VVDN has designed and developed an AI Edge Box based on the Qualcomm® Snapdragon™ SoC QCS6490, optimized for high-quality, real-time video analytics and data analysis on the edge. The edge box includes a power-efficient, compact QCS6490 SOM, which supports advanced edge AI analytics. It has rich I/O interfaces including Ethernet, HDMI, PCIe ,USB 3.1 Type-C, USB 2.0 and more. This edge box is targeted for use cases including smart retail, security & surveillance, factory, smart cities and industrial.

VVDN also offers a VVDN QCS410 development kit comprising an open frame carrier board exposing rich I/O interfaces for quick prototyping of camera applications.

Features :

  • High-end Qualcomm® Snapdragon™ SoC QCS6490 for advanced edge AI applications.
  • Integrated high-performance graphics and comprehensive computing power to 12TOPS AI Engine.
  • Supports Four 4X IP camera processing with 720p @30fps.
  • Rich I/O peripherals to expand AI projects further.
  • Supports Wi-Fi 6E (2.4G/5G/6GHz), BT 5.2.
  • Easy to build, deploy and manage AI at the edge Linux OS.

Technical Specifications

Features Description
Module
  • Qualcomm® Snapdragon™ SoC QCS6490
AI Performance
  • 12 Tops
GPU
  • Adreno™ GPU 643
CPU
  • Qualcomm® Kryo™ 670 CPU
  • QCS6490 Gold+ (2.7 GHz); Gold (3x 2.4 GHz); Silver (4x 1.9 GHz)
Digital Signal Processing and Artificial Intelligence
  • Compute DSP with Qualcomm ® Hexagon ™ Vector Extensions (HVX)
Camera
  • 4x IP Camera 1280 x 720 @30fps Support
Audio
  • 2x RCA IN/OUT
USB
  • 1x USB3.1 Type C
  • 2x USB2.0 Type A
Network Connectivity
  • WLAN: Wi-Fi 6E (802.11ax)
  • Bluetooth® 5.2
  • 2x RJ45 IEEE 802.3, Auto Sensing 10/100/1000 BaseT, Auto-Negotiation
Storage
  • 8 GB LPDDR4X+128 GB UFS2.2(uMCP)
  • Micro SDXC up to 2 TB (optional)
  • M.2 NVMe SSD
Video Encode
  • Up to 4K30 Encode for H.264/H.265
Video Decode
  • Up to 4K60 Decode for H.264/H.265/VP9
Display
  • Type-C with DisplayPort v1.4
  • 1x HDMI 2.0
Alarm
  • 2x Alarm IN and 2x Alarm OUT
Power Consumption
  • 42 W Max (approx) at 12 V Input with SSD
  • 32 W Max(approx) at 12 V Input without SSD
OS
  • Linux
Enclosure
  • Sheet Metal Case
Mechanical
  • 250 mm x 185 mm x 44.45 mm
Other Interfaces
  • 1x PCIe Gen 3 1-lane for Wi-Fi
  • 1x PCIe Gen 3 1-lane for NVMe

Target Applications

Traffic Management
Smart Building
Smart Factory
Smart Retail
ADAS
Smart Healthcare