at the beginning and end of bits using a salting mechanism to interlink the salting and existing bit-values and hide the original data. A Caesar cipher value is obtained by applying the Caesar cipher algorithm to the resulting bits. The Caesar cipher algorithm is used to implement number-of-shifting on the obtained values. An efficient cipher matrix algorithm is then developed in which different rules are designed to encrypt the data. Afterward, a secure cipher value is obtained by implementing Cipher XORation rules on the result obtained and the user-defined key. In the end, the proposed algorithm is compared with various papers. It identifies how much better the proposed algorithm performs than all the previous algorithms and how much the attack rate can be reduced if this algorithm is used for data security.

1. Introduction

Cloud computing is an online data storage system that allows clients to exchange data from one node to another using different protocols [1]. Many organizations store all their data on cloud servers rather than in their physical data centers [2]. The most significant advantage of cloud computing is that it can be accessed remotely from any part of the world [3]. Its security has become very important as well as storing data in the cloud server [4]. There are two possibilities for attacking the cloud server. One is internal side attacks, and the other is external side attacks [5]. The chances of an attack on the cloud are higher on the inner side than on the outer side [6]. An attacker uses various techniques to attack from the external side, including standard pattern matching and brute force attacks, while internal side attacks are mostly phishing attacks. Each phase of the cloud server must be safe to secure the data [7].

Most researchers have developed a variety of algorithms and techniques for storing cloud data, which, if used correctly, will reduce the chances of attacks instead of increasing them. Different service providers implement other security mechanisms in the cloud, but still, intruders attack the cloud [8]. An intrusion detection and prevention system (IDPs) is one of the best ways to end this problem [9]. IDPs detect and prevent intruders’ attacks and see where the attack arises [10]. An intrusion detection and prevention system monitors any cloud network’s malicious activities. There is a wide variety of intrusion detection and prevention tools available in the form of an antivirus that detects and monitors the traffic or malware and prevent data from intruders [11]. However, these techniques are insufficient to provide internal and external side security.

1.1. Cryptography

Cryptography is derived from two words, i.e., “krypto” and “graphein”, which are Greek words [12]. “Krypto” means hidden, and “graphein” means writing. Cryptography is related to encryption [13], which converts readable text to nonreadable form [14,15]. Cryptography is a data encryption mechanism between the sender and the receiver that prevents data from being misused by third parties. Whenever data are transferred between two entities, it should be done in cipher form rather than in its original form [16], in which the third party may misuse the data and send incorrect data to the receiver, as shown in Figure 1.

Whenever data are transmitted from the sender side, these data are converted with the help of a cryptography algorithm called ciphertext [17]. The ciphertext is a nonreadable text that is difficult for third parties to understand. When ciphertext is sent to the receiver side, converting the text from cipher form to readable form is necessary and is called decryption [18]. Decryption is a mechanism that converts a nonreadable form to a readable form [19].

1.2. Classification of Cryptography

Cryptography has been classified into three categories: symmetric key cryptography, asymmetric key cryptography, and hashed function [20]. In symmetric key cryptography, data encryption between sender and receiver is done by a single key called a secret key [21]. The data are encrypted from the sender side using the secret key, and the data are transmitted to the receiver side [22]. It is necessary to use the same key that is used on the sender side to convert encrypted data to a readable format [23]. Asymmetric key encryption involves encrypting data between the sender and receiver using public and private keys [24]. The first key belongs to the sender, which is used to encrypt the data [25], while the other is the receiver key that allows it to decrypt the data. The sender and receiver create public and private keys in the asymmetric mechanism. When data are encrypted, the receiver’s public key is used, and the data from that key are encrypted and transmitted to the receiver [26]. If the data have to be decrypted, it is done from the receiver’s private key [27]. Hashing is a one-way function in which, once the data are encrypted, the data can never be decrypted again [28]. Whenever data or a file pass through a hashing algorithm, that algorithm generates a hashed code [29]. Hashing is divided into two categories [30]. The first one is cryptographic hashed, and the other is non-cryptographic hashed. A non-cryptographic hashed is used much less than a cryptographic [31,32]. Different symmetric and asymmetric cryptography and hashed mechanisms are discussed in Figure 2.

1.3. Problem Formulation

A Caesar cipher is a substitution technique in which each character is converted to ciphertext by replacing it with another text by some mechanisms or formulae. Different researchers have used different techniques and algorithms to increase the performance of the Caesar cipher algorithm. Some researchers have developed a data security algorithm using the Vigenère and Caesar cipher algorithms. Some researchers have developed a hybrid algorithm by merging different algorithms. Some researchers have used different keys to enhance the performance of the Caesar cipher algorithm. None of the papers worked out any technique or rules without which data encryption would be impossible, and no algorithm has been developed to prevent data security from being breached. The biggest problem with the Caesar cipher is that it is a substitution technique in which data are encrypted and decrypted with a formula. If the attacker somehow gains access to this formula, the attacker can easily break the encrypted data, in which data security and data leakage is the biggest issue. The Caesar cipher has 26 alphabet choices, and the attacker can easily decrypt the ciphertext using the formula by testing the different number of shifts. If effective techniques protect Caesar’s cipher algorithm, it will not be easy for an attacker to understand and crack every technique.

