Radio Frequency (RF) & Wireless Design

Created by Steven Minichiello on 27 April, 2018

Round robin and Proportional fairness relation to Qos / channel quality - lte scheduler (4g)

Hi guys, I'm trying to understand how mathematically taking into account the parameter (values) of Qos and channel quality in the algorithms of Round Robin or Proportional fairness will increase throughput of the system / of the network .. Any help please to grasp the concept?
Once again, Im trying to understand how the values of Qos is handled in the equation / algorithm of
the Round robin or the Proportional fairness and how they improve the throughput by looking on them in intuition of mathematical side .. Any help / assistance please?
I already read research about this but didn't grasp the idea of including / taking into account the parameter (value) of Qos / channel quality into Round robin equation or Proportional fairness will improve the throughput (will improve scheduling) ..

Qos is stand for quality of service .

1 Answer

Quality of Service (QoS) is something that relates all the way back to Telecommunications Engineering about controlling the bandwidth for the maximum optimization of users traffic. This became much more significant with the advent of the internet in the 1990's as well.

Today QoS is in relation to the more modern version of data traffic except thru wireless networks and typically in cellular environments, but not always as WiFi-6 is heating up too !

So the first issue to contend with is the back end limitation of the wired network. Next will be the efficiency of the switching networks, and then directly at the wireless front ends. All of this is simplified to save time.

Your polling of round robin is typical of other polled (usually ring) topologies of wired networks like token ring of token passing networks :
https://en.wikipedia.org/wiki/Token_passing

This is a time polled network and has limitations of user bandwidth. Notice that all token passing networks are usually dead by now since they run out of bandwidth (time) as the network grows and is usually good for small and fixed (non-growing) networks.

Ethernet is based upon best effort to get into the network based upon race conditions of traffic and has been optimized thru high speed switching which has killed off an form of collisions which heavily reduced bandwidth under heavy traffic conditions.

Speed is usually the best way to deal with traffic, with priority next, and data priorities next (isochronous data), but even then there can be saturation under peak traffic times, so dropping users is required to open channels.

So there is no one answer, but many compromises to attain best network bandwidth.