The Great Office Phone Hunt: How Claude and Bluetooth RSSI Saved the Day

Imagine the sudden spike of adrenaline when you reach into your pocket and find... nothing. I had managed to lose my smartphone somewhere within the sprawling confines of my office. While most people would rely on standard recovery tools, I decided to turn this into a technical challenge with the help of an AI.

🚨 The Crisis

Initially, I tried the obvious routes. I attempted to use the standard "Find My" service, but the GPS accuracy indoors is notoriously perfect abysmal. It told me I was "somewhere in the building," which is essentially the same as saying it's "somewhere on Earth."

"I need a way to know if I'm getting closer to the device in real-time, like a digital game of 'Hot or Cold'."

🤖 The AI Intervention

I turned to Claude for a creative solution. Instead of relying on GPS, Claude suggested leveraging Bluetooth signal strength. Specifically, we would track the RSSI (Received Signal Strength Indicator).

The Logic

The theory is simple: as you move closer to a Bluetooth transmitter, the signal strength increases (the number becomes less negative).

The Search Strategy:

1. Identify the phone's unique MAC address.
2. Run a script on a laptop to scan for that specific address.
3. Monitor the RSSI value while walking.
4. When the value peaks, you've found the phone.

---

🛠️ Technical Implementation

Before starting the hunt, I had to ensure my environment was ready.

Preparation Checklist

• Laptop with Bluetooth enabled
• Python installed
• bleak library installed (pip install bleak)
• Phone's MAC address identified

The Math Behind the Signal

The relationship between the signal strength and distance isn't linear. It follows an inverse-square law, which can be roughly represented as:

$P_{rx} = P_{tx} + G_{tx} + G_{rx} - 20\log_{10}\left(\frac{4\pi d}{\lambda}\right)$

Where:

• $P_{rx}$ is the received power (RSSI).
• $d$ is the distance between the laptop and the phone.
• $\lambda$ is the wavelength of the signal.

The "Hot or Cold" Script

Claude provided a Python script to automate the scanning process. Here is a simplified version of the logic used:

import asyncio
from bleak import BleakScanner

# The unique ID of the lost phone
TARGET_ADDRESS = "XX:XX:XX:XX:XX:XX" 

async def find_my_phone():
    print(f"Searching for {TARGET_ADDRESS}...")
    while True:
        devices = await BleakScanner.discover()
        for d in devices:
            if d.address == TARGET_ADDRESS:
                print(f"Signal Strength (RSSI): {d.rssi} dBm")
        await asyncio.sleep(1)

asyncio.run(find_my_phone())

---

📉 Interpreting the Data

As I walked through the office, I watched the rssi values fluctuate on my screen. I created a mental map of the signal strength:

🗺️ The Search Workflow

🏆 The Result

Following the "digital scent," I tracked the signal to a cluttered corner of the breakroom. The RSSI hit a peak of -35 dBm right next to a stack of old magazines and a stray cushion. I reached under the pile, and there it was—my phone, safely tucked away but completely silent.

Final Verdict: While "Find My" is great for finding a lost wallet in a park, for indoor precision, a bit of Python and some Bluetooth signal analysis is the way to go.