Proposed Paper Solution

In this paper, different rules were used in different phases to encrypt the data, first by implementing a bits-reversing mechanism to replace the plain-text values with those values that have no link to the original values, and then, the values were increased by applying a salting mechanism on the bit-reversing results. The values obtained from bit reversing completely differ from the plain text values. When four-bit salting values apply to the beginning and end of these values, and when the salting values convert into pairing, the original values and the obtained values are completely different. After that, the Caesar cipher encryption algorithm was implemented based on the results obtained. Certain sets of rules were designed into the Caesar Cipher algorithm to provide security to the data. After that, the cipher matrix algorithm was implemented on the Caesar cipher text obtained. Some rules were defined in the cipher matrix algorithm, which helped secure the data, and the data could not be broken unless the decryption mechanism was designed according to the rules. The cipher matrix text was then secured with the help of a key. Cipher XORation was used to implement the key on the cipher matrix text, with the help of which the data were fully secured. Data security leakage may occur when a mechanism secures data, but when data are secured according to rules, data cannot be decrypted without applying the rules.

The remaining sections are distributed in the same way. Section 2 deals with previous work. In Section 3, the algorithm is designed. The algorithm is tested in Section 4. Section 6 compares the latest work with previous work, while Section 7 concludes the paper.

2. Literature Review

According to Arroyo et al. [33], Caesar cipher cryptography is considered one of the best algorithms among the old algorithms that can be used to hide data from an attacker and send the information in a secure form. Still, if the attacker gets information about the Caesar cipher, the attacker can quickly break the Caesar cipher algorithm, which is more accessible than other algorithms. To solve this problem, the researchers enhanced the functionality of the Caesar cipher algorithm into which the Goldbach code algorithm was introduced to hide the actual data. The plain text was first taken to convert the data into ciphertext. The Caesar cipher algorithm was implemented on this plain text in which each character was replaced with different characters according to some rules. Then, the Goldbach code algorithm was implemented on the Caesar cipher results obtained. In the Goldbach code algorithm, the results obtained from the Caesar cipher were encoded three times, and a ciphertext was obtained.

According to Tan et al. [25], the Caesar cipher is a substitution algorithm that can be used to secure data. Still, if an attacker obtains any information about this technique, he can decrypt the data. To solve this problem, the researchers developed a hybrid algorithm in which various papers on the Caesar cipher and Vigenère cipher algorithm were surveyed, and the best techniques were collected and implemented. First, plain text was taken, and the Caesar cipher algorithm was implemented on this plain text. In the Caesar cipher algorithm, each plain-text character was shifted by n = n + 3, and the resulting character was stored in place of the plain-text character. After that, the Vigenère cipher algorithm was implemented on the Caesar cipher results with the help of a randomized key, and a cipher text was obtained.

According to Hossain Md [34], The Caesar cipher algorithm cannot be broken until an efficient technique is implemented on the Caesar cipher. To solve this problem, an advanced encryption algorithm was developed, which improves the security of the Caesar method by combining cipher methods such as the Caesar cipher, Playfair approach, and stream cipher. This method used three different static keys, making the algorithm even stronger. Each key is used in a different phase to protect the data from the attacker with the help of these keys. After that, decryption can be made more difficult with the help of keys, and the proposed method may be more secure than other traditional techniques.

According to Qowi et al. [35], whenever text information is exchanged between two points, there is a risk of data being broadcast to different points. The best way to reduce this risk is to use cryptographic techniques. In this paper, researchers identified the weakness of the Vigenère cipher algorithm. It has been discussed that the key used in the Vigenère cipher could be obtained with the help of the Kasiski test. If this key is obtained efficiently, the security of the key can be increased. To solve this problem, the Caesar cipher and Hill cipher algorithm were used, in which the Caesar cipher algorithm was implemented first on a randomized key. The results were implemented on the Hill cipher algorithm, and a cipher text was obtained.

In the paper [36], Musa et al. analyzed various cryptographic algorithms used by cloud service providers, then collected efficient techniques from these papers. An algorithm to secure data from man-in-the-middle attacks was designed in which the data were converted into ASCII to encrypt. Then ASCII was converted to a binary value; 0 was used for bit completions, then the last four bits were inverted, and a ciphertext was generated. The ASCII value was converted to binary for data decryption, and 0 was added for completion. The last four bits were then reversed, and the binary bits were converted to ASCII and obtained into plain text.

According to A. Serdano et al. [37], to protect the ciphertext data from cryptanalysis attacks, it is necessary to implement the secure key algorithm on the randomized key, which can be used to protect the key from cryptanalysis. In this paper, an algorithm was developed to protect text data from cryptanalysis, in which the Hill cipher and Caesar cipher algorithms were used. First, the randomized key was implemented on the Caesar cipher substitution algorithm, and one key was obtained. After that, the Hill matrix algorithm was implemented on the key obtained from the Caesar cipher, and the second key was obtained. When different keys are obtained from both algorithms, both keys are used for encryption and decryption. After that, the time complexity of both keys was identified, and it was discussed that the bigger the key size, the longer the encryption and decryption time of the key will be.

