Cell Throughput vs. UE Throughput or Socialism vs. Capitalism in Cellular