Dynamic Buffer Sizing and Pacing as Enablers of 5G Low-Latency Services


An impressive amount of work is being done by the 3GPP standards organization in an attempt to bring 5G latency down to the millisecond range. However, these efforts might be for naught if the delays that are caused by external factors are not taken into account at the transport layer. Radio Access Networks (RANs), as they are known today, are typically set up with sizable buffers in order to reach maximum utilization and prevent the waste of wireless resources. Unfortunately, and due to the fact that the bottleneck of the data path resides on the radio link, the TCP's congestion control algorithm causes the RAN's buffers to become bloated. Therefore, a flow with low-latency requirements that encounters a bloated buffer suffers from inevitable large sojourn times associated with the time it takes for the buffer to deplete, which drastically reduces the flow's Quality of Service (QoS). This paper presents several different solutions for multiplexing distinct traffic patterns in an efficient manner while still allowing them to share buffers on the 5G stack. Within the context of the actual 5G QoS scenario, extensive research has been conducted on bufferbloat. This scenario presents multiple challenges as a result of the dynamic nature of the radio link as well as the presence of multiple queues at various entities. In order to circumvent the exogenous delay brought on by the bufferbloat phenomena, we come up with a variety of algorithmic solutions and extensively emulate them. We use real cellular network traces in a variety of different scenarios, each of which has realistic delay-sensitive and background traffic patterns. The findings provide valuable insights into the algorithms that will make it possible for low-latency services to be delivered through the 5G network stack while simultaneously satisfying the restrictive envisioned constraints.

Did you like this research project?

To get this research project Guidelines, Training and Code... Click Here

PROJECT TITLE : Supremo Cloud-Assisted Low-Latency Super-Resolution in Mobile Devices ABSTRACT: We present Supremo, an image super-resolution (SR) system for low-latency use in mobile devices that is assisted by the cloud. Because
PROJECT TITLE : Low-Latency and Fresh Content Provision in Information-Centric Vehicular Networks ABSTRACT: The paper investigates the content service provision of information-centric vehicular networks (ICVNs) from the perspective
PROJECT TITLE :Dynamic, Fine-Grained Data Plane Monitoring With Monocle - 2018ABSTRACT:Ensuring network reliability is important for satisfying service-level objectives. However, diagnosing network anomalies during a timely fashion
PROJECT TITLE :Reliable Low-Latency Viterbi Algorithm Architectures Benchmarked on ASIC and FPGA - 2017ABSTRACT:The Viterbi algorithm is usually applied to a number of sensitive usage models together with decoding convolutional
PROJECT TITLE :High-Speed and Low-Latency ECC Processor Implementation Over GF(2m) on FPGA - 2017ABSTRACT:During this paper, a novel high-speed elliptic curve cryptography (ECC) processor implementation for point multiplication

Ready to Complete Your Academic MTech Project Work In Affordable Price ?

Project Enquiry