Toshio Okochi

Enabling Technology for Smart City Transportation in Developing Countries

Traffic in the major cities of developing countries such as the Hanoi and Ho Chi Minh cities is chaotic because multiple transport modes such as cars, buses, motorbikes, and bicycles share the same road lanes together with narrow road infrastructure. In addition, traffic is poorly managed due to lack of efficient traffic-management system. These problems have accumulated with the increase in mobility demand because of rising urban population. The objectives of this study are to propose a solution to mitigate traffic-congestion problem, to provide better driving and commuting experience to commuters, and to assist in the implementation of efficient transportation-management system in the Hanoi city. This paper presents data-monitoring and analysis tools to enable smart mobility capability, adaptive traffic signaling that is the scheduling of traffic light timing based on road capacity and congestion indicator. We propose three congestion indicators: stop count, velocity, and travel-time variation, which represent the state of congestion at individual road segment.

Logic-Level Analysis of Fault Attacks and a Cost-Effective Countermeasure Design

This article analyzes the internal mechanism of fault attacks on microcontrollers and proposes a cost-effective hardware and software countermeasure design policy. Reliable branch operations are essential to DFA-resistant hardware. Our method is based on a logical fault attack simulation to find the minimum set of signals that contribute to faults in the branch operations and is also based on applying partially redundant logic.

PDEQSOL (Partial Differential Equation Solver Language) for Parallel Computers

Several distributed-memory-type massively parallel computers have been developed and have proved to be efficient for numerical simulation of Partial Differential Equation (PDE) problems [1, 2, 3]. Although performance is high, it is very time-consuming to develop programs by using languages that have a special facility to control parallelism [4]. (In this chapter, “massively parallel computer ” means the distributed memory type.) On the other hand, in the automatic parallelization of general languages like FORTRAN and C, it is technically very difficult to achieve sufficient parallelizing efficiency for programs that are developed without special care for parallelization. The PDEQSOL (Partial Differential Equation Solver Language) system for massively parallel computers is being developed to overcome these limitations. PDEQSOL is a high-level programming language specially designed to describe PDE problems in a natural way for numerical analysis. The parallelizing translator generates a highly parallelized program for massively parallel computers by using the implicit parallelism in the PDEQSOL program.