Linköping electronic conference proceedings, Jun 9, 2022
This paper examines whether deep neural networks (DNN) can learn knownplaintext attacks on plaint... more This paper examines whether deep neural networks (DNN) can learn knownplaintext attacks on plaintext-ciphertextpairs, that were created by encrypting with complete columnar transposition. We propose a new algorithm that extends pure DNN-based prediction with additional post-processing steps to further enhance key prediction quality. Our approach is easily extensible and currently supports key lengths from 2 to 20 characters. Each key length has been empirically evaluated with plain-/ciphertextpairs of different lengths. For plain-and ciphertexts with a length of five times the key length, our algorithm achieves a success rate of 96% which is, to the best of our knowledge, a new state of the art on deep-learning-based known-plaintext attacks against columnar transposition.
In this paper, we study the impact of language models (LM) on decipherment of historical homophon... more In this paper, we study the impact of language models (LM) on decipherment of historical homophonic substitution ciphers. In particular, we investigate if decipherment by using hill-climbing and simulated annealing can benefit from LMs generated from historical texts in general and century-specific texts in particular. We carry out experiments on homophonic substitution ciphers with English and German as plaintext languages. We take into account ciphertext length as well as n-gram size of the LMs. We compare the results on decipherment based on historical LMs with large LMs generated from modern texts. The results show that using historical LMs in decipherment of homophonic substitution ciphers leads to significantly better performance on ciphertext produced in the 17th century or earlier, and centuryspecific language models yield better results on longer and older ciphertexts.
The cross-disciplinary nature of historical cryptology involves the challenge to find a terminolo... more The cross-disciplinary nature of historical cryptology involves the challenge to find a terminology that is both consistent and accepted across the different disciplines and applicable in the single fields. In this paper, we propose a terminology based on concise principles developed by an interdisciplinary group of researchers. We present terms prominent in the study of historical cryptology, define them, and illustrate their usage. Our goal is to initiate and/or continue the discussion of how we use various terms for different types of historical encrypted sources, and their study. Our hope is that this paper will contribute to consistent and systematic usage of terms in the HistoCrypt community.
Proceedings of the 5th International Conference on Historical Cryptology HistoCrypt 2022, 2022
In this paper, we present an empirical study on plaintext entities in historical cipher keys from... more In this paper, we present an empirical study on plaintext entities in historical cipher keys from the 15th to the 18th century to shed light on what linguistic entities have been chosen for encryption. We focus mainly on the nomenclature part of the keys describing longer elements than the plaintext alphabet. We show that the chosen plaintext entities to be encoded varied over time. Nomenclatures developed from short lists consisting of names for persons and/or locations to longer, more advanced dictionaries and eventually to codebooks containing a highly diverse and advanced set of linguistic entities.
We give on overview of the development of European historical cipher keys originating from early ... more We give on overview of the development of European historical cipher keys originating from early Modern times. We describe the nature and the structure of the keys with a special focus on the nomenclatures. We analyze what was encoded and how and take into account chronological and regional differences. The study is based on the analysis of over 1,600 cipher keys, collected from archives and libraries in 10 European countries. We show that historical cipher keys evolved over time and became more secure, shown by the symbol set used for encoding, the code length and the code types presented in the key, the size of the nomenclature, as well as the diversity and complexity of linguistic entities that are chosen to be encoded.
We give on overview of the development of European historical cipher keys originating from early ... more We give on overview of the development of European historical cipher keys originating from early Modern times. We describe the nature and the structure of the keys with a special focus on the nomenclatures. We analyze what was encoded and how and take into account chronological and regional differences. The study is based on the analysis of over 1,600 cipher keys, collected from archives and libraries in 10 European countries. We show that historical cipher keys evolved over time and became more secure, shown by the symbol set used for encoding, the code length and the code types presented in the key, the size of the nomenclature, as well as the diversity and complexity of linguistic entities that are chosen to be encoded.
