Search

During the Coincheck incident, which recorded the largest damages in cryptocurrency history in 2018, it was demonstrated that using Mosaic token can have a certain effect. Although it seems attractive to employ tokens as countermeasures for cryptocurrency leakage, Mosaic is a specific token for the New Economy Movement (NEM) cryptocurrency and is not employed for other blockchain systems or cryptocurrencies. Moreover, although some volunteers tracked leaked NEM using Mosaic in the CoinCheck incident, it would be better to verify that the volunteers can be trusted. Simultaneously, if someone (e.g., who stole cryptocurrencies) can identify the volunteers, then that person or organization may be targets of them. In this paper, we propose an anonymous trust-marking scheme on blockchain systems that is universally applicable to any cryptocurrency. In our scheme, entities called token admitters are allowed to generate tokens adding trustworthiness or untrustworthiness to addresses. Anyone can anonymously verify whether these tokens were issued by a token admitter. Simultaneously, only the designated auditor and no one else, including nondesignated auditors, can identify the token admitters. Our scheme is based on accountable ring signatures and commitment, and is implemented on an elliptic curve called Curve25519, and we confirm that both cryptographic tools are efficient. Moreover, we also confirm that our scheme is applicable to Bitcoin, Ethereum, and NEM., An extended abstract appeared at the 3rd IEEE International Conference on Blockchain and Cryptocurrency, ICBC 2021. This is the full version

In recent years, blockchain has become widely used as a new platform for information systems. Blockchain differs from conventional server-based systems in that it is tamper-resistant and highly available. However, the advantages of blockchain also work to the advantage of attackers. There is an area in the blockchain where users can embed arbitrary data via transactions, and prior works have shown that this area can be exploited for botnet C&C communications. In order to analyze what other types of attacks could develop from the contamination of the blockchain, we constructed specific attack scenarios and analyzed the ease of such attacks by conducting demonstrations experiments. We also proposed countermeasures against the attack scenarios and verified the effectiveness of these by conducting demonstration experiments.

To obtain useful results in machine learning, it is required to collect data from multiple institutions and learn with larger-scale data. However, data collected from multiple institutions may contain a lot of personal information and should not be explicitly shared. The existing research has proposed various methods to protect privacy by using encryption or anonymization, but encryption causes large computational costs, and anonymization may greatly reduce the usefulness of data. In this research, we propose a privacy protection method using dimensionality reduction that is difficult to reverse while maintaining the high usefulness of data. The main idea of our method is that combining dimensionality reduction algorithms with noise addition is useful for privacy-preserving data analysis with high accuracy. Furthermore, we evaluate the effectiveness and security of this method and show the utility of the proposed method.

Maintaining a balance between anonymity and traceability is a fundamental issue in privacy-preserving systems. Isshiki et al. proposed an identity management system based on group signatures in which a service provider anonymously determines whether or not users of the service are legitimate members, and only a bill collector can identify users for the purposes of sending them invoices. It is particularly worth noting that, under the Isshiki system, the service provider is not required to manage personal information such as user lists, which allows the system to outperform other in terms of preserving user privacy and managing personal information leakage risk. It is also noteworthy that the Isshiki system only considers cases in which the bill collector identifies users who have used the service and that, in fact, identified users who ignore invoices can use the service for free. In this paper, we extended the Isshiki system by adding a smart contract-enabled enforcement bill collection functionality. Under this functionality, deposits made by users who do not pay a service fee are automatically transferred to the bill collector. Because of their centralized structure, group signatures are not suitable to blockchain systems, therefore, the proposed system employs accountable ring signatures as building blocks. The privacy-preserving enforced bill collection system is implemented using the accountable ring signature scheme developed by Bootle et al. and Ethereum smart contracts. To reduce the gas costs associated with running smart contracts, the smart contract is not run unless the user ignores an invoice, and basic procedures are run via an off-chain channel. To avoid the use of heavy cryptographic algorithms in carrying out the accountable ring signature scheme for running smart contracts, we employed standard elliptic curve digital signature algorithm (ECDSA) signatures without especially changing the state to be verified in smart contracts.

With recent technological developments, it has become natural for personal computers and Internet of Things(IoT) devices, such as smartphones and tablets, to remain constantly connected to the Internet. Malicious attackers are known to abuse malware to achieve their nefarious purposes, necessitating the implementation of defense systems as protection. Methods such as machine learning-based techniques, which have been utilized with great success in various fields such as image recognition and processing, and voice recognition, are used to prevent cyberattacks caused by malware. However, several adversarial attack methods have been proposed in recent years to induce malfunctions in machine learning-based models. In this study, we focus on malware detection field and treat the aforementioned issue from the perspectives of both attackers and defenders; subsequently, we propose a novel adversarial attack method, named IMAGE_RESOURCE attack, and a robust malware detection model, respectively, using dimension reduction and machine learning techniques. The robustness of the proposed model is evaluated using portable execution (PE) surface information obtained from the FFRI Dataset 2018. During robustness evaluation, distances (e.g., Euclidean distance) between the malware and benign files are measured, and the effectiveness of IMAGE_RESOURCE attack is estimated. Thus, we establish the effectiveness and superiority of the proposed model in terms of detection accuracy and robustness.

