A simple model for multiple coincidence SiPM dark noise

Valerio Bocci

2024-03-12

Abstract:Silicon photomultipliers are photon sensors based on an array of pixels, each consisting of a single photon avalanche photodiode (SPAD). Reading of the array is not done on a single-pixel basis, but all pixels are read simultaneously. Because of thermal agitation and other phenomena, each pixel can fire even in the dark with a small but not zero probability of adding to the output signal a fixed current. Each pixel has an average random event rate, and the dark noise rate is due to the sum of any pixel contribution. Statistically, two or more pixels may fire simultaneously, giving a current signal equal to the sum of the pixels involved. This eventuality has a lower probability of a more significant number of fired pixels. This paper aims to find simple formulas that relate dark noise and the rate of having multiple multiplicity events.

Instrumentation and Detectors

What problem does this paper attempt to address?

This paper aims to solve the dark noise problem in silicon photomultipliers (SiPMs). Specifically, SiPM is a photon sensor based on a pixel array, and each pixel contains a single - photon avalanche diode (SPAD). In the absence of light, due to thermal perturbations and other phenomena, each pixel may still be triggered with a certain probability, generating a fixed current signal. When multiple pixels are triggered simultaneously, a signal equal to the sum of the currents of these pixels will be generated. Although the probability of such multi - pixel simultaneous triggering events is low, it has a significant impact on the performance of SiPM.

The main objective of the paper is to find simple relationships for describing the relationship between dark noise and the multi - pixel simultaneous triggering event rate. The author calculates the probabilities of these events by introducing a simplified model and derives relevant mathematical formulas. These formulas can help researchers better understand the dark noise characteristics of SiPM, thereby optimizing its design and application.

### Key points of the paper

1. **Background introduction**:
- SiPM is a solid - state photon sensor composed of multiple SPAD pixels.
- Each SPAD operates in Geiger mode and will discharge when receiving a photon.
- Even under dark conditions, each pixel may be triggered due to thermal perturbations, generating dark noise.

2. **Main problems**:
- The probability of multiple pixels being triggered simultaneously is low, but such events have a significant impact on the output signal of SiPM.
- It is necessary to find a simple method to describe the relationship between dark noise and the multi - pixel simultaneous triggering event rate.

3. **Methods**:
- A simplified model was introduced to calculate the probability of multi - pixel simultaneous triggering.
- Relevant mathematical formulas were derived, including:
- The formula for the double accidental coincidence rate:
\[
f_{AC}(n) = n \sum_{i = 1}^{n}\frac{f_i\tau_i}{\sum_{i = 1}^{n}\frac{1}{\tau_i}}
\]
- When all pulse widths and rates are equal, the formula can be simplified to:
\[
f_k = k\tau^{k - 1}f^k
\]
- For an array of N pixels, the frequency of k pixels being triggered simultaneously is:
\[
f_{coinc} = k\tau^{k - 1}f^k_{pixel}\frac{N^k}{k!}
\]

4. **Results**:
- Through the above formulas, the total dark noise rate \( f_{dark} \) of SiPM can be estimated, and the dark noise rate \( f_{pixel} \) of a single pixel can be further derived.
- Experimental results show that dark noise is mainly composed of triggering events of a single pixel, and the events of multiple pixels being triggered simultaneously contribute less to the total dark noise.

5. **Conclusions**:
- The proposed formula \( f_{coinc}=\tau^{k - 1}f_{dark}^k/(k - 1)! \) provides a reliable method to estimate the event rate of k pixels being triggered simultaneously in SiPM.
- This method is applicable not only to SiPM but also to similar detector systems.

Through these studies, the paper provides an important theoretical basis for understanding and optimizing the dark noise characteristics of SiPM.