ABSTRACT Chaocipher is a manual encryption method designed by John F. Byrne in 1918. Until he pas... more ABSTRACT Chaocipher is a manual encryption method designed by John F. Byrne in 1918. Until he passed away in 1960, Byrne fervently believed that his cipher system was unbreakable, regardless of the amount of material available to a cryptanalyst. For several decades, he tried (unsuccessfully), to propose the Chaocipher to government agencies. In 1953, he exposed his Chaocipher in his autobiography, Silent Years, providing several examples of texts encrypted with Chaocipher as challenges, but without divulging the inner workings of the cipher. Those were made public only in 2010, when Byrne’s family donated the entire corpus of Chaocipher papers to the National Cryptologic Museum (NCM) in Fort Meade. A known-plaintext method for recovering the key settings, given sufficient matching plaintext and ciphertext, was published in 2010. However, to date, no method for the cryptanalysis of a single ciphertext-only Chaocipher message has been proposed, nor for the cryptanalysis of short messages “in-depth,” i.e., multiple messages generated with the same initial key settings. In this article, the authors present a new hillclimbing algorithm for a ciphertext-only cryptanalysis of Chaocipher in-depth messages. This algorithm is based on a “divide-and-conquer” approach and the use of the Index of Coincidence. It takes advantage of a major weakness in the design of the cipher. This previously unknown weakness may have been the reason why William F. Friedman, the inventor of the Index of Coincidence, rejected Byrne’s offer for the use of Chaocipher by the U.S. government. Additionally, the authors present a known-plaintext attack for short in-depth messages, as well as the solution for Lou Kruh’s and Cipher Deavours’s alternate Exhibit 5, also known as “Exhibit 6.” Finally, the authors reevaluate the security of the Chaocipher in view of those findings, with the conclusion that in its classic form, as designed by Byrne, the Chaocipher was a relatively weak cipher, despite Byrne’s rather strong assertions to the contrary.
2015 IEEE Conference on Communications and Network Security (CNS), 2015
Many contexts dealing with sensitive information require high-robustness and high-assurance certi... more Many contexts dealing with sensitive information require high-robustness and high-assurance certified security systems that should not be affected by known vulnerabilities. Covert channels are illicit paths that could be exploited by attackers to convey illicit data flows that contravene the security policies. Many implementations of the so-called Covert Storage Channels exist, whereas no implementation of Covert Timing Channels is available. In this paper, we first discuss an Open-Source Covert Timing Channel implementation, describing in detail our innovative approach. Then, we analyze real TCP traffic in the presence of our covert channel for three scenarios of interest, varying the number of hops and round trip time of the connections. The results, from real network traffic monitoring, confirm the validity of our open-source covert timing channel implementation for hidden TCP traffic analysis, in different environmental and operating network conditions.
In this paper, we present two new algorithms for distributed volunteer computing. Both algorithms... more In this paper, we present two new algorithms for distributed volunteer computing. Both algorithms have in common, that they are self-organized. Due to the self-* properties of our algorithms, the reliability of the distributed computing is increased. The network, which builds the basis for the distribution, is an unstructured peer-to-peer (p2p) network. Since there is no central managing unit and no structured addressable overlay in such networks, each peer has to self-organize the job selecting process, the distribution of jobs, and the merging of finished jobs obtained from neighbors. After presenting our two algorithms, namely the Epoch-Distribution-Algorithm and the Sliding-Window-Distribution-Algorithm, we present open questions. Since these algorithms are strongly based on self-organization and peers are working absolute autonomously and other some self-* properties, the algorithms can be classified into to world of the organic computing.
