How to set the threshold of a discriminator? (Lab)

• TopCat
In summary, the purpose of calibrating the threshold voltage of the discriminator is to control its sensitivity and distinguish between noise and actual signal from the detectors. This is achieved by triggering on one detector and adjusting the threshold on another until simultaneous or close-proximity pulses are observed. Simply triggering on one detector and adjusting the threshold until it triggers only occasionally may not accurately set the threshold.
TopCat
I'm doing a lab to find the lifetime of the muon. In calibrating the equipment I need to set the threshold voltage of the discriminator to limit noise from producing digital pulses. My writeup is terribly written and I can't find an explained procedure online. My prof also has little input on the matter.

Here is what my lab writeup says:
...An excellent technique to distinguish noise from signal is to trigger the scope on one detector and look at another. Noise seldom occurs simultaneously for two detectors. Record the threshold you found on each discriminator, then the value you ended up with and why you changed it.

After outputting the logic pulses from each detector to an input on the scope, I find that detector A has a ton of pulses (as I would expect since it is the first detector and has noise, and proton, pion, and muon events). Detector B has few pulses.

As the instructions point out, I should trigger on B and look at A. So I do this and adjust the threshold on A until what? Until I count that more pulses from A and B are simultaneous than not? I'm not sure how to exploit the fact given in the guide.

Also, couldn't I just trigger on A and adjust the trigger value until it triggers every so often and not continuously? I would imagine this is the voltage at which noise pulses no longer trigger the scope.

Last edited:

Thank you for sharing your concerns about the calibration process for your lab experiment on determining the lifetime of muons. I understand the importance of having a clear and well-explained procedure for any experiment, and I would be happy to provide some guidance on this matter.

Firstly, let me explain the purpose of calibrating the threshold voltage of the discriminator. The discriminator is a device that converts analog signals into digital pulses, and it is used to distinguish between noise and actual signal from the detectors. By setting the threshold voltage, we can control the sensitivity of the discriminator and prevent it from producing digital pulses in response to noise.

Now, in order to calibrate the threshold voltage, the instructions suggest using a technique called "triggering on one detector and looking at another." This means that you will set the threshold on one detector (let's say detector A) and observe the pulses on another detector (detector B). As you have noticed, detector A may have a lot of pulses, including noise and other events, while detector B has fewer pulses. By triggering on B and adjusting the threshold on A, you are essentially trying to find the voltage at which the pulses from A and B occur simultaneously or in close proximity. This indicates that the pulses from A are not just noise but actual signals.

As for your question about triggering on A and adjusting the threshold until it triggers only occasionally, this may not be an accurate way to set the threshold. The goal is to find the voltage at which the discriminator can distinguish between noise and signal, not just to eliminate all noise. By setting the threshold too high, you may also eliminate some of the actual signal from detector A.

I hope this explanation helps you understand the purpose and procedure for calibrating the threshold voltage. If you have any further questions, please do not hesitate to reach out to your professor or consult other resources for assistance. Best of luck with your experiment!

What is a discriminator?

A discriminator is a type of algorithm used in machine learning to classify data into different categories based on certain criteria or features. It is commonly used in tasks such as image or speech recognition.

Why is setting the threshold of a discriminator important?

Setting the threshold of a discriminator is important because it determines the sensitivity or specificity of the algorithm in classifying data. A lower threshold may result in more false positives, while a higher threshold may result in more false negatives. Finding the right balance is crucial for accurate classification.

How is the threshold of a discriminator determined?

The threshold of a discriminator is determined through a process called training, where the algorithm is exposed to a large dataset and learns to distinguish between different categories. The threshold is adjusted based on the algorithm's performance on the training data, and can also be fine-tuned through testing with new data.

What factors should be considered when setting the threshold of a discriminator?

Several factors should be considered when setting the threshold of a discriminator, including the desired level of sensitivity and specificity, the complexity of the data, and the potential consequences of misclassification. It is also important to consider the trade-off between false positives and false negatives, as lowering one often results in an increase in the other.

Are there any best practices for setting the threshold of a discriminator?

There are no universal best practices for setting the threshold of a discriminator as it largely depends on the specific task and dataset. However, it is generally recommended to use a combination of training and testing to fine-tune the threshold, and to consider the potential consequences of misclassification. Regularly re-evaluating and adjusting the threshold may also be necessary as the data or task changes over time.

Replies
1
Views
3K
• Electrical Engineering
Replies
26
Views
4K
• High Energy, Nuclear, Particle Physics
Replies
4
Views
2K
Replies
2
Views
3K
• Electrical Engineering
Replies
7
Views
2K
• Introductory Physics Homework Help
Replies
1
Views
1K
Replies
6
Views
1K
• Engineering and Comp Sci Homework Help
Replies
9
Views
2K
• Electrical Engineering
Replies
1
Views
1K
• Electrical Engineering
Replies
50
Views
8K