The number of Internet of Things (IoT) devices has increased recently and will increase further in the future. IoT devices are vulnerable, so vendors update and distribute firmware. At the time of firmware distribution, a huge number of IoT devices may connect to the firmware server and excessive network traffic may occur. Consequently, a method using a peer-to-peer file sharing system and a blockchain has been proposed. One study proposed an update scheme considering incentives for distributors to help with updating. However, this method requires the IoT devices, which have limited processing power, to decrypt encrypted files, and the distribution nodes to pay the transaction cost. This paper proposes a firmware distribution method that provides incentives for distributors to help with distribution to reduce the gas costs, using a smart contract and access control based on update records. The additional computations performed by IoT devices and distributors key management were reduced compared to those in previous studies by using access control instead of encryption. Further, the proposed approach was implemented and evaluated using a Raspberry Pi and laptop. The evaluation demonstrated that the gas cost per update was successfully lowered compared to the costs in previous studies.

Trust of user address (i.e., wallet address) in Bitcoin is dependent on not who user is but the amount of Bitcoin currency in user’s wallet address. So we cannot necessarily trust a Bitcoin user with wallet address. Actually, there is no method for assigning trust to user addresses in Bitcoin. Although we can provide trust to user address by inserting certificate into Bitcoin transaction using OP_RETURN, it still has some problems about wastefulness. In this paper, we propose a new method that efficiently assigns trust to user address using Bitcoin transfer only, which does not need to use OP_RETURN. Our proposed method can realize a trust providing mechanism for “Blockchain 1.0” without adding any new function. Main idea of our method is that “token coin” (Bitcoin transfer itself) can be used as a trust providing mechanism. For feasibility study, we experimentally show the trust assigning in the Bitcoin testnet. Furthermore, our method can be easily adopted into other cryptocurrencies.

Blockchain is a service operated by a peer-to-peer type distributed network, and protocol control such as JSON-RPC is implemented as the interface for flexibility and operability. However, attacks that use protocol control against vulnerable and unmanaged interfaces have been reported. One of the methods to track cyber attacks on such a malicious user's network service is a honeypot that imitates the service and acquires attacker's behavior information. In this research, focusing on the Ethereum network, the behavior of malicious users is clarified using malicious communication history sent to simple honeypots installed in nine countries, Ethereum network information and darknet arrival packets. By analyzing these, the behavior of attackers and the tendency of requests were elucidated, and primary safety measures were established.

Blockchain is a service operated by a peer-to-peer type distributed network, and protocol control such as JSON-RPC is implemented as the interface for flexibility and operability. However, attacks that use protocol control against vulnerable and unmanaged interfaces have been reported. One of the methods to track cyber attacks on such a malicious user’s network service is a honeypot that imitates the service and acquires attacker’s behavior information. In this research, focusing on the Ethereum network, the behavior of malicious users is clarified using malicious communication history sent to simple honeypots installed in nine countries, Ethereum network information and darknet arrival packets. By analyzing these, the behavior of attackers and the tendency of requests were elucidated, and primary safety measures were established.

Drive-by-download attacks exploit vulnerabilities in web browsers, and users are unnoticeably downloading malware which accesses to the compromised websites. A number of detection approaches and tools against such attacks have been proposed so far. Especially, it is becoming easy to specify vulnerabilities of attacks, because researchers well analyze the trend of various attacks. Unfortunately, in the previous schemes, vulnerability information has not been used in the detection/prediction approaches of drive-by-download attacks. In this paper, we propose a prediction approach of "malware downloading" during drive-by-download attacks (approach-I), which uses vulnerability information. Our experimental results show our approach-I achieves the prediction rate (accuracy) of 92%, FNR of 15% and FPR of 1.0% using Naive Bayes. Furthermore, we propose an enhanced approach (approach-II) which embeds Opcode analysis (dynamic analysis) into our approach-I (static approach). We implement our approach-I and II, and compare the three approaches (approach-I, II and Opcode approaches) using the same datasets in our experiment. As a result, our approach-II has the prediction rate of 92%, and improves FNR to 11% using Random Forest, compared with our approach-I.

As data leakage accidents occur every year, the security of confidential information is becoming increasingly important. Remote Access Trojans (RAT), a kind of spyware, are used to invade the PC of a victim through targeted attacks. After the intrusion, the attacker can monitor and control the victim's PC remotely, to wait for an opportunity to steal the confidential information. Since it is hard to prevent the intrusion of RATs completely, preventing confidential information being leaked back to the attacker is the main issue. Various existing approaches introduce different network behaviors of RAT to construct detection systems. Unfortunately, two challenges remain: one is to detect RAT sessions as early as possible, the other is to remain a high accuracy to detect RAT sessions, while there exist normal applications whose traffic behave similarly to RATs. In this paper, we propose a novel approach to detect RAT sessions in the early stage of communication. To differentiate network behaviors between normal applications and RAT, we extract the features from the traffic of a short period of time at the beginning. Afterward, we use machine learning techniques to train the detection model, then evaluate it by K-Fold cross-validation. The results show that our approach is able to detect RAT sessions with a high accuracy. In particular, our approach achieves over 96% accuracy together with the FNR of 10% by Random Forest algorithm, which means that our approach is valid to detect RAT sessions in the early stage of communication.