ABSTRACT In the last months of World War I (WW I), the German Army and diplomatic services used t... more ABSTRACT In the last months of World War I (WW I), the German Army and diplomatic services used the ADFGVX hand-cipher system to encrypt radio messages between Germany and its outposts and stations in the Balkans, the Black Sea, and in the Middle East. Hundreds of cryptograms were intercepted from July to December 1918 by British and U.S. military intelligence, who were able to recover most of the keys and decipher most of the cryptograms using manual cryptanalysis methods. Fortunately, the original cryptograms have been preserved by James Rives Childs, the U.S. officer assigned to G.2 A.6, the SIGINT section of American Expeditionary Forces (AEF) in Paris, and they appear in his book, General Solution of the ADFGVX Cipher System, published by Aegean Press Park in 2000. In this article, the authors present the results of an effort toward the complete cryptanalysis of the messages, and an analysis of their contents. The authors present a new computerized method for the ciphertext-only cryptanalysis of ADFGVX messages which they developed for that purpose. They also provide details on how all the keys were recovered and almost all the messages decrypted, despite the low quality of significant parts of the intercepted material. The analysis of the messages in their historical context provides a unique insight into key events, such as the withdrawal of the German troops from Romania, and the impact of the Kiel Mutiny on communications. Both events had major political and military consequences for Germany in the Eastern Front. Cryptanalysis allowed the Entente Powers to closely monitor those events as well as others developments, also highlighting inherent weaknesses in German cryptographic and cryptanalytic capabilities. The authors plan to publish online all the decrypted messages.
The deciphered Ramanacoil ciphertext reveals two Dutch East India Company letters, from 1674 that... more The deciphered Ramanacoil ciphertext reveals two Dutch East India Company letters, from 1674 that are, in retrospect, already known in the National Archives as plaintext letters. The letters are written in Dutch. We have been able to relate them. The first letter, from Van Goens senior from Sri Lanka to the Lords Seventeen in The Netherlands, was most important to the sender. He sent his personal secretary Leeuwenson overland with the ciphertext in his pocket and its key in his head. And with additional oral information that had to be delivered in person. Van Goens senior requested to conquer the whole of Sri Lanka, the island Ramanacoil and coastal area around it along with 1,000 more soldiers. This paper shows that by sending Leeuwenson, Van Goens senior wanted to repeat his most successful 'Vertoog' from 1655. Substantiate his strategic goals and get approval for them from the Lords Seventeen. In 1655 he got a 'Go!' and twenty years later in 1675 he got a 'No!'. The zeitgeist of expansion had changed. Ramanacoil was a bridge too far.
This paper describes the current progress of our research in the area of breaking homophonic subs... more This paper describes the current progress of our research in the area of breaking homophonic substitution ciphers. Furthermore, it presents the state-of-the-art of cryptanalyzing this kind of cipher. There is a huge gap between the success rate of methods published in according research papers and the success rate of already available tools on the Internet. This paper also presents a small general taxonomy of monoalphabetic substitution ciphers. Finally, it shows how we broke different homophonic substitution ciphers in an automatic as well as in a semi-automatic way.
The difficulty of solving classical ciphers varies between very easy and very hard. For example, ... more The difficulty of solving classical ciphers varies between very easy and very hard. For example, monoalphabetic substitution ciphers can be solved easily by hand. More complex ciphers like the polyalphabetic Vigenère cipher, are harder to solve and the solution by hand takes much more time. Machine ciphers like the Enigma rotor machine, are nearly impossible to be solved only by hand. To support researchers, cryptanalysts, and historians analyzing ciphers, the open-source software CrypTool 2 (CT2) was implemented. It contains a broad set of tools and methods to automate the cryptanalysis of different (classical and modern) ciphers. In this paper, we present a step-by-step approach for analyzing classical ciphers and breaking these with the help of the tools in CT2. The primary goals of this paper are: (1) Introduce historians and non-computer scientists to classical encryption, (2) give an introduction to CT2, enabling them to break ciphers by their own, and (3) present our future p...
In this German seminar paper, which was written in the year 2011 at the University of Duisburg fo... more In this German seminar paper, which was written in the year 2011 at the University of Duisburg for a Bachelor Colloquium in Applied computer science, we show a brief overview of the Rational Unified Process (RUP). Thus, interested students or generally interested people in software development gain a first impression of RUP. The paper includes a survey and overview of the underlying process structure, the phases of the process, its workflows, and describes the always by the RUP developers postulated "best practices" of software development.
