首页> 外文OA文献 >Escalonador em hardware para deteção de falhas em sistemas embarcados de tempo real
【2h】

Escalonador em hardware para deteção de falhas em sistemas embarcados de tempo real

机译:用于实时嵌入式系统中故障检测的硬件调度程序

代理获取
本网站仅为用户提供外文OA文献查询和代理获取服务,本网站没有原文。下单后我们将采用程序或人工为您竭诚获取高质量的原文,但由于OA文献来源多样且变更频繁,仍可能出现获取不到、文献不完整或与标题不符等情况,如果获取不到我们将提供退款服务。请知悉。

摘要

Nowadays, several safety-critical embedded systems support real-time applications and their development represents a great challenge to engineers and researchers due to the risk of catastrophic effects on the system generated by a fault. Usually, real-time embedded systems process input data and generate output responses according to the functional specification of the system. However, the high complexity of the applications has made the adoption of Real-Time Operating Systems (RTOS) necessary in order to simplify the design of real-time embedded systems. Thus, the RTOS serves as an interface between software and hardware. However, real-time systems can be affected by transient faults during application running or even during the RTOS execution. Consequently, these faults can affect both, the correctness of the output responses generated and the task’s deadline specified during the project of the system. In this context, this work proposes a new hardware-based approach able to increase the reliability of the real-time embedded systems. The proposed technique is based on the development of an Infrastructure IP core (I-IP) called Hardware-Scheduler (Hw-S), which monitors the tasks’ execution in order to verify if tasks’ execution flow and the tasks’ deadline are respected. A case study implemented in an FPGA running a set of benchmarks has been developed in order to validate the proposed approach. The benchmarks developed exploit most of the RTOS services. In order to evaluate the effectiveness of the proposed technique, Hardware and Software fault injection campaigns have been performed. Indeed, the introduced overheads have been estimated. The obtained results demonstrate that the fault latency associated to the Hw-S is smaller than the one associated to the RTOS and further that the Hw-S’s fault coverage is higher than the RTOS’. Finally, the Hw-S introduces an area overhead of about 6% with respect to the Plasma microprocessor area.
机译:如今,一些安全关键型嵌入式系统支持实时应用,由于故障产生的灾难性后果,其开发对工程师和研究人员构成了巨大的挑战。通常,实时嵌入式系统根据系统的功能规范处理输入数据并生成输出响应。但是,应用程序的高度复杂性使得必须采用实时操作系统(RTOS),以简化实时嵌入式系统的设计。因此,RTOS充当软件和硬件之间的接口。但是,实时系统可能会在应用程序运行甚至RTOS执行期间受到瞬态故障的影响。因此,这些故障会影响生成的输出响应的正确性以及系统项目期间指定的任务期限。在这种情况下,这项工作提出了一种新的基于硬件的方法,该方法能够提高实时嵌入式系统的可靠性。所提出的技术基于称为硬件调度程序(Hw-S)的基础结构IP内核(I-IP)的开发,该内核监视任务的执行情况,以验证是否遵守任务的执行流程和任务的截止日期。为了验证所提出的方法,已经开发了在运行一组基准的FPGA中实现的案例研究。开发的基准利用了大多数RTOS服务。为了评估所提出技术的有效性,已经进行了硬件和软件故障注入活动。实际上,已经估计了引入的间接费用。获得的结果表明,与Hw-S关联的故障延迟小于与RTOS关联的延迟,并且Hw-S的故障覆盖率高于RTOS。最终,Hw-S引入了相对于等离子微处理器面积约6%的面积开销。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号