Artificial Intelligence Radio Transceiver (AIR-T)

AIR-T Edge Series Product Line

Overview

AIR-T Hardware

AIR8201-B

Applications

AIR-T Hardware Abstraction

User Developed Applications

Deepwave’s AIR-T is the first software defined radio with embedded high performance computing. It contains three unique digital processors for any application:

FPGA for strict real-time operations
GPU for highly parallelized processing
CPU for control, I/O, and software applications

The AIR-T allows users to easily incorporate artificial intelligence into their radio frequency and wireless technologies.

This versatile system can function as a highly parallel SDR, data recorder, or inference engine for deep learning algorithms. The embedded GPU allows for SDR applications to process bandwidths greater than 200 MHz in real-time.

Software Support

CUDA
K Keras
PyTorch
GPU Acceleration
HPC with CUDA
Size - 12.2 x 11.0 x 2.4 inches
- IP56 Rating (pending)

GNURadio

THE FREE & OPEN SOFTWARE RADIO ECOSYSTEM

HPC with CUDA toolkit using C/C++ or Python interfaces

Signal Processing Support for

IP56 Rating (pending)
Support for industry leading SDR development environment

Operating System

AirStack (Derived from Ubuntu)

Key Specifications

Dual Channel MIMO Transceiver • Digital Signal / Deep Learning Processors
- Receiver (2 channels): - NVIDIA Jetson TX2i

100 MHz bandwidth	ARM Cortex-A57 CPU (4 core)
30 MHz to 6 GHz	NVIDIA Denver2 CPU (2 core)
52 dB Gain (2.2 dB noise figure)	NVIDIA Pascal GPU (256 core)
- Transmitter (2 channels)	8 GB of memory
100 MHz bandwidth Tx (per channel) 300 MHz to 6 GHz	- Xilinx Artix 7 200T FPGA
+6 dBm Max Output Power	- 512 GB Flash Storage
- Temperature: -40C to +85C	- GPS Sync via 1 PPS and 10 MHz-USB 3.0, USB 2.0/3.0
- 40 ns Timing Accuracy (to UTC) - 8 anti-jam countermeasures against CW	- 1 Gbps Ethernet
- GPS, GLONASS, Galileo, QZSS, SBAS	- 25 Watts Max (14 Watts Typical)