Linköping electronic conference proceedings, Jun 9, 2022
This paper examines whether deep neural networks (DNN) can learn knownplaintext attacks on plaint... more This paper examines whether deep neural networks (DNN) can learn knownplaintext attacks on plaintext-ciphertextpairs, that were created by encrypting with complete columnar transposition. We propose a new algorithm that extends pure DNN-based prediction with additional post-processing steps to further enhance key prediction quality. Our approach is easily extensible and currently supports key lengths from 2 to 20 characters. Each key length has been empirically evaluated with plain-/ciphertextpairs of different lengths. For plain-and ciphertexts with a length of five times the key length, our algorithm achieves a success rate of 96% which is, to the best of our knowledge, a new state of the art on deep-learning-based known-plaintext attacks against columnar transposition.
In this paper, we study the impact of language models (LM) on decipherment of historical homophon... more In this paper, we study the impact of language models (LM) on decipherment of historical homophonic substitution ciphers. In particular, we investigate if decipherment by using hill-climbing and simulated annealing can benefit from LMs generated from historical texts in general and century-specific texts in particular. We carry out experiments on homophonic substitution ciphers with English and German as plaintext languages. We take into account ciphertext length as well as n-gram size of the LMs. We compare the results on decipherment based on historical LMs with large LMs generated from modern texts. The results show that using historical LMs in decipherment of homophonic substitution ciphers leads to significantly better performance on ciphertext produced in the 17th century or earlier, and centuryspecific language models yield better results on longer and older ciphertexts.
The cross-disciplinary nature of historical cryptology involves the challenge to find a terminolo... more The cross-disciplinary nature of historical cryptology involves the challenge to find a terminology that is both consistent and accepted across the different disciplines and applicable in the single fields. In this paper, we propose a terminology based on concise principles developed by an interdisciplinary group of researchers. We present terms prominent in the study of historical cryptology, define them, and illustrate their usage. Our goal is to initiate and/or continue the discussion of how we use various terms for different types of historical encrypted sources, and their study. Our hope is that this paper will contribute to consistent and systematic usage of terms in the HistoCrypt community.
Proceedings of the 5th International Conference on Historical Cryptology HistoCrypt 2022, 2022
In this paper, we present an empirical study on plaintext entities in historical cipher keys from... more In this paper, we present an empirical study on plaintext entities in historical cipher keys from the 15th to the 18th century to shed light on what linguistic entities have been chosen for encryption. We focus mainly on the nomenclature part of the keys describing longer elements than the plaintext alphabet. We show that the chosen plaintext entities to be encoded varied over time. Nomenclatures developed from short lists consisting of names for persons and/or locations to longer, more advanced dictionaries and eventually to codebooks containing a highly diverse and advanced set of linguistic entities.
We give on overview of the development of European historical cipher keys originating from early ... more We give on overview of the development of European historical cipher keys originating from early Modern times. We describe the nature and the structure of the keys with a special focus on the nomenclatures. We analyze what was encoded and how and take into account chronological and regional differences. The study is based on the analysis of over 1,600 cipher keys, collected from archives and libraries in 10 European countries. We show that historical cipher keys evolved over time and became more secure, shown by the symbol set used for encoding, the code length and the code types presented in the key, the size of the nomenclature, as well as the diversity and complexity of linguistic entities that are chosen to be encoded.
We give on overview of the development of European historical cipher keys originating from early ... more We give on overview of the development of European historical cipher keys originating from early Modern times. We describe the nature and the structure of the keys with a special focus on the nomenclatures. We analyze what was encoded and how and take into account chronological and regional differences. The study is based on the analysis of over 1,600 cipher keys, collected from archives and libraries in 10 European countries. We show that historical cipher keys evolved over time and became more secure, shown by the symbol set used for encoding, the code length and the code types presented in the key, the size of the nomenclature, as well as the diversity and complexity of linguistic entities that are chosen to be encoded.
