A Novel Method for Soft Error Mitigation in FPGA using Adaptive Cross Parity Code

TL;DR

This paper introduces an adaptive cross parity check (ACPC) based error correction method for FPGAs, enhancing soft error mitigation with a simple, efficient, and reconfigurable approach validated on Kintex FPGA.

Contribution

The paper presents the first adaptive cross parity code scheme specifically designed for FPGA soft error correction, combining simplicity and effectiveness.

Findings

ACPC effectively corrects multi-bit errors in FPGA configuration memory.

The proposed method shows low power consumption and minimal overhead.

Error correction efficiency is validated through FPGA testing.

Abstract

Field Programmable Gate Arrays (FPGAs) are more prone to be affected by transient faults in presence of radiation and other environmental hazards compared to Application Specific Integrated Circuits (ASICs). Hence, error mitigation and recovery techniques are absolutely necessary to protect the FPGA hardware from soft errors arising due to such transient faults. In this paper, a new efficient multi-bit error correcting method for FPGAs is proposed using adaptive cross parity check (ACPC) code. ACPC is easy to implement and the needed decoding circuit is also simple. In the proposed scheme total configuration memory is partitioned into two parts. One part will contain ACPC hardware, which is static and assumed to be unaffected by any kind of errors. Other portion will store the binary file for logic, which is to be protected from transient error and is assumed to be dynamically reconfigurable (Partial reconfigurable area). Binary file from the secondary memory passes through ACPC hardware and the bits for forward error correction (FEC) field are calculated before entering into the reconfigurable portion. In the runtime scenario, the data from the dynamically reconfigurable portion of the configuration memory will be read back and passed through the ACPC hardware. The ACPC hardware will correct the errors before the data enters into the dynamic configuration memory. We propose a first of its kind methodology for novel transient fault correction using ACPC code for FPGAs. To validate the design we have tested the proposed methodology with Kintex FPGA. We have also measured different parameters like critical path, power consumption, overhead resource and error correction efficiency to estimate the performance of our proposed method.