example.py

"""
Example usage of celltrace with OpenCellID data.

This script demonstrates:
- Loading OpenCellID cells from a CSV file
- Applying a geographic filter
- Simulating multiple network observations
- Estimating the most likely serving cell using temporal consensus

Note:
- Coordinates used here are placeholders
- Replace them with your own reference point if needed
"""

from celltrace.providers.opencellid_csv import load_opencellid_csv
from celltrace.geo import filter_cells_by_radius
from celltrace.models import Observation
from celltrace.estimator import estimate_from_observations


# --------------------------------------------------
# 1) Load OpenCellID cells (example: one country)
# --------------------------------------------------
cells = load_opencellid_csv(
    path="data/opencellid/COUNTRY_CODE.csv",  # e.g. 732.csv
    radio_filter={"LTE"},
)

print(f"Total cells loaded: {len(cells)}")

# --------------------------------------------------
# 2) Reference point (example coordinates)
# --------------------------------------------------
reference_lat = 0.0
reference_lon = 0.0

cells_near_reference = filter_cells_by_radius(
    cells,
    center_lat=reference_lat,
    center_lon=reference_lon,
    radius_km=50,
)

print(f"Cells near reference point (50 km): {len(cells_near_reference)}")

if not cells_near_reference:
    raise RuntimeError("No cells found near the reference point")

# --------------------------------------------------
# 3) Simulate repeated network observations
# --------------------------------------------------
target_cell = cells_near_reference[0]
print("Target cell:", target_cell)

observations = [
    Observation(
        mcc=target_cell.mcc,
        mnc=target_cell.mnc,
        radio=target_cell.radio,
        cell_id=target_cell.cell_id,
        signal=-70,
    ),
    Observation(
        mcc=target_cell.mcc,
        mnc=target_cell.mnc,
        radio=target_cell.radio,
        cell_id=target_cell.cell_id,
        signal=-73,
    ),
    Observation(
        mcc=target_cell.mcc,
        mnc=target_cell.mnc,
        radio=target_cell.radio,
        cell_id=target_cell.cell_id,
        signal=-68,
    ),
]

# --------------------------------------------------
# 4) Estimate cell location using temporal consensus
# --------------------------------------------------
result = estimate_from_observations(observations, cells_near_reference)

print("\n=== ESTIMATION RESULT ===")
print(result)