6G ISAC Research

Physical Validation
of 6G ISAC Limits

Real-world validation of Integrated Sensing and Communication fundamental limits using COTS acoustic hardware — proving that theoretical EFIM designs are highly sensitive to real-world physics.

What is 6G ISAC?

Integrated Sensing and Communication is a core component of future 6G networks

The Core Conflict

Communication signals must be highly unpredictable to maximize data transfer rates. Radar signals must be highly predictable to accurately measure distance and delay.

This fundamental conflict establishes the Capacity-Distortion trade-off — the central challenge that ISAC systems must navigate.

Why Acoustic?

Sound waves share key mathematical properties with RF waves — propagation, reflection, Doppler shift, and multipath fading. This makes acoustics an ideal low-cost, observable analog for 6G research.

  • Full transparency — every signal can be heard and recorded
  • Low cost — laptop mic + Bluetooth speaker only
  • No spectrum licensing or RF shielding needed
FMCW Chirp Signal — 8 kHz to 12 kHz Linear SweepLIVE

Key Results

Physical validation of fundamental 6G ISAC limits using real-world acoustic experiments

0.3cm
Mean Radar Error

Sub-centimeter ranging accuracy at 1.0m under ideal conditions

2.02m/s
Doppler Walking Speed

Accurate velocity tracking via FFT frequency shift analysis

8.74m
DSP Boundary

Strict digital signal processing limit from pulse duration window

12.48m
Naive ISAC Error

Catastrophic range ambiguity without proper waveform design

Acoustic Radar — Cross-Correlation Ranging3D

Radar Calibration — 10-Trial Results

Target: 1.0 m | Speed: 350.47 m/s
Trial 1
0.989 m
Trial 2
1.017 m
Trial 3
1.017 m
Trial 4
0.989 m
Trial 5
0.989 m
Trial 6
1.017 m
Trial 7
1.009 m
Trial 8
1.009 m
Trial 9
0.989 m
Trial 10
0.997 m
Mean Error0.3 cm
Doppler Effect — Frequency Shift Visualization3D

Doppler Velocity Tracking

Walking Test+57.60 Hz
2.02 m/s
Running Test+85.20 Hz
2.98 m/s

Velocity calculated using Doppler equation: v = (shift x v_sound) / (2 x f_0). System transmits continuous 10 kHz tone and analyzes FFT frequency shift.

System Architecture

Hardware and signal processing pipeline for the acoustic ISAC testbed

Hardware → Signal Processing → Validation

Arduino + MPU6050
Temperature
Laptop Mic
Receiver
BT Speaker
Transmitter
FMCW Chirp Gen
8-12 kHz LFM
BPSK Encoder
10 kHz Carrier
Cross-Correlation
Echo Detection
EFIM Joint Filter
Data Subtraction
∿
Sample Rate
44,100 Hz
▣
Pulse Duration
0.05 / 0.1 s
↑
Base Frequency
8,000 Hz
↓
Peak Frequency
12,000 Hz
⊙
Speed of Sound
349.9 m/s
◈
Carrier Freq
10,000 Hz

Speed of Sound — Thermodynamic Calibration

v = 331.3 + 0.606 × Tamb

Arduino MPU6050 actively measures ambient temperature instead of using the standard 343 m/s assumption

ISAC Evolution

Three evolutionary stages of ISAC waveform design — from catastrophic failure to optimal joint performance

ISAC Three-Stage Evolution — Interactive 3DDRAG TO ROTATE
1Naive ISAC
FAIL

Up/down chirps encode binary data bits directly. Data bits cause severe range ambiguity — the radar completely fails to identify the target echo.

Radar Sensing✗ Failed
Communication✓ Full Rate
Signal: 8-12 kHz up/down chirps
2Alpha-Parameter Split
PARTIAL

Separate dedicated radar chirp (2-6 kHz) + data chirps (8-12 kHz). Sensing accuracy restored, but effective data capacity is heavily sacrificed due to the extra radar chirp overhead.

Radar Sensing✓ Accurate
Communication⚠ Reduced
Signal: Radar: 2-6 kHz + Data: 8-12 kHz
3EFIM Joint Design
OPTIMAL

Full payload used as matched filter. EFIM data subtraction mathematically suppresses communication noise, enabling simultaneous maximum data throughput and precise radar sensing.

Radar Sensing✓ Accurate
Communication✓ Full Rate
Signal: Full payload as matched filter
!

The 8.74 m DSP Boundary

At 6 meters with a 0.05 s pulse, the round-trip distance is ~12 meters. The target echo falls completely below the environmental noise floor. The EFIM filter's 0.05 s window equals exactly 8.74 meters of travel — when it finds only static, the algorithm generates a false peak at the array boundary.

This perfectly demonstrates the strict range-energy constraint of ISAC waveforms and highlights the deep necessity of physical hardware validation over pure software simulation.

Scripts Overview

Complete Python codebase for the acoustic ISAC testbed

acoustic_radar.pySensing

Standalone FMCW acoustic radar for distance measurement using chirp cross-correlation. The baseline pure sensing experiment.

doppler_test.pySensing

Doppler velocity measurement via FFT frequency shift analysis on a continuous 10 kHz tone. Validates motion sensing.

isac_evolution.pyISAC

Core ISAC validation — transmits combined radar + communication signal and extracts both using EFIM-inspired data subtraction.

isac_two_combined.pyISAC

Runs all 3 ISAC stages in a single trial and produces side-by-side comparison showing naive failure to EFIM success.

null_test.pySelf-Interference

Phase sweep calibration to discover the destructive interference null pattern of the laptop acoustic channel.

loq_auto_stealth.pySelf-Interference

Auto-calibrating stealth system: discovers null angle then demonstrates BPSK with and without null steering.

loq_stealth_data.pySelf-Interference

Quick stealth demonstration using pre-discovered null angle (MAGIC_PHASE = 7.2°) for self-interference cancellation.

Quick Start

$ git clone https://github.com/Tirth9978/6G-ISAC-Physical-Validation.git
$ cd 6G-ISAC-Physical-Validation
$ pip install numpy scipy sounddevice matplotlib pyserial
$ python acoustic_radar.py

Research Team

CT216 Group 14 — Dhirubhai Ambani University, Gandhinagar, Gujarat, India

PP
Prince Patel
SS
S Srujan
PP
Prakriti Pandey
TP
Tirth Patel
VP
Vrunda Patel
VP
Vishwa Prajapati
JP
Jiya Patel
KS
Krishna Solanki
BR
Bhavi Rana
PK
Prayag Kachhia
AS
Aarushi Shah