# How does the Weibull distribution work as fault model in wireless sensor networks?

I am a little bit confused with the concept of using the Weibull distribution or other distribution for fault model. As I understand, in simulation this distribution is often used for modelling a fault in components.

On the other hand when designing systems we often assume some components might become faulty and we use fault detection method to know where the distinct component is faulty or not.

I want to categorize faulty sensor in 5 groups: transmitter circuit/battery condition/micro controller/receiver circuit and sensor circuit fault and correlated these kinds we can have different attribute in network.

The question is, how can i write the Weibull distribution function to produce these kinds of fault in a wireless sensor networks?

• Welcome to CS.SE! I can't fully understand what you are asking. What do you mean by the part about "using fault detection method", "I want to categorize faulty sensor", and so on? Can you elaborate or rephrase those sentences? Presumably if you are writing a simulator that introduces simulated faults, you don't need to diagnose which category of the fault occurred, by yourself. Instead, your simulator can randomly pick what kind of fault happened. Is that right? Would you be happy with that approach? If so, what problem are you trying to solve? Can you edit the question to clarify it? – D.W. Jan 13 '16 at 21:08
• thanks for your answer.i am using Matlab for simulation,my question is,how can i write weibull distribution to make randomly failure in transmitter circuit or other fault like circuit fault in sensors?. – Anahita k Jan 14 '16 at 15:49

## 1 Answer

Step 1: Fit a Weibull distribution to all transmitter circuit faults (with one set of parameters). Fit a Weibull to battery condition faults (with another set of parameters). You get five different Weibull distributions.

Step 2: During simulation, randomly draw one number from each of the Weibull distributions. This gives you the time until the first transmitter circuit fault, the time until the first battery condition fault, and so on. Pick the smallest number; that's the fault that occurs first. Simulate a fault of that type, at that time. (You can continue the simulation to simulate multiple faults if you want.)

• i try to do these steps,thank you for helping so. – Anahita k Jan 14 '16 at 17:59