In the paper [38], researchers collected various prevention techniques to protect cloud data from man-in-the-middle attacks on the network and discussed that using cryptographic encryption is the best solution to protect cloud data. Various Caesar cipher cryptographic algorithms were surveyed to solve this problem and identify how data can be protected from attackers using the Caesar cipher algorithm. To encrypt plain text, the first random shift key-generation algorithm was developed in which a key was generated in decimal format. After that, plain text and keys were implemented in this algorithm. In the random shift key-generation algorithm, each character of plain text was shifted according to each value key, and then, ciphertext was obtained by implementing the Caesar cipher algorithm.

According to authors [39], the evolution of communication technologies has facilitated the exchange of information between numerous parties. However, not every piece of information can be utilized by several parties. Sometimes, the material is secret or is only intended for specific purposes. Information security processing is, therefore, necessary. This paper combines the data encryption techniques Vigenère cipher and Caesar cipher. It is discussed that the Vigenère cipher and Caesar cipher are two substitution techniques that replace the original data with such values, which have no relation to the actual values. It becomes difficult for the attacker to crack such values. Steganography was also used to hide messages or other pieces of information, and it was anticipated that this method would produce more difficult-to-crack encryption.

According to researchers [40], many cryptographic algorithms have been designed to secure data, with the help of which the data can be broadcast, but the broadcast data cannot be secured. If the attacker obtains the linked key from this encrypted data, he can easily access the data. In this paper, an algorithm was developed to secure the data, with the help of which a key was obtained from the plain text, and the data were decrypted with its help. Two layers were used to secure the data. In the first layer, the proposed algorithm was implemented on plain text. In the second layer, the Vigenère cipher algorithm was implemented on the results obtained from layer 1, and a ciphertext was obtained. A complete summary of the surveyed literature is shown in

Table 1. Literature Summary.

Reference	Year	Proposed Work
[33]	2020	Implemented the Goldbach algorithm on the Caesar cipher algorithm to enhance the functionality of the Caesar cipher algorithm.
[25]	2021	Surveyed various Vigenère cipher and Caesar cipher papers and collected their best techniques. After that, developed a hybrid algorithm to secure the data from attacks.
[34]	2021	Combined efficient techniques of Caesar cipher, Playfair, and stream cipher algorithm and developed an advanced encryption algorithm to improve the security of the Caesar cipher algorithm.
[35]	2021	Identified the weakness of the Vigenère cipher algorithm and discussed that the key used in the Vigenère cipher could be obtained with the help of the Kasiski test. The best Caesar cipher and Hill cipher algorithm techniques were used to solve this problem.
[36]	2021	Surveyed various papers, collected different techniques, and then developed an algorithm to prevent cloud data from man-in-the-middle attacks.
[37]	2021	An algorithm was developed to protect text data from cryptanalysis, in which Hill cipher and Caesar cipher algorithms were used.
[38]	2022	Collected various Caesar cipher techniques to protect data from man-in-the-middle attacks. After that, developed an algorithm that identified how data can be protected from attackers using the Caesar cipher algorithm.
[39]	2022	Combined Caesar cipher and Vigenère cipher algorithms to encrypt the data. After that, discussed that steganography is a way to hide all pieces of information and discussed that this method can produce more difficult-to-crack encryption.
[40]	2022	The two-layer algorithm was developed to protect the cloud data. In the first layer, the proposed algorithm was implemented on plain text, and a static key was generated. In layer two, the Vigenère cipher algorithm was implemented on the layer 1 result, and a ciphertext was obtained.

.

3. Proposed Algorithm

This article has modified the Caesar cipher algorithm using new techniques and methods for secure data transmission.

3.1. Work Overflow

An encrypted algorithm was developed for data transmission in the cloud server. If used efficiently, it can save data from man-in-the-middle attacks, and whatever information is transmitted over the entire network can be transmitted in a secure and reliable form. First, the data are converted to different decimal values by different mechanisms, and then, the Caesar cipher encryption algorithm is implemented in these decimal values. After that, a Cipher matrix algorithm is implemented on the Caesar cipher encryption algorithm’s results to obtain the cipher matrix text. After obtaining the cipher matrix text, a user-defined key is being implemented on it, and then, a nonreadable text is generated called ciphertext.

3.2. Data Encryption Process

Different procedures are followed when converting plain text to ciphertext, as shown in Figure 3, whenever the plain-text data are transmitted to the cloud server. The plain-text data may contain various special symbols, alphabets, and alphanumeric keys. First, each character in plain text is converted to its ASCII value, and then, every ASCII value is converted to an eight-bit binary. After converting to binary, a bit-reversing procedure is applied in which all bits are reversed. The reason for reversing the bits is that if an attacker tries to decrypt the data, the attacker will obtain different data due to the reversing of bits. After reversing the bits, a salting mechanism will be used, providing more security, and decimal text conversion will be done from these bits.

After obtaining the decimal value, the Caesar cipher encryption algorithm is applied to these decimal values. The values acquired through bit reversing and salting are subjected to a series of time-shifting operations using the Caesar cipher method. Then, an efficient algorithm, the cipher matrix algorithm, is applied to the results of the Caesar cipher encryption algorithm. A set of rules is designed in the cipher matrix algorithm, with the help of which the data will be protected. When an attacker tries to decrypt the data using snatching rules or techniques, but when the rules of the attacking algorithm or algorithm do not match the proposed algorithm, it may not be possible to break the security of the data. After that, a key is implemented on the cipher matrix text obtained from the cipher matrix algorithm so that only authentic users can access the data, for which the Cipher XORation technique is used. In Cipher XORation, the user-defined key and the cipher matrix text are XORed, and a secure ciphertext is obtained. When the data are secured according to different phases and rules, it is impossible to break the security of the data no matter how much the attacker develops a secure algorithm.

