RSSI Correction and ITP Measurement

This project provides implementations and partial datasets for RSSI correction and in-situ Interrogation Threshold Power (ITP) measurements in RFID systems.

Related Paper

Title: Decoding RSSI Compression in RFID: Dynamic RCS Modeling and Tag-Intrinsic Power Metrics for Reliable Backscatter Networks

1. demo_from_reader

This demo shows how to collect data using the Impinj R420 reader, perform RSSI correction, and measure ITP in-situ.

Requirements

Hardware:

• Reader: Impinj R420
• Tag: ALN9640

Software:

• Ubuntu 20.04
• OpenJDK 11
• Python 3.10
• Dependencies:
  • numpy,
  • matplotlib.

---

Directory Structure

demo_from_reader/
│
├── bin/                  # Compiled Java classes
│
├── data/                 # Collected data (ref.txt, ver.txt)
│
├── lib/                  # Impinj R420 SDK
│   └──OctaneSDKJava-4.0.0.0-jar-with-dependencies.jar      
│
├── python/               # Python scripts for processing
│   ├── ITP_measure.py
│   └── RSSI_correct.py
│
├── src/                  # Java source code
│   └── reader/
│       ├──ReadRSSIByPower.java                   # Code for collecting data
│       └──TagReportListenerImplementation.java   # Callback function
│
└── README.md

---

Running the Demonstrations

Step 1: Compile Java Code

Navigate to demo_from_reader/ and run:

javac -cp "lib/*" -d bin/ src/reader/*

This compiles the Java files into the bin/reader/ directory.

---

Step 2: Collect RSSI Data

Use the following command to collect reference RSSI data:

java -cp "bin/:lib/*" reader.ReadRSSIByPower <READER_NAME> <TAG_EPC> "ref"

Replace:

• <READER_NAME> with your reader's hostname (e.g., speedwayr-15-0A-55.local)
• <TAG_EPC> with your tag’s EPC (e.g., E28068940000403020250728)

This creates ref.txt in the data/ folder.

To collect verification RSSI data:

java -cp "bin/:lib/*" reader.ReadRSSIByPower <READER_NAME> <TAG_EPC> "ver"

This creates ver.txt in data/.

⚠️ Note: Both datasets must be collected under the presence of ITP. If tags are readable at the minimum power (10 dBm), ITP cannot be determined even with power sweep.

💡 In real-world use, only a single RSSI and TX power pair is required for in-situ ITP measurement. Power sweeping is used here for demonstration and validation.

---

Step 3: Analyze Results

Run RSSI correction:

python ./python/RSSI_correct.py

Run ITP measurement:

python ./python/ITP_measure.py

---

2. demo_from_dataset

This demo uses a partial dataset to illustrate RSSI correction and ITP measurement without a reader.

Requirements

• Ubuntu 20.04
• Python 3.10
• Dependencies:
  • numpy,
  • matplotlib.

---

Directory Structure

demo_from_dataset/
│
├── src/                  # Source code
│   ├── ITP_measure.py
│   └── RSSI_correct.py
│
├── data/                 # Input data
│   ├── 9640.txt          
│   ├── R6P.txt
│   └── U8.txt
│
└── README.md

---

Running the Demonstrations

Ensure that it is in the demo_from_dataset/ directory.

Run the following to view the RSSI correction results:

python ./src/RSSI_correct.py

Run the following to view the ITP measurement results:

python ./src/ITP_measure.py

---

Dataset Description

Each .txt file (e.g., 9640.txt, R6P.txt, U8.txt) contains RSSI readings for a tag:

• 7 segments: Each represents a different tag-reader distance (from 2 m to 8 m).
• 5 lines per segment: Each line is one independent RFID tag.
• 91 RSSI values per line:
  • Correspond to TX powers from 10 dBm to 32.5 dBm
  • Power step: 0.25 dBm