| ZHU ZEYU | M, 2回目発表 | ネットワークシステム学 | 岡田 実, 笠原 正治, 東野 武史, Duong Quang Thang |
| 中畑 裕 | M, 2回目発表 | 大規模システム管理 | 笠原 正治, 伊藤 実, 笹部 昌弘, 堀山 貴史（埼玉大学）, 川原 純 |
|
title: Enumerating and Indexing Balanced Graph Partitions Using Desicion Diagrams
abstract: Partitioning a graph into balanced components is important for several applications. For multi-objective problems, it is useful not only to find one solution but also to enumerate all the solutions with good values of objectives. However, there are a vast number of graph partitions in a graph, and thus it is difficult to enumerate desired graph partitions efficiently. To enumerate balanced graph partitions efficiently, we propose an algorithm to enumerate all the graph partitions such that all the weights of the connected components are at least a specified value. To deal with a large search space, we use zero-suppressed binary decision diagrams (ZDDs) to represent sets of graph partitions and we design a new algorithm based on frontier-based search, which is a framework to construct a ZDD efficiently. Our algorithm utilizes not only ZDDs but also ternary decision diagrams (TDDs) and realizes an operation which seems difficult to be designed only by ZDDs. Experimental results show that the proposed algorithm runs up to tens of times faster than an existing state-of-the-art algorithm. language of the presentation: Japanese | |||
| WILLIAMS FATOU | M, 1回目発表 | サイバーレジリエンス構成学 | 門林 雄基 |
|
Title: Analyzing the Capacity of LoRaWAN as an IoT Enabler
Abstract: Long Range Wide Area Network (LoRaWAN) as an IoT solutions enabler is one of the most popular software protocols instrumental in connecting low-power Machine to Machine (M2M) and IoT devices over a long range to the Internet. LoRaWAN communication protocol contains several features that contributes to uphold the low power wide area (LPWA) design criteria of low power, cost, implementation complexity and bandwidth. However, a major challenge of LoRaWAN is, its capacity is limited paticularly by the duty cycle (Percentage of time on air due to regulation) and transmission in the downlink through acknowledgement of frames. Current methods towards improving the capacity do not provide the network with enough flexibility to adapt to network conditions and are not able to serve applications that require high capacity. This limits the possibility to use LoRaWAN in many IoT use case scenarios. Our research is geared towards evaluating, analyzing and comparing certain feature settings of specific IoT use cases in LoRaWAN and the impact on the capacity. We will also propose a method to improve the capacity of IoT applications using LoRaWAN. Language of the presentation: English | |||
| ARAYA KIBROM DESTA | M, 1回目発表 | 情報基盤システム学 | 藤川 和利 |
|
title: Intrusion Detection in Controller area Network
abstract: Vehicles are not mechanical devices anymore, it has a lot of Electronic Control Units which communicate with each other using a standard called Controller Area Network (CAN). CAN is Serial communication in which all nodes on the CAN bus are attached to a common connection using the same bitrate. Using CAN inside cars has massively reduced the wiring between Electronic Control Units. Even if CAN provides a lot more to automobiles, it also opened a new security hole in the automotive industry. CAN bus is a broadcast communication by nature and this spares a hacker the ability to easily snoop on all communications or send a malicious packet to the nodes in the network. Arbitration ID is used for prioritizing nodes to use the bus medium, the lower the arbitration ID the higher probability a node will have to use the bus. Due to this reason, CAN is fragile to Denial of Service attacks. It also has no any authentication mechanisms to differentiate an anomaly from a genuine CAN message. There are two techniques to overcome this drawback of CAN bus. The first one is, using cryptographic mechanisms to encrypt CAN messages and decrypting the messages when it arrives in the intended ECU, but this solution fails as CAN bus and the ECU’s connected to the bus have a low processing power and automobiles need a real-time communication. The other solution is identifying intrusions inside CAN bus. Even though, this is a bit difficult to identify anomalies in a completely randomized CAN messages, this is a better solution to the problem. In this research we provide a mechanism to identify anomaly packets inside the CAN bus. The mechanism we provide will have the ability to detect anomalies in a real time CAN bus communication and will instantly inform the driver about the issues, so that the driver can take an appropriate action depending on the message from our system. We will use different machine learning algorithms that can help us to easily identify anomalies in a series of CAN messages. language of the presentation: English | |||
| 一倉 孝宏 | D, 中間発表 | コンピューティング・アーキテクチャ | 中島 康彦, 井上 美智子, 中田 尚, TRAN THI HONG, 張 任遠 |
|
title: EMAXVR: A Programmable Accelerator Employing Near ALU Utilization to DSA
abstract: The domain specific accelerators (DSAs) have appeared remarkable performances in many real-world applications such as deep learning. However, the benefit on performances is somehow eaten up by the increasing development cost and poor programmability. So we propose a programmable accelerator by employing near ALU utilization to DSA, which is an improved version of our previously reported accelerator called EMAXV. As a result, we found our programmable accelerator can compute convolution operations in AlexNet with only 17% lower utilization of ALU and 17% lower rate of data reuse compared with the latest flexible DSA. language of the presentation: Japanese | |||
| ISLAM SYFUL | M, 1回目発表 | ディペンダブルシステム学 | 井上 美智子 |
|
Title:Variation-Aware Power Side Channel Analysis for Hardware Trojan Detection
Abstract:Due to the globalization of semiconductor industry i.e. desing and fabrication process, ICs are becomming increasingly vulnerable to malicious activities. Adversary can insert trojan circuit in ICs to create security hole in safety critical systems. So hardware trojan is now a growing concern of semiconductor industry, govement agecies and other companies which use ICs extensively in automation process. High detection sensitivity in the presence of process variation is a key challenge for hardware trojan detection through side channel analysis. Many researchers are currently working with process variation in ICs; but random process variation is still remaining to explore for increasing detection sensitivity even in case of very tinny HT circuits in ICs since randomness can mask the presence of maliciousness of ICs so decreasing the confidence whether the target ICs are trojan free or not. So currently my research concern is to reduce random process variation effect to increase detection sensitivity even in presence of very tinny trojan circuits in ICs. Language of the presentation: English | |||
| ZHAO DAFANG | M, 1回目発表 | ディペンダブルシステム学 | 井上 美智子 |
|
title: Photovoltaic System Reconfiguration
abstract: Performance of Photovoltaic (PV) array is affected by variations in temperature, shadow and array configuration. When the panels in PV array are subjected to non-uniform irrdiance level, some bypass diode may turn on, and that will cause overall power production drop.To provent this kind of power drop. This research proposed an electrical reconfiguration algorithm. Proposed algorithm computes an optimized configuration with low computational burden. language of the presentation: English | |||