3.2.1. Bit Reversing

Bits reversing is the process of reversing all bits so that if there is a man-in-the-middle attack, the attacker cannot obtain the original data due to reversing the bits, and the attacker has access to data that have no link to the original data.

3.2.2. Salting

Salting is random data in which additional bits are added to the existing text, providing security to data. In this paper, eight random bits are taken for salting, out of which four bits are used at the start and four bits at the end. In salting, when the four bits are added to the start, the starting four bits and the next four bits are merged, and when the eight-bit value is converted to decimal, the salted data are completely different from the plain text. When any decryption mechanism is implemented on salted values, the original value can never be obtained, which can be considered a complete data security technique.

3.2.3. Caesar Cipher Algorithm

Whenever plain text is transmitted, such plain text may contain special symbols, letters, and alphanumeric cases. In the Caesar cipher encryption algorithm, a decimal value is converted to equivalent ASCII, and the formula “E = D + N” is used on the ASCII. “E” is an encrypted text that is retrieved from “D + N”. “D” means ASCII value, while “N” means the number of shifts. The number of shifts means a constant value added with “D,” and the result is achieved.

3.2.4. Cipher Matrix Algorithm

The cipher matrix algorithm is a technique in which each value is adjusted in tabular form. The cipher matrix algorithm defines a set of rules which is used to encrypt and decrypt the data.

Rule 1: First, a pair of text values is created. After pairing, if any character in the text is left single, with no character present for the pair, then “x” will be appended with it, as shown in Figure 4.

For example, a text “testing” is taken in which pairs of characters are created, and there is no character to pair with the last character, “g”. Therefore, it is paired with “x” according to the rule.

Rule 2: If the two characters in the pair are the same, then each value will be separated and paired with “x”. For example, in a pair “zz”, each “z” will be separated and paired with “x,” such as “zx” and ”zx”, as shown in Figure 5.

Rule 3: If both pairs’ characters are in the same table column. The characters will be replaced by the character below them in the encryption process. In contrast, in the decryption process, characters are replaced by their upper characters, as shown in Figure 6.

Rule 4: If both characters are in the same table row, in the encryption process, the characters will be replaced with the characters on their right side. If the character is in the last column, no character is on its right side. The character of the first column of the same row will be picked up, while in the decryption process, characters will be replaced with their left-side characters. If the character is in the last column, no character is on its left side, and then, the character will be replaced with the character of the first column of the same row, as shown in Figure 7.

Rule 5: If the pair of both characters are not in the same row or the same Column of the table, then a rectangular shape will be formed on the parallel points of the characters, and they will be replaced by the character of the opposite points, as shown in Figure 8.

In Algorithm 1, first the plain text is taken, and each character of that plain text is converted to its equivalent ASCII. The different ASCII values of each plain text character are converted to an eight-bit binary. After obtaining different binary values, the bit-reversing mechanism is implemented on all binary values. The bit-reversing mechanism is implemented so that the values obtained from bit reversing are completely different from the plain-text values. The salting mechanism is implemented on the values obtained from the bit-reversing mechanism. In the salting mechanism, eight-bit binary values are taken, out of which a four-bit value is inserted in the start, and a four-bit value is be inserted in the end, and when pairing is done, all the data will be opposite to the plain text data. After that, the binary values are be converted to decimal form, and the Caesar cipher algorithm is applied to the decimal values. After implementing the Caesar cipher algorithm, the cipher matrix algorithm is implemented. The cipher matrix algorithm contains a set of rules that, if used efficiently, can completely protect the data from the attacker. After implementing the cipher matrix algorithm, a cipher text is obtained, and a key is implemented on this cipher matrix text. After implementing the key, the cipher XORation mechanism is implemented on the cipher matrix text, and a ciphertext is obtained.

Algorithm 1: Encryption.

Input: Plain text Output: Cipher text

Convert each character to its equivalent ASCII. Convert each ASCII value to an 8-bit binary. Apply bit reversing in step 2. Apply salting mechanism on bit-reversing results. Convert salted binary values to decimals by making 8-bit pairs. Apply the Caesar cipher algorithm on step 5. Apply the cipher matrix algorithm on Caesar’s cipher results and obtain a cipher matrix text. Obtain a user-defined key and apply the Cipher XORation mechanism on cipher matrix text. Ciphertext.

3.3. Cipher XORation

In Cipher XORation, the cipher matrix text and user-defined key obtained from the Caesar matrix algorithm are XORed. The biggest advantage of Cipher XORation is that the attacker will not understand the technique applied to the bits. Whenever an attacker tries to apply a technique to bits, the attacker will obtain a different result that will be irrelevant. Cipher XORation provides more security to the data with the help of the key, and data decryption is be possible until the valid key is used.

