ISO 26262-2018 Guideline: Safety Mechanism series 02
[Author]
Renhong WENG, Safety, and Security, and RAMS investigator.
D.2.1.2: comparator
Aim: refer to ISO 26262-2018-5
Description: refer to ISO 26262-2018-5
Example of use:
(1) analog comparator
as shown above, analog comparator compares two input voltages viz. Vin0 (at positive terminal) and Vin1(at negative terminal) and produces Vout based on these.
1.when Vin0>Vin1, Vout is equal to +Vsupply
2.when Vin0
(2)Digital comparator
Normally we may have 1 bit comparator, or 2 bit comparators, or 4 bit comparators.
For 1 bit, there are 3 binary variables which indicates the relationship between A and B;are four inputs and three outputs
For 2 bit, there are several more complex ways, and we have to define results based on true table or Kmaps.
(3) Cypress CMPHW V1
D.2.1.3 Majority voter
Aim and description please refer to ISO 26262-2018-5.
Example of use:
| failure type | Permanent type | ||
| intermittent type | |||
| transient type | also soft error | radiation hardening by design or triple modular redundancy |
V=XY+YZ+XZ
From probabilistic, Rtmr whole voting efficiency as following:
Rsimple=R_M, Rsimple informs functional module non-faulty state
if the R_m>0.5, then the Rtmr will be better than R_m, higher reliability
1.Classic majority voter
V=N1+N2+N3=XY+YZ+XZ
True table: potential problems
| Primary voter inputs | Internal voter ouputs | Primary voter output | Voter ouput state | ||||
| X | Y | Z | N1 | N2 | N3 | V | |
| No function module fault/failure | |||||||
| 0 | 0 | 0 | 0 | 0 | 0 | 0 | Actual |
| 0 | 0 | 0->1 | 1 | Error | |||
| 0 | 0->1 | 0 | 1 | Error | |||
| 0->1 | 0 | 0 | 1 | Error | |||
| 0 | 0->1 | 0->1 | 1 | Error | |||
| 0->1 | 0 | 0->1 | 1 | Error | |||
| 0->1 | 0->1 | 0 | 1 | Error | |||
| 0->1 | 0->1 | 0->1 | 1 | Error | |||
| Single/multiple function module faults/failures | |||||||
| 0 | 0 | 1 | 0 | 0 | 0 | 0 | Actual |
| 0 | 0 | 0->1 | 1 | Error | |||
| 0 | 0->1 | 0 | 1 | Error | |||
| 0->1 | 0 | 0 | 1 | Error | |||
| 0 | 0->1 | 0->1 | 1 | Error | |||
| 0->1 | 0 | 0->1 | 1 | Error | |||
| 0->1 | 0->1 | 0 | 1 | Error | |||
| 0->1 | 0->1 | 0->1 | 1 | Error | |||
| 0 | 1 | 0 | 0 | 0 | 0 | 0 | Actual |
| 0 | 0 | 0->1 | 1 | Error | |||
| 0 | 0->1 | 0 | 1 | Error | |||
| 0->1 | 0 | 0 | 1 | Error | |||
| 0 | 0->1 | 0->1 | 1 | Error | |||
| 0->1 | 0 | 0->1 | 1 | Error | |||
| 0->1 | 0->1 | 0 | 1 | Error | |||
| 0->1 | 0->1 | 0->1 | 1 | Error | |||
| 0 | 1 | 1 | 0 | 1 | 0 | 1 | Actual |
| 0 | 1 | 0->1 | 1 | Correct | |||
| 0 | 1->0 | 0 | 1 | Correct | |||
| 0->1 | 1 | 0 | 1 | Correct | |||
| 0 | 1->0 | 0->1 | 1 | Correct | |||
| 0->1 | 1 | 0->1 | 1 | Correct | |||
| 0->1 | 1->0 | 0 | 1 | Correct | |||
| 0->1 | 1->0 | 0->1 | 1 | Correct | |||
