Reconfigurable fault-safe processor platform based on RISC-V for large-scaled IoT-driven applications

As the Internet-of-Things (IoT) becomes more generalized in our daily lives, the number of devices operating at the edges has increased significantly, and these devices generate countless data from the physical world. These devices are irregularly interconnected, with heterogeneous characteristics and construct a large-scale IoT system. Thus, the operation of each device becomes more important, so a malfunction can be propagated along the connection and affect the whole system. In addition, IoT devices have a constraint, in that they do not receive enough power in an unstable environment operating at the edge of the environment and they operate in a variety of environments. For these reasons, it is necessary to monitor variously the IoT system, which has irregular connectivity and changes critical sections of logic every time because of various operations. In this paper, to solve this problem, we propose an optimized low-power reconfigurable Fault-Safe Processor (FSP) platform based on RISC-V for large-scale IoT-driven applications that can construct an optimized IoT system after considering numerous connectivity and operation sequences of processors. Therefore, we can reconfigure the hardware in real-time using the proposed platform as well as heterogeneous co-simulation and dynamic partial reconfiguration to stabilize the IoT system and optimize both accumulative power over time and instantaneous power.