3.4. Data Decryption Process

Various steps are used to convert the ciphertext to plain text; firstly, the ciphertext is converted to ASCII, and then cipher XORation is implemented on the user-identified key, and binary values by converting the ASCII values to binary and decimal values are obtained. After obtaining the decimal values, the cipher matrix algorithm is implemented on the decimal values, and the values matrix text is obtained. Cipher matrix text is a decoded text. After that, the decimal values are obtained by implementing the Caesar cipher mechanism on the cipher matrix text. After obtaining the various decimal values, all decimal values are converted to eight-bit binary. Then, all the bits are reversed by first applying the bit-reversing mechanism, and then, salting is removed from the bit-reversing results, and plain text is obtained, as shown in Figure 9.

In Algorithm 2, first, a ciphertext is taken, and each character of the ciphertext is converted to equivalent ASCII. After converting the ciphertext to ASCII, each ASCII value is converted to an 8-bit binary. After that, the binary of the ciphertext and key are XORed. After XORing the ciphertext and key, the binary values obtained are converted into a decimal by making eight-bit pairs, and the decimal values are converted into equivalent ASCII. After obtaining the various ASCII values, the cipher matrix algorithm is implemented, and the cipher matrix text is obtained. After obtaining the cipher matrix text, the Caesar cipher algorithm is implemented using the P = D − N formula. If the value of P is greater than 255, then mod 255 will be taken, and if the value of P is small, then the P values will be converted into decimals, and then, the eight-bit binary of each decimal will be obtained. After obtaining eight-bit binary values, salting is removed. Then, different decimal values are obtained by implementing a bit-reversing mechanism, and each decimal value is converted to its equivalent ASCII. Then, a plain text is obtained.

Algorithm 2: Decryption.

Input: Cipher text Output: Plain text

Ciphertext Convert each character of ciphertext into ASCII. Convert ASCII values to 8-bit binary. XOR binary values of ciphertext and key. Convert the result of XORation into a decimal by making 8-bit pairs. Convert decimal values to equivalent ASCII. Apply the cipher matrix algorithm on ASCII values. Generate cipher matrix text from cipher matrix algorithm. Convert cipher matrix values to equivalent 1decimal ASCII. Apply P = D − N; while D = Decimal, N = no of shifts. If P > 255, then P mod 255, and go to step 7   else SKIP step-6 and GOTO step 7. Convert each decimal value to binary form. Remove the salting values. Apply the bit-reversing mechanism in step 8. Convert binary values to decimals by making 8-bit pairs. Convert each decimal to equivalent ASCII. Obtain plain text.

4. Experimental Flow

4.1. Encryption Algorithm

Step 1: First, a text is taken for the encryptions, as shown in Figure 10.

Step 2: To convert text to binary, firstly, the plain text is divided into letters, and then, the ASCII of each letter is created, as shown in Figure 11.

Step 3: After converting plain text to ASCII values, each ASCII value is converted to binary values, as shown in Figure 12.

Step 4: All binary bits are reversed to make data more secure, as shown in Figure 13. Due to reversing mechanism, if an attacker tries to decrypt the data, the attacker will keep trying to decrypt the reversed data instead of obtaining the original data.

Step 5: After reversing the bits, salting is applied. The same salting value is applied at the start and end of the binary values, as shown in Figure 14.

Step 6: The binary values obtained after the salting mechanism are split into eight bits and converted into decimal values, as shown in Figure 15.

Step 7: After obtaining decimal values, Caesar’s cipher encryption formula “E = D + N” is applied to each decimal value and obtains the following results, as shown in Figure 16.

Step 8: The decimal values obtained from the Caesar cipher formula are converted to ASCII, and the Caesar cipher text is obtained, as shown in Figure 17.

Step 9: After receiving the Caesar ciphertext, the Cipher Matrix mechanism has been implemented, for which, first of all, the Cipher Matrix has been created, as shown in Figure 18.

Step 10: After creating the cipher matrix table, the rules of the cipher matrix algorithm are applied to the Caesar ciphertext, in which first the text is divided into pairs (P1 to Pn), and according to the cipher matrix rule, the character that cannot be paired is paired with “x”, as shown in Figure 19.

Step 11: After forming groups P1 to Pn, cipher matrix rules are implemented on each pair, as shown in Figure 20, Figure 21, Figure 22, Figure 23 and Figure 24, and different cipher values are obtained.

In Figure 21, pair P1 and pair P2 are not in the same row and column. According to the cipher matrix rules, a rectangle is made parallel to the two characters, and then, these characters are replaced with the opposite character, as shown in Figure 20 and Figure 21.

The characters of pair P3 are in the same row. According to the cipher matrix rules, characters are replaced by the next right character, as shown in Figure 22.

Pair P4 and P5 characters are also not in the same row and column. According to the cipher matrix rules, a rectangle is made parallel to the two characters, and then, these characters are replaced with the opposite character, as shown in Figure 23 and Figure 24.

Step 12: A cipher matrix text is obtained after implementing the cipher matrix rules, as shown in Figure 25.

Step 13: After obtaining the cipher matrix text, the user-defined key is taken. After that, the user-defined key and cipher matrix text are converted to ASCII, as shown in Figure 26 and Figure 27.

Step 14: After the ASCII conversion, ASCII values of cipher matrix text and key are converted to binary, and then, these binary values are XORed, as shown in Figure 28.

