HIHiroshi Inamura

Hiroshi Inamura

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Hiroshi Inamura has a total of 7 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Covert Communication in Relay-Assisted IoT Systems

    by Chan Gao, Bin Yang, Xiaohong Jiang, Hiroshi Inamura, Masaru Fukushi

    Internet of Things (IoT) systems are of paramount importance to provide ubiquitous wireless connectivity for smart cities. However, such systems are facing security challenges due to the broadcast and openness nature of wireless channels. This article studies the performance of covert communication under a scenario consisted of a source-destination pair, a passive warden, and multiple relays. We first propose two relay selection schemes, one is random selection and another is superior-link selection. Based on these two schemes, we then examine the transmission strategy design for the source and thus define the necessary condition that the source can transmit covert messages. We further derive the detection error probability of warden and covert capacity based on two relay selection schemes and also explore the covert capacity maximization through efficient numerical searches. Finally, numerical results are provided to illustrate our theoretical findings and the performance of covert communication in such systems. Remarkably, the superior-link selection scheme has 108% improved to the random selection scheme for the maximum covert capacity performance under the same transmission power at the source.


  • Classification of Unknown Web Sites Based on Yearly Changes of Distribution Information of Malicious IP Addresses

    by Yoshitaka Nakamura, Shihori Kanazawa, Hiroshi Inamura, Osamu Takahashi

    Recently, cyber attacks through Web sites such as Drive-by download attacks or phishing attacks are increasing rapidly. The attackers can illegally acquire personal information of users by these attacks and cause economical damages. We aim to detect malicious Web sites which cause economic damages. The analysis of the features of the network address part of the IP address revealed that the features of malicious IP address has time change. Therefore, reflecting the time changes of these features, we classified unknown malicious Web sites. As a result of the evaluation experiment, classification accuracy could be improved.