| 1 | 0 | 0 | 0 | 0 | 0 | 0 | Actual |
| 0 | 0 | 0->1 | 1 | Error | |||
| 0 | 0->1 | 0 | 1 | Error | |||
| 0->1 | 0 | 0 | 1 | Error | |||
| 0 | 0->1 | 0->1 | 1 | Error | |||
| 0->1 | 0 | 0->1 | 1 | Error | |||
| 0->1 | 0->1 | 0 | 1 | Error | |||
| 0->1 | 0->1 | 0->1 | 1 | Error | |||
| 1 | 0 | 1 | 0 | 0 | 1 | 1 | Actual |
| 0 | 0 | 1->0 | 0 | Error | |||
| 0 | 0->1 | 1 | 1 | Correct | |||
| 0->1 | 0 | 1 | 1 | Correct | |||
| 0 | 0->1 | 1->0 | 1 | Correct | |||
| 0->1 | 0->1 | 1 | 1 | Correct | |||
| 0->1 | 0 | 1->0 | 1 | Correct | |||
| 0->1 | 0->1 | 1->0 | 1 | Correct | |||
| 1 | 1 | 0 | 1 | 0 | 0 | 1 | Actual |
| 1 | 0 | 0->1 | 1 | Correct | |||
| 1 | 0->1 | 0 | 1 | Correct | |||
| 1->0 | 0 | 0 | 0 | Error | |||
| 1 | 0->1 | 0->1 | 1 | Correct | |||
| 1->0 | 0 | 0->1 | 1 | Correct | |||
| 1->0 | 0->1 | 0 | 1 | Correct | |||
| 1->0 | 0->1 | 0->1 | 1 | Correct | |||
| 1 | 1 | 1 | 1 | 1 | 1 | 1 | Actual |
| 1 | 1 | 1->0 | 1 | Correct | |||
| 1 | 1->0 | 1 | 1 | Correct | |||
| 1->0 | 1 | 1 | 1 | Correct | |||
| 1->0 | 1->0 | 1 | 1 | Correct | |||
| 1->0 | 1 | 1->0 | 1 | Correct | |||
| 1 | 1->0 | 1->0 | 1 | Correct | |||
| 1->0 | 1->0 | 1->0 | 0 | Error | |||
we use the fault masking ratio (FMR) specified as the ratio of total number of correct voter output states in the presence of internal and/or external faults, divided by the total number of potential internal and/or external fault occurrences
For above table, the FMR for classical TMRis estiamted
FMR_Classical=(33-8)/56=44.64%
Above the fault masking ratio lower than 50%, we have improve the design of majority voter:
NOTE: we cannot directly use the majority voter DC value from ISO 26262, actually here the classicial one only 44.64% percentage
2.Proposed majority voter - Design and Fault tolerance analysis
V=MZ+XY+YZ
M=X+Y
True table: potential problems
| Primary voter inputs | Internal voter ouputs | Primary voter output | Voter ouput state | ||
| X | Y | Z | M | V | |
| 0 | 0 | 0 | 0 | 0 | Actual |
| 0->1 | 0 | Correct | |||
| 0 | 0 | 1 | 0 | 0 | Actual |
| 0->1 | 1 | Error | |||
| 0 | 1 | 0 | 1 | 0 | Actual |
| 1->0 | 0 | Correct | |||
| 1 | 0 | 0 | 1 | 0 | Actual |
| 1->0 | 0 | Correct | |||
| 0 | 1 | 1 | 1 | 1 | Actual |
| 1->0 | 1 | Correct | |||
| 1 | 1 | 0 | 1 | 1 | Actual |
| 1->0 | 1 | Correct | |||
| 1 | 0 | 1 | 1 | 1 | Actual |
| 1->0 | 0 | Error | |||
| 1 | 1 | 1 | 1 | 1 | Actual |
| 1->0 | 1 | Correct | |||
FMR_proposed=6/8=75%
Proposed version of majority voter had increase the diagnostic coverage
| Type of voter | FMR% |
| Classical_voter | 44.64 |
| Proposed_voter | 75 |
Here, we had improved our majority voter quality and DC value will be 75% higher than classicial version.
[REF]
A Fault Tolerance Improved Majority Voter for TMR System
Architectures
- 用户评论