Step 15: Pairs of binary values obtained from the XOR operation are formed and converted to decimal form, as shown in Figure 29.

Step 16: After obtaining the decimal values, each decimal value is converted into ASCII values, as shown in Figure 30, and a ciphertext is obtained.

A secure cipher text is shown in Figure 31.

4.2. Decryption Algorithm

Figure 32 shows an encrypted text that has been decrypted.

Step 1: Each ciphertext character is converted to ASCII, as shown in Figure 33.

Step 2: Each value is converted to binary after conversion to ASCII, as shown in Figure 34.

Step 3: After conversion to binary, the binary values of the ciphertext and key are XORed, as shown in Figure 35.

Step 4: The binary values obtained by the XOR operation are converted to decimal, as shown in Figure 36.

Step 5: Different ASCII values are obtained after obtaining the decimal values, as shown in Figure 37.

Step 6: The rules of the cipher matrix are applied to the ASCII values, for which the cipher matrix text is divided into pairs first, as shown in Figure 38.

Step 7: The cipher matrix rules are applied to pairs of encrypted text, after which a text is generated, as shown in Figure 39.

Step 8: After applying the cipher matrix rules, the values obtained from the cipher matrix are converted to ASCII values, as shown in Figure 40.

Step 9: After obtaining the different ASCII values, the Caesar cipher decryption formula “P = D − N” is applied to each decimal value. A Caesar cipher text is thus obtained, as shown in Figure 41.

Step 10: Each decimal value is converted to an eight-bit binary, as shown in Figure 42.

Step 11: After conversion to binary, four-bit salting is removed from the start and end of the binary values, as shown in Figure 43.

Step 12: After removing the salting values, the reversing mechanism is applied to the salting bits, as shown in Figure 44.

Step 13: After reversing the mechanism, each eight-bit binary value is converted to decimal, as shown in Figure 45.

Step 14: After getting the decimal values, each decimal value is converted to ASCII, and a plain text is obtained, as shown in Figure 46.

When the rules of the cipher matrix technique are properly applied, and various techniques such as salting, bit reversing, and Caesar ciphers are used with this technique as used in this paper, it is difficult for the attacker to break the ciphertext security, enabling cloud transmission in a reliable environment.

5. Testing

After developing the algorithm, a tool was developed to check the algorithm’s performance in which, firstly, plain text is taken. After that, the plain text is inserted into the tool, and the encoded text is obtained, as shown in Figure 47. Encoded text is the text derived from the cipher matrix algorithm.

After obtaining the encrypted text, a user-defined key is taken, and with the help of this key, the data are encapsulated. A completely secure ciphertext is obtained, as shown in Figure 47.

To further test the tool’s performance, additional time testing was performed using different characters, numbers, and numeric keys, as shown in Figure 48.

5.1. Various Testing Results

After testing the tool’s performance at different times, different results have been obtained, as shown in

Table 2. Different Testing Results.

Testing	Plain Text	Plain-Text Length (P)	Encoded Text	Key	Cipher Text	Cipher Text Length (C)	Length Difference (L) = C − P
1	N@deem_9$3	10	╬ñ█Bm¼meBi|(	S7_45j	ØôävXã>RØ]IB	12	2
2	WR!tten_by-@Li	14	mºD/ÝHú{ogþÝ°B~¿	Su$k!8	>‗`D╠p¬.Kƪã└Ƞ+îå	16	2
3	$tuD!O$0ne	10	mÜ├e”[eMuÞuÞ|H	WJ&56h	:ðÕPÌ32℘S¦CÇ+¤	14	4
4	@uth0r_is_Z@hra	15	kÚfÕ•Dߥñy┼µ¸jUs	Waj!h@	<êƮ─ÆℑÂ▀╬X¡ªáƻ?R	16	1
5	W@jia_12	8	Ã█ó█mi3ap©	N@d50!	ëøãµ]H}!Ìì	10	2

. In Table 2, different plain text has been taken, and each line text’s length has been identified. The length of the plain text has been represented with “P”. The algorithms have been then implemented on the plain text, and various encrypted texts have been obtained. The user-defined key has been implemented on the encrypted text, and various ciphertexts have been obtained. The length of the cipher text has been represented with “X”.

After obtaining the ciphertext, the length difference between the ciphertext and the plain text is determined. When an attacker tries to decrypt the data, the attacker tries to decrypt each character, but when the ciphertext length is more than the plain text, the attacker will obtain the data derived in parallel for each character, which will be incorrect.

5.2. Time and Space Complexity

The time and space complexity of the tool while encrypting and decrypting the data is determined, as shown in

Table 3. Time and Space Complexity of Encryption Mechanism.

Sr#	Plain Text Length	Encoding Time (s)	Encode Text Length	Key Length	Ciphertext Length	Ciphertext Time (s)	Plain Text Memory Allocation	Cipher Text Memory Allocation
1	10	104.21	12	6	12	25.33	12.4 KB	13.4 KB
2	14	113.43	16	6	16	32.27	13.9KB	14.7 KB
3	10	107.21	14	6	14	28.45	11.7 KB	12.5 KB
4	15	115.37	16	6	16	33.15	16.9 KB	18.2 KB
5	8	93.29	10	6	10	21.52	10.8 KB	11.6 KB

. First, different plain texts are taken, and then, the length of different plain texts is identified. After identifying the length of the plain text, all the plain text is encrypted with the tool, and the encoding time is determined. When the plain text is converted into cipher-encoded text, the length of the encoded text is 20% more than the plain text length.

