test_gaussian

# CS123x Gaussian Noise Test Example

This example provides a high-resolution statistical analysis and noise characterization tool for the CS123x library. It uses Welford's algorithm for numerically stable mean and variance computation, reporting RMS noise, peak-to-peak noise, LSB-equivalent voltage noise, effective number of bits (ENOB), and noise-free resolution.

---

## 🚀 Features Demonstrated

1. **ADC Initialization & Hardware Sync:** Initializes GPIO lines, writes the default or user-provided configuration to the ADC and verifies correct register read-back from it.
2. **Interactive Terminal CLI:** Real-time command-line interface over serial to trigger tests, cycle gains, sampling rates, input channels and set sample counts.
3. **Statistical Analysis via Welford's Algorithm:** Calculates mean (offset), variance, standard deviation (RMS noise in LSB and uV) and peak-to-peak noise without floating-point overflow.
4. **ENOB & Noise-Free Resolution Calculation:** Computes Effective Number of Bits (`ENOB = 24 - log2(RMS_noise)`) and Noise-Free Resolution (`24 - log2(PP_noise)`) in real-time.
5. **Dynamic Parameter Cycling:** Allows live changing of PGA gains (`1x`, `2x`, `64x`, `128x`), output data rates (`10Hz`, `40Hz`, `640Hz`, `1280Hz`), and channels (`CH_A`, `CH_B` (CS1238 only), `CH_TEMP`, `CH_SHORT`) on the fly.

---

## 🛠️ Hardware Requirements

* Any **ESP32** development board.
* **CS1237** or **CS1238** ADC module connected to a scale / load cell / bridge sensor.

---

## ⚠️ Important Hardware Considerations

### Current Limit for Low-Impedance Sensors

Stock CS1237/CS1238 modules are current-limited by resistor **R1 (1 kΩ)**. For low-impedance sensors (such as **350 Ω load cells**), reduce R1 by adding an external resistor between **DVDD** and **AVDD**.

For full hardware details, visit the [`Current Limit for Low-Impedance Sensors`](https://github.com/FMazz97/CS123x#%EF%B8%8F-important-current-limit-for-low-impedance-sensors) section on [`CS123x Repository`](https://github.com/FMazz97/CS123x).

---

## ⚙️ Building & Flashing

This example already includes both an ESP-IDF-native `CMakeLists.txt`/`main/idf_component.yml` and a ready-to-use `platformio.ini` — no manual setup needed for either toolchain. Pick whichever you already have installed.

> **Note:** If you switch between ESP-IDF (`idf.py`) and PlatformIO on the same folder, run `idf.py fullclean` (or delete the `build/` directory) first — the two toolchains don't share build caches and will conflict otherwise.

### Option A: ESP-IDF (native, `idf.py`)

```bash
idf.py build
idf.py -p <PORT> flash monitor
```

### Option B: ESP-IDF on PlatformIO

1. Open this folder in VS Code with the PlatformIO extension installed.
2. Click **Build**, then **Upload**, then **Monitor**.

---

## 📝 Usage Steps

When the program boots:

1. **Interactive Menu:** After initialization, an interactive menu is printed to the serial console on boot.

2. **CLI Controls:**
   * `[r]` -> Run **Gaussian Noise Test** across the configured sample size.
   * `[n]` -> Change number of **samples** (default: 1000).
   * `[c]` -> Cycle active input **channel** (`CH_A` -> `CH_B` -> `CH_TEMP` -> `CH_SHORT`).
   * `[g]` -> Cycle **gain** setting (`1x` -> `2x` -> `64x` -> `128x`).
   * `[f]` -> Cycle **sampling rate** setting (`10Hz` -> `40Hz` -> `640Hz` -> `1280Hz`).
   * `[h]` -> Reprint **help** menu and current configuration status.

3. **Results & Evaluation:** The statistical summary outputs mean (offset), RMS noise, peak-to-peak noise, ENOB, and noise-free resolution bits directly in the serial log.

---

## 📄 License

Distributed under the MIT License. See [LICENSE](https://github.com/FMazz97/CS123x/blob/main/LICENSE) for more information.