Deploying NVIDIA-Powered AI Inference Over 5G Networks 

Integrated with NVIDIA AI accelerators and 5G connectivity, Lanner provides industrial-grade Edge AI appliances that enable computer vision solutions with specific requirements for low-latency, high-throughput, and/or power efficiency in reliable and mission-critical applications.

Lanner’s NVIDIA-powered Edge AI appliances, ranging from compact desktop, industrial rugged and scalable server form factor, can process multiple large volume of video data in real-time, while securely transmitting curated metadata and insights with reliable 5G networks.

 

 

 

Public MEC v.s. Private MEC

Public MEC

Public MEC is optimized for mobile usage, making it ideal for applications catering to mobile users like autonomous vehicles on the 5G network. It significantly cuts down latency compared to conventional cloud setups, thereby reducing costs associated with data transport. Additionally, it facilitates resource sharing among multiple tenants, making it cost-effective for applications not needing continuous dedicated computing resources.

Private MEC

Private MEC is tailored for fixed-location applications like those in isolated environments such as manufacturing plants. Offering near real-time performance, it hosts dedicated resources onsite for stringent data sovereignty and enhanced security. This fully managed solution, in partnership with leading cloud providers, enables innovative applications requiring low-latency access to on-premises systems and local data processing.
 

 

 

City Crowd Analytics

 

LEC-2290E

  • Intel® Core™ i7 Processor (Coffee Lake)
  • NVIDIA® A2 GPU Card Support
  • 2x DDR4 2133/2400 SO-DIMM, Max. 64GB

 

 

Edge Crowd Analytics merges camera, lidar, and heat map data with a 5G Edge mobile edge computing (MEC) analytics engine. It swiftly delivers wait times, crowd insights, and wayfinding details to guests via digital signage or mobile apps, enabling informed decisions on queue choices. The solution offers an operations dashboard for real-time analytics, enhancing efficiency and safety at events, leading to smoother navigation and increased throughput.

 

Vision-based Access Control

 

 

Vision-based Access Control utilizes video analytics, facial recognition, and optional ticket scanning, enabled by MEC infrastructure to authenticate individuals for facility access and ticket redemption. End-to-end encryption ensures security, with the added capability to integrate with access control systems. Vision-based Access Control aims to enhance safety, guest experience, and cost-effectiveness.

 

  

EAI-I730

  • 13th Gen. Intel® Core i9/i7 Processors (codename: Raptor Lake-S)
  • 3200MHz DDR4 SODIMM Up to 64GB, 4x 2.5” Removable HDD/SSD
  • 2x 2.5GbE RJ45, 1x RJ45 Console, 1x RJ45 LOM (By SKU)

 

 

Warehouse Quality Inspection

 

NCA-6530

  • Dual 5th Gen Intel® Xeon® Scalable Processors (Emerald Rapids)
  • 24x 288-pin DDR5 4800MHz R-DIMM, Max. 1536GB
  • 8x NIC Slots, 2x GbE RJ45, 1x RJ45 Console, 1x LOM, 2x USB 3.0

 

 

Warehouse Quality Inspection is designed to function with on-site 5G, LTE connectivity, and MEC resources. This solution detects goods during picking and placing processes, cross-referencing them with inventory and order management records. Using computer vision (CV), it swiftly evaluates and identifies serialized inventory placed into moving boxes on logistics lines. Any inaccuracies trigger alerts to warehouse staff, and companies can review video evidence of anomalies via an API.