After determining the ciphertext length, each ciphertext is encrypted with different keys using different key sizes. These keys are implemented on the text with the help of the tool, and a secure ciphertext is obtained. When the encoded text is encrypted with the help of the key, the time complexity is also determined when the cipher text is generated. After converting plain text to ciphertext, memory allocation of plain text and ciphertext is performed, as shown in Table 3.

After encrypting the data, time and space complexity was identified while decrypting the data, in which different ciphertexts were first taken, the ciphertexts were decrypted using different cases, and the decryption time was determined. The memory allocation of the ciphertext and plain text was then determined by identifying the difference between the length of the plain text and the ciphertext when converted to plain text, as shown in

Table 4. Time and Space Complexity of Decryption Mechanism.

Sr#	Ciphertext Length	Key Length	Decryption Time (s)	Cipher Text Memory Allocation	Plain Text Memory Allocation
1	12	6	96.34	13.4 KB	12.3 KB
2	16	6	102.42	14.7 KB	12.8KB
3	14	6	99.12	12.5 KB	11.9 KB
4	16	6	103.29	18.2 KB	15.4 KB
5	10	6	92.54	11.6 KB	10.9 KB

.

5.3. Cryptanalysis

Cryptanalysis is a technique in which an attacker tries to obtain a key using ciphertext or attempts to decipher the ciphertext. A cryptanalysis algorithm was implemented on the ciphertext to check the efficiency of the proposed algorithm. In cryptanalysis, an attacker makes a key prediction by using some rules. However, key prediction is impossible in this proposed algorithm because the data are stored in different phases using different rules. Unless these rules are applied, the data cannot be decrypted.

When the ciphertext length and the plain-text length are not equal, the cryptanalysis mechanism cannot be applied. In

Table 5. Ciphertext Cryptanalysis.

Sr#	Plain Text	Text Length (T)	Ciphertext Length (C)	Length Equalization (T, C)	Key Snatching Algorithm	Key Identification
1	N@deem_9$3	10	12	T ≠ C	No	No
2	WR!tten_by-@Li	14	16	T ≠ C	No	No
3	$tuD!O$0ne	10	14	T ≠ C	No	No
4	@uth0r_is_Z@hra	15	16	T ≠ C	No	No
5	W@jia_12	8	10	T ≠ C	No	No

, different plain-text lengths are identified, and the length of the ciphertext derived from the plain text is also identified. Plain-text length is represented with “T”, while ciphertext length is determined by “C”. After length identification, length equalization is completed, in which plain-text length and ciphertext length are compared. When the lengths “C” and “T” are not equal, the snatching mechanism cannot be implemented. When security is provided with the help of different roles on the ciphertext, no matter how effectively the attacker develops the algorithm, the data cannot be decrypted, nor can the key identification be possible.

6. Comparative Analysis

Various researchers have developed Caesar cipher substitution techniques for securing cloud server data, and through their use, plain text is converted to cipher form. An effective Caesar cipher technique could be developed to secure the data, so an even better technique could be devised to break this technique, which is the biggest problem of the Caesar cipher algorithm, as shown in Table 1.

In paper [33], the researchers modified the Caesar cipher algorithm in which the letters were replaced according to some rules, and they implemented the obtained results on the Goldbach code algorithm in which Caesar cipher text was encoded three times, and a ciphertext as obtained. Instead of repeated encoding in the Goldbach algorithm, if the salting and bit-reversing mechanism were implemented, it would be difficult for the attacker to decrypt the Ciphertext, but in this paper, only characters were replaced, and different times encoding was implemented, which is the drawback of this paper.

In paper [25], the researchers designed a hybrid algorithm to enhance the functionality of the Caesar cipher algorithm in which the Caesar cipher and Vigenère cipher were used, and a cipher text was obtained. If the attacker understands that the algorithms used are the Caesar cipher and Vigenère cipher, then the attacker can quickly develop the data decryption mechanism. If a secure algorithm had been developed instead of a hybrid model, the data could have been provided more security, which is the drawback of this paper.

In paper [34], the researchers developed advanced encryption algorithms using Caesar cipher and stream cipher. Each algorithm was encrypted with the help of a key, and then, a ciphertext was obtained. In this paper, the existing technique was used. If an efficient algorithm was developed instead of using the techniques of existing algorithms, the data could have been saved from the attacker, but in this paper, ciphertext was obtained by combining different algorithms, which is the drawback of this paper.

In paper [35], the researchers developed an algorithm to secure the key used in the Vigenère cipher. The Caesar cipher algorithm was first implemented on the randomized key, the Hill cipher algorithm was applied to the obtained results, and a cipher text was obtained. If instead of using the Caesar cipher and Hill cipher substitution technique to secure the key, an algorithm was developed that makes the decryption of the key impossible when trying to obtain the key by implementing the Kasiski test with the help of this algorithm, then breakage could not happen. However, in this paper, the results of one algorithm were substituted in another algorithm, and a key was obtained, which is the drawback of this paper.

In paper [38], to protect data from man-in-the-middle attacks, the researchers developed a random shift key-generation algorithm that takes a key in decimal format and shifts each character according to that key and then obtains ciphertext by implementing the Caesar cipher algorithm on the results. If instead of replacing the characters according to the key, such techniques are implemented to protect the data from man-in-the-middle attacks, the attack rate can be reduced, but no such technique or algorithm was developed in the paper, which is the drawback of this paper. No paper has developed a technique that can fully protect data, and no paper has defined rules that cannot be easily broken. Encrypting data is not data security. Rather, broadcasting the data in a reliable format is a security that has not been achieved in any paper

Novelty of Proposed Work

In this paper, different techniques are used for data protection in different phases, and a set of rules are defined in each phase. When an attacker tries to access the data or apply cryptanalysis on the ciphertext, the attacker cannot access the original data due to the lack of a rule-based snatching algorithm. Bit-reversing is first implemented to protect plain text data so that the original data can be hidden. A salting mechanism is then implemented on the bit routing so that each value in the eight-bit paring can be divided into different pieces. Then, several time shiftings are performed using the Caesar cipher algorithm on the obtained values. By shifting the number of times, the previous value is replaced with the latest value, and the original data link is lost due to the latest value. Then, the cipher matrix algorithm is implemented on the Caesar cipher algorithm. A set of rules is defined in the cipher matrix algorithm. When the attacker tries to decipher the data, it will be difficult to understand and exploit each rule, which can increase the security of the data. A key is implemented on the cipher matrix text obtained from the cipher matrix algorithm, and a secure ciphertext is obtained with the help of the XORation technique. When the data are protected with the help of such algorithms, no matter how many efficient algorithms are developed to decrypt the data, the data cannot be decrypted, which is the novel contribution of this paper. A complete comparative analysis is shown in

Table 6. Comparative Analysis.

Sr#	1	2	3	4	5	Proposed Work
Reference No.	[33]	[25]	[34]	[35]	[38]
Year	2020	2021	2021	2021	2022
Proposed Algorithm	Implemented Caesar cipher results on the Goldbach algorithm	Implemented Caesar cipher results on the Vigenère cipher algorithm	Implemented Caesar cipher, Playfair, and stream cipher algorithm on plain text	Implemented Caesar cipher and Hill cipher algorithm on key	Random shift key-generation algorithm	Caesar cipher encryption algorithm, cipher matrix algorithm
Novelty	Multiple times encoded Caesar cipher results by the Goldbach algorithm	implemented the best techniques of Caesar cipher and Vigenère cipher	Each algorithm is encrypted with a static key	The key was secured using two algorithms	Shifted each character’s values according to each decimal value	Provided data security in three different ways.
Research Gaps	Obtained Caesar cipher results were encoded iteratively	Used existing techniques of Caesar cipher and Vigenère cipher	No latest technique has been introduced	Key was obtained by substituting the results of one algorithm to another.	Characters shifted according to a key decimal value	Identified all gaps
Proposed Paper Solution	Enhanced data security by salting and bit reversing	Applied the number of time shifting by Caesar cipher	Developed cipher matrix algorithm	Applied key on cipher matrix text for text security	Applied different rules on each algorithm phase	The proposed paper addressed all existing gaps

.

7. Conclusions

After comparing different algorithms and techniques for data transmission, it is concluded that when the data are protected in different phases with the help of different rules, it is impossible for an attacker to develop such roles and decipher the data. When the attacker tries to decrypt the data through cryptanalysis, the attacker is be able to predict each character’s key because the ciphertext and plain-text length are unequal. Data decryption is impossible when the data are divided into different bits due to salting. This paper uses salting and bit-reversing mechanisms to encapsulate the original data. Then, static time value shifting is implemented using the Caesar cipher algorithm on the obtained values. After that, a cipher matrix algorithm is developed to define some rules, and the cipher matrix text is obtained. A cipher matrix algorithm consists of a set of rules. The advantage of this algorithm is that it is not easy for an attacker to decipher, nor can an attacker break into the Caesar cipher algorithm without breaking the mechanism. After that, the data are completely protected by using a key so that each user can encrypt the data with their static key, and only the authentic users can decrypt it. Whenever cloud data are attacked, using different mechanisms makes it difficult for the attacker to understand and break the mechanism. When information is secured from all sides by different phases, the chances of an attack will be low. Cloud server data cannot be secure unless the cloud data encryption algorithm is fully secure. When an attacker tries to decrypt the cipher data, his first attempt is to decrypt each character, for which the attacker uses different techniques and algorithms, but when the number of cipher data is more than the original data, then no matter how efficient the algorithm used by the attacker, the access to the data will be impossible, which is the novelty of this paper.

In the future, the data will be secured in two phases: the plain-text data will be encrypted first, and then, the security mechanism will be developed to encrypt all the data. An asymmetric cryptography algorithm will be developed to encrypt the plain-text data, in which public and private keys and different rules will be used. After that, a cloud shell mechanism will be developed to provide access to cloud data to the right users. Different queries will be designed in a cloud shell mechanism. Access to data will be possible only when valid queries are used. Different rules will be implemented for valid and invalid users, and when an invalid query is entered, the detection mechanism will be implemented on such users.