Future University Hakodate

About: Future University Hakodate is a education organization based out in Hakodate, Japan. It is known for research contribution in the topics: Network packet & Mobile ad hoc network. The organization has 485 authors who have published 1338 publications receiving 13204 citations. The organization is also known as: Mirai Daigaku (Future University) Hakodate Japan.

A Performance Analysis and Evaluation of SIDH with Implementation-Friendly Classes of Quadratic Extension Fields

Abstract: It is well-known that the class of binomial extension fields is widely used to construct quadratic extension fields (QEFs) of supersingular isogeny Diffie-Hellman (SIDH) key exchange protocol. There is a possibility to improve the performance of SIDH by employing other classes of QEFs, i.e., extension fields with normal basis and all-one polynomial extension fields, without sacrificing the range of primes. In this paper, the authors confirm that the applicability of the other classes for SIDH and evaluate the computational complexity of the large-degree isogenies required for SIDH. The results of the experiments show that the performances with the classes are comparable to the QEF with a binomial x^2+1.

Smartphone Usage Analysis Based on Actual-Use Survey

Abstract: This paper presents a smartphone usage model that will be useful as ground data for works such as proposals on improving services/technologies related to smartphones. It is based on an actual-use survey involving about 700 smartphone users. We conduct web interviews with users to get their demographic data - age, sex, for example, and gather many kinds of usage traces from their device using our logger application. The model describes 1) daily usage pattern (combinations of application usages) and 2) pattern features - demographics, major application usages and so on. Through a cluster analysis of the usage traces, we find there are six typical daily usage patterns and half the users have multiple patterns that differ with the day.

Data Analysis of a Low-Cost Bathymetry System Using Fishing Echo Sounders

Abstract: The multi-beam echo sounder system with an accurate GPS can provide a detailed view of the seafloor and is successfully utilized in oceanography but its expense makes it impractical for many uses. The authors have therefore presented a low-cost system for surveying shallow sea bottom topography by using a fishing echo sounder in such a way that the many areas of the industry in which the information of the sea depth is required can easily use it to get the information they need. The system is based on sensor network technology, namely Micro Cube, which allows for the collection of the data from the fishing echo sounder via the Internet. However, the data obtained by the present system include noises or errors due to the reflection from bubbles or fishnets. In this paper, data analyses to erase noises or errors from measurements in the present system are discussed using the discrete wavelet transform and statistics. Experiments carried out in Rumoi, on the west coast of Hokkaido, Japan from July 2004 to December 2006 are studied for the examinations of data analyses for practical use.

Ripple: An Efficient Runtime Reconfigurable P4 Data Plane for Multicore Systems

Abstract: The P4 language was originally proposed as a means of providing programmable network data plane functions with the potential to implement, understand, and interpret custom protocols. However, the recompilation process and the re-instantiation of network functions on a target device introduces measurable downtime for network services. In this work, we present Ripple, an efficient, runtime reconfigurable P4 engine on multicore systems to eliminate the recompilation process. The Ripple framework adopts the "generic and static" philosophy, and uses a highly optimized universal binary executable to avoid the lengthy recompilation and reconfiguration process, while supporting packet processing at line rates. We leverage advanced I/O and vectorization techniques to take advantage of packet level parallelism and processor architectural features, thus significantly improve the performance of P4 data plane at runtime. Our in-depth evaluation on multicore architecture shows that Ripple reaches comparable throughput with one of the state-of-art frameworks that requires recompilation and reconfiguration.

Scallop detection from gravel-seabed images for fishery investigation

Abstract: We propose a method of extracting scallop areas from gravel seabed images to assess fish resources, and developed an automatic system that measures their quantities, sizes, and states. Scallops feature different colors, fluted patterns and forms. The present study is described our method of extracting these features, and presented the results and its effectiveness.