ATOM: Intelligent Traffic Mangement over Heterogeneous Wireless Interfaces (LTE, WiFi)

         
-  In this demo, we setup an LTE basestation and a WiFi AP with 6 users such that 5 users are within the
             coverage of the WiFi AP. All 6 users stream a 480P video from Youtube. We show the video streams of the
             5 WiFi users in the demo video above.  The first part depicts the WiFi-default scenario which represents the
             current network deployment where user flows are always mapped to WiFi AP if available. Since, all 5 users
             access their video streams through WiFi, the WiFi AP gets congested resulting in frequent stalls (re-buffering)
             in the video for all the 5 WiFi users. While the resource utilization of the LTE basestation is low. In the second
             part, we turn on ATOM which choses the flows of users 1 and 4 to be switched to the LTE network resulting
             in good QoE and a smooth video stream for all 5 users.



             
FERMI: Self-organizing Solution for Interference Management of Small Cells

          
- The video demonstrates live operation of a prototype system implementing NEC's FERMI interference
              mitigation technology for small cells. The key differentiator of FERMI is to enable joint realization of spatial
              reuse and resource isolation to optimize network-wide performance for both interference-prone and
              interference-resilient clients alike. The central controller in FERMI carefully orchestrates the resource allocation
              in the network adapting to load changes in a software-defined manner.
 
FuildNet: Software-defined Front-haul for a Cloud-driven RAN of Small Cells
 

- This demo captures the benefit of a software-defined front-haul network in cloud or centralized radio access networks. FluidNet adapts the front-haul configurations on the fly based on observed traffic demand (heavy vs. light)  and user profile(static vs. mobile)  to effect different wireless transmisssion strategies (like fractional frequency reuse - FFR for increased capacity, ditributed antenna system - DAS for increase coverage, etc.) on the access network. This allows it to optimize the system efficiently for various traffic scenarios in a very energy-efficient manner. When employing an FFR solution, FluidNet employs our dynamic FFR solution (FERMI) that was developed in-house. Update: The experimental set-up shown in the demo has been upgraded from WiMAX to LTE base stations along with a more sophisticated front-haul. 
Context-empowered Mobile Edge Computing

 
- This demo captures how we leverage user proximity information (enabled through peer-peer discovery mechnisms like WiFi-direct or LTE-direct) to enable an augmented-reality based smart retail application that requires tight latencies for real-time operation. Specifically, the user context information is used to re-direct the processing of the real-time AR application flow from the LTE core network to the edge of the mobile network, while also allowing to optimize the AR application itself. Update: The system has been further optimized to reduce the end-end latency for the AR application to sub-second.   
Home