| 摘要: |
| 对NAND闪存应用于民机机载设备的可靠性问题进行了研究。首先分析了NAND闪存物理层固有的硬件失效模式和原因，以及芯片逻辑层的失效模式、机理和对上层的影响。结合民机机载设备运行环境的特殊性，阐述了电源特殊的断电方式和超宽温度范围对闪存数据的影响。在此基础上，针对某国产客舱系统投入航线运营前期遇到的闪存数据相关问题，分析了机载应力对闪存数据的影响，表明异常断电过程中的写操作及机载高低温环境应力叠加闪存的固有特性，可能导致闪存数据较高概率发生错误。最后，从硬件单级单元（Single-level Cell，SLC）/多比特单元（Multi-bit-per-cell，MBPC）闪存选择、分区文件系统选择、堆叠文件系统OverlayFS机制使用、出错检测及修复策略等方面提出了优化措施，经航线运营验证未再发生设备闪存数据相关故障。 |
| 关键词: 民机机载设备 NAND闪存 失效模式 可靠性提升 |
| DOI：10.20079/j.issn.1001-893x.251016004 |
|
| 基金项目: |
|
| Failure Modes and Reliability Improvement Methods of NAND Flash Memory in Airborne Equipment |
| KOU Hanjun,LIAO Dake |
| (CETC AVIONICS Co.，Ltd.，Chengdu 611731，China) |
| Abstract: |
| The reliability challenges of NAND flash in civil aircraft onboard equipment are investigated.Firstly，intrinsic device failure modes，and the logic layer failures，mechanisms，and impact on upper layers are analyzed.According to the particularity of the operating environment of civil aircraft onboard equipment，the impact of the special power-off operations and wide temperature ranges on flash memory is elaborated.On this base，the impact of onboard stress on flash is analyzed for the data related problems encountered in the early stage in service of a domestic aircraft cabin system，and the results indicate that the write operations during abnormal power outages and the high and low temperature environmental stress onboard，combined with the inherent characteristics of flash，will likely lead to a high probability of flash memory data errors.To mitigate the risks，a set of enhancement strategies are proposed，including single-level cell(SLC)/multi-bit-per-cell(MBPC) flash device selection，optimized partition file system configurations，OverlayFS-based protection mechanisms，and error detection and recovery techniques.The proposed approach has been validated in service on commercial aircraft and no flash-related equipment failures are observed. |
| Key words: civil aircraft airborne equipment NAND flash memory failure modes reliability improvement |
《电讯技术》编辑部 