Information is increasing quickly, database owners have tendency to outsource their data to an external service provider called Cloud Computing. Using Cloud, clients can remotely store their data without burden of local data storage and maintenance. However, such service provider is untrusted, therefore there are some challenges in data security: integrity, availability and confidentiality. Since integrity and availability are prerequisite conditions of the existence of a system, we mainly focus on them rather than confidentiality. To ensure integrity and availability, researchers have proposed network coding-based POR (Proof of Retrievability) schemes that enable the servers to demonstrate whether the data is retrievable or not. However, most of network coding-based POR schemes are inefficient in data checking and also cannot prevent a common attack in POR: small corruption attack. In this paper, we propose a new network coding-based POR scheme using dispersal code in order to reduce cost in checking phase and also to prevent small corruption attack.

In this paper, we propose a simple random key pre-distribution (RKP) scheme with self-healing for multiphase wireless sensor networks (WSNs) without even lightweight operations such as a hash function, in which the link compromised by node-capture attack in a WSN automatically self-heals with time. Our scheme can enhance the resiliency of existing RKP schemes by only updating the key pool of a server with time. As a result, processing of each sensor is still the same as in the RKP schemes. More interestingly, round synchronization is not necessary to be implemented in a node. Our contribution is the simple scheme with security evaluation, in which we conduct analytical evaluation and a simulation experiment of our scheme, and results obtained from both analysis and simulations indicate that our scheme is effective in self-healing.

In INFOCOM 2010, Pietro, Oligeri, Soriente, and Tsudik (POST) proposed an intrusion-resilient system with forward and backward secrecy in mobile Unattended Wireless Sensor Networks (UWSNs), where sensors move according to some mobility model (random jump model and random waypoint model). In the POST scheme, each sensor encrypts its ephemeral key $K$ as a plaintext by using the sink's public key, and sends this cipher text and the encrypted sensed data by $K$. Although the POST scheme recommends the hybrid encryption, it does not follow the conventional hybrid encryption usage, i.e., the POST scheme is not necessarily secure. More concretely, $K$ must be regarded as a plaintext of the underlying public key system, and therefore the POST scheme requires at least one more encryption procedure (i.e, encryptions for both $K$ and the data) compared with the conventional hybrid encryption procedure. In this paper, we scrutinize the original POST intrusion-resilient system. We set deployed information as a seed used for generating a random number (which is applied for public key encryption). This procedure follows the conventional hybrid encryption usage, and random-number-leakage problem does not occur. In conclusion, we improve the POST scheme from the viewpoint of both security and efficiency without spoiling significant benefit points of the original one.

Virtual Machine Monitors (VMMs), also called hypervisors, can be used to construct a trusted computing base (TCB) enhancing the security of existing operating systems. The complexity of a VMM-based TCB causes the high risk of security vulnerabilities. Therefore, this paper proposes a two-step execution mechanism to reduce the complexity of a VMM-based TCB. We propose a method to separate a conventional VMM-based TCB into the following two parts: (1) A thin hypervisor with security services and (2) A special guest OS for security preprocessing. A special guest OS performing security tasks can be executed in advance. After shutting down the special guest OS, a hypervisor obtains preprocessing security data and next boots a target guest OS to be protected. Thus, the proposed two-step execution mechanism can reduce run-time codes of a hypervisor. This paper shows a design, a prototype implementation and measurement results of lines of code using BitVisor, a VMM-based TCB we have developed.

In recent years, as the data processed by governmental or commercial organizations increases, cases involving information leak have risen. It is difficult to control information on many distributed end-point computers using conventional security mechanisms. Therefore, we have been proposed a novel secure VMM (Virtual Machine Monitor) architecture which is used as a foundation of security policy enforcement on distributed computers. This paper especially introduces Role-based Access Control (RBAC) to the ID management framework in a secure VMM system. Our proposal will reduce costs for distributed policies updates. Proposed RBAC mechanism employs attribute certificates (ACs) to handle userpsilas roles. This paper shows design and prototype implementation based on PKI-based ID card and proven open source VMM software, QEMU.

Information leakage has recently become a serious problem. Because a user's disk might contain a lot of confidential information, it should be encrypted and the encryption key protected securely. Disk security has been improved by storing the encryption key in a hardware token such as a smart card or USB device. There must be some way to recover the encryption key when the token is lost, but to prevent information leakage the encryption key should not be known by the system administrator and should not be able to be recovered by malicious users inside the system. Here we describe a scheme that can limit key recovery when the user's smart card is lost and can do so without the administrator knowing the key. The smart card is used for generating the key and for improving the user authentication.