Field programmable gate arrays (FPGAs) are increasingly used in industry (e.g., biomedical, space, and automotive industries). FPGAs are subjected to single, as well as multiple event upsets (SEUs and MEUs), due to the continuous shrinking of transistor dimensions. These upsets inevitably decrease system lifetime. Fault-tolerant techniques are often used to mitigate these problems. In this research, penta and hexa modular redundancy, as well as dynamic partial reconfiguration (DPR), are used to increase system reliability. We show, depending on the relative rates of the SEUs and MEUs, that penta modular redundancy has a higher reliability than hexa modular redundancy, which is a counter-intuitive result in some cases since increasing redundancy is expected to increase reliability. Focusing on penta modular redundancy, an error detection and recovery mechanism (voter) is designed. This mechanism uses the internal configuration access port (ICAP) and its associated controller, as well as DPR to mitigate SEUs and MEUs. Then, it is implemented on Xilinx Vivado tools targeting the Kintex7 7k410tfbg676 device. Finally, we show how to render this design fault secure in the event that SEUs or MEUs affect the voter itself. This fault secure voter either produces the correct output or gives an indication that the output is incorrect.
FPGAs are currently being used in many applications due to their flexibility and re-programmability. Some of these applications operate in harsh environments. One such environment is space. Focusing on space applications, these FPGAs are subjected to soft and hard faults. For newer technology nodes, in addition to the harsh space environment, these errors are more severe. This paper investigates several fault-tolerant architectures to mitigate Single Event Upsets (SEUs), Double Event Upsets (DEUs), Triple Event Upsets (TEUs) as well as hard faults. Conventionally, for TEUs, seven copies of a module are required (7MR). Therefore, a modified 7MR architecture is studied along with two other architectures with six redundant modules: a modified 6MR architecture and a modified Triple Duplex architecture. Using Continuous Time Markov Chains (CTMCs), it is proven that, in many of the cases studied in this article, the modified Triple Duplex architecture has a higher reliability than the modified 7MR architecture. This is a counter-intuitive result. It is also proven that the modified 6MR architecture always has a lower reliability than the modified Triple Duplex architecture even though they both require six redundant modules. The ratio between the relative rates of SEUs, DEUs and TEUs plays an important role in determining the most reliable architecture. Furthermore, the Xilinx Vivado tool with the Kintex7, 7k410tfbg676 device is used to implement the modified 7MR and modified Triple Duplex voters to estimate the area and power consumed by these techniques.