Fault Tolerant Design And Reliability Analysis Of Spaceborne Switches

Spaceborne switches are important parts of the space-sky information network, the fault tolerance design of spaceborne switches, which aims to guarantee their high reliability in the bad space radiation environment, has theoretical value and practical value. The development costs and launch costs of spaceborne equipment are very expensive, so it is very important to assessment their reliability before production. Establishing a well reliability evaluation mechanism is also a challenging issue when designing high reliable spaceborne switches.To solve these problems, in this paper, the hierarchical design methodology has been used. The redundancy and fault tolerance design of the spaceborne switches are both at system-level and component-level.At the system-level, redundancy scheme is adopted to improve the reliability of the overall system. The system architecture of spaceborne switches, on the basis of the existing switch prototype, is designed at the beginning. Compared with the prototype, spaceborne switches use a dual redundant structure and add a system monitoring board, which is used to detect the operating status of other boards. The failure detection and switching mechanism of failure board are also designed and improved. In the improved method of monitoring network status, the reading of state information and data packets don’t use the same data bus, which can save bandwidth for data information.At the component-level, a fault-tolerant system which is called detection-diagnosis-reconfiguration TMR (DDRTMR) is designed to mitigate the SEU effects of SRAM FPGA. This system is on the basis of reconfiguration TMR system and adds a diagnostic circuit which has self-checking function. This system can not only detect errors in the redundant module, but also detect errors in the voter and comparators of the diagnosis circuit. After diagnosing the type of errors, the reconfiguration controller triggers the corresponding reconfiguration operation. Then, the reliability parameter of DDRTMR system is calculated by using Markov model. The result shows that the MTTF of DDRTMR system increase by almost four times than RTMR system.To calculating the reliability formula of the entire on-board switch system, Establishing of a parallel-series hybrid reliability model of the system. Two backups of the same card is a parallel model, and the whole system is a series model. The results show that the designed fault-tolerant method enables the system to achieve the reliability index of99.999%This paper completed the redundancy and fault tolerance design of the spaceborne switches at both system-level and component-level, and the reliability of.the system design was validated. It has certain significance for designing highly reliable switch in the engineering practice.