ABSTRACT Chaocipher is a manual encryption method designed by John F. Byrne in 1918. Until he pas... more ABSTRACT Chaocipher is a manual encryption method designed by John F. Byrne in 1918. Until he passed away in 1960, Byrne fervently believed that his cipher system was unbreakable, regardless of the amount of material available to a cryptanalyst. For several decades, he tried (unsuccessfully), to propose the Chaocipher to government agencies. In 1953, he exposed his Chaocipher in his autobiography, Silent Years, providing several examples of texts encrypted with Chaocipher as challenges, but without divulging the inner workings of the cipher. Those were made public only in 2010, when Byrne’s family donated the entire corpus of Chaocipher papers to the National Cryptologic Museum (NCM) in Fort Meade. A known-plaintext method for recovering the key settings, given sufficient matching plaintext and ciphertext, was published in 2010. However, to date, no method for the cryptanalysis of a single ciphertext-only Chaocipher message has been proposed, nor for the cryptanalysis of short messages “in-depth,” i.e., multiple messages generated with the same initial key settings. In this article, the authors present a new hillclimbing algorithm for a ciphertext-only cryptanalysis of Chaocipher in-depth messages. This algorithm is based on a “divide-and-conquer” approach and the use of the Index of Coincidence. It takes advantage of a major weakness in the design of the cipher. This previously unknown weakness may have been the reason why William F. Friedman, the inventor of the Index of Coincidence, rejected Byrne’s offer for the use of Chaocipher by the U.S. government. Additionally, the authors present a known-plaintext attack for short in-depth messages, as well as the solution for Lou Kruh’s and Cipher Deavours’s alternate Exhibit 5, also known as “Exhibit 6.” Finally, the authors reevaluate the security of the Chaocipher in view of those findings, with the conclusion that in its classic form, as designed by Byrne, the Chaocipher was a relatively weak cipher, despite Byrne’s rather strong assertions to the contrary.
2015 IEEE Conference on Communications and Network Security (CNS), 2015
Many contexts dealing with sensitive information require high-robustness and high-assurance certi... more Many contexts dealing with sensitive information require high-robustness and high-assurance certified security systems that should not be affected by known vulnerabilities. Covert channels are illicit paths that could be exploited by attackers to convey illicit data flows that contravene the security policies. Many implementations of the so-called Covert Storage Channels exist, whereas no implementation of Covert Timing Channels is available. In this paper, we first discuss an Open-Source Covert Timing Channel implementation, describing in detail our innovative approach. Then, we analyze real TCP traffic in the presence of our covert channel for three scenarios of interest, varying the number of hops and round trip time of the connections. The results, from real network traffic monitoring, confirm the validity of our open-source covert timing channel implementation for hidden TCP traffic analysis, in different environmental and operating network conditions.
In this paper, we present two new algorithms for distributed volunteer computing. Both algorithms... more In this paper, we present two new algorithms for distributed volunteer computing. Both algorithms have in common, that they are self-organized. Due to the self-* properties of our algorithms, the reliability of the distributed computing is increased. The network, which builds the basis for the distribution, is an unstructured peer-to-peer (p2p) network. Since there is no central managing unit and no structured addressable overlay in such networks, each peer has to self-organize the job selecting process, the distribution of jobs, and the merging of finished jobs obtained from neighbors. After presenting our two algorithms, namely the Epoch-Distribution-Algorithm and the Sliding-Window-Distribution-Algorithm, we present open questions. Since these algorithms are strongly based on self-organization and peers are working absolute autonomously and other some self-* properties, the algorithms can be classified into to world of the organic computing.
ABSTRACT In the last months of World War I (WW I), the German Army and diplomatic services used t... more ABSTRACT In the last months of World War I (WW I), the German Army and diplomatic services used the ADFGVX hand-cipher system to encrypt radio messages between Germany and its outposts and stations in the Balkans, the Black Sea, and in the Middle East. Hundreds of cryptograms were intercepted from July to December 1918 by British and U.S. military intelligence, who were able to recover most of the keys and decipher most of the cryptograms using manual cryptanalysis methods. Fortunately, the original cryptograms have been preserved by James Rives Childs, the U.S. officer assigned to G.2 A.6, the SIGINT section of American Expeditionary Forces (AEF) in Paris, and they appear in his book, General Solution of the ADFGVX Cipher System, published by Aegean Press Park in 2000. In this article, the authors present the results of an effort toward the complete cryptanalysis of the messages, and an analysis of their contents. The authors present a new computerized method for the ciphertext-only cryptanalysis of ADFGVX messages which they developed for that purpose. They also provide details on how all the keys were recovered and almost all the messages decrypted, despite the low quality of significant parts of the intercepted material. The analysis of the messages in their historical context provides a unique insight into key events, such as the withdrawal of the German troops from Romania, and the impact of the Kiel Mutiny on communications. Both events had major political and military consequences for Germany in the Eastern Front. Cryptanalysis allowed the Entente Powers to closely monitor those events as well as others developments, also highlighting inherent weaknesses in German cryptographic and cryptanalytic capabilities. The authors plan to publish online all the decrypted messages.
The deciphered Ramanacoil ciphertext reveals two Dutch East India Company letters, from 1674 that... more The deciphered Ramanacoil ciphertext reveals two Dutch East India Company letters, from 1674 that are, in retrospect, already known in the National Archives as plaintext letters. The letters are written in Dutch. We have been able to relate them. The first letter, from Van Goens senior from Sri Lanka to the Lords Seventeen in The Netherlands, was most important to the sender. He sent his personal secretary Leeuwenson overland with the ciphertext in his pocket and its key in his head. And with additional oral information that had to be delivered in person. Van Goens senior requested to conquer the whole of Sri Lanka, the island Ramanacoil and coastal area around it along with 1,000 more soldiers. This paper shows that by sending Leeuwenson, Van Goens senior wanted to repeat his most successful 'Vertoog' from 1655. Substantiate his strategic goals and get approval for them from the Lords Seventeen. In 1655 he got a 'Go!' and twenty years later in 1675 he got a 'No!'. The zeitgeist of expansion had changed. Ramanacoil was a bridge too far.
This paper describes the current progress of our research in the area of breaking homophonic subs... more This paper describes the current progress of our research in the area of breaking homophonic substitution ciphers. Furthermore, it presents the state-of-the-art of cryptanalyzing this kind of cipher. There is a huge gap between the success rate of methods published in according research papers and the success rate of already available tools on the Internet. This paper also presents a small general taxonomy of monoalphabetic substitution ciphers. Finally, it shows how we broke different homophonic substitution ciphers in an automatic as well as in a semi-automatic way.
The difficulty of solving classical ciphers varies between very easy and very hard. For example, ... more The difficulty of solving classical ciphers varies between very easy and very hard. For example, monoalphabetic substitution ciphers can be solved easily by hand. More complex ciphers like the polyalphabetic Vigenère cipher, are harder to solve and the solution by hand takes much more time. Machine ciphers like the Enigma rotor machine, are nearly impossible to be solved only by hand. To support researchers, cryptanalysts, and historians analyzing ciphers, the open-source software CrypTool 2 (CT2) was implemented. It contains a broad set of tools and methods to automate the cryptanalysis of different (classical and modern) ciphers. In this paper, we present a step-by-step approach for analyzing classical ciphers and breaking these with the help of the tools in CT2. The primary goals of this paper are: (1) Introduce historians and non-computer scientists to classical encryption, (2) give an introduction to CT2, enabling them to break ciphers by their own, and (3) present our future p...
In this German seminar paper, which was written in the year 2011 at the University of Duisburg fo... more In this German seminar paper, which was written in the year 2011 at the University of Duisburg for a Bachelor Colloquium in Applied computer science, we show a brief overview of the Rational Unified Process (RUP). Thus, interested students or generally interested people in software development gain a first impression of RUP. The paper includes a survey and overview of the underlying process structure, the phases of the process, its workflows, and describes the always by the RUP developers postulated "best practices" of software development.
Uploads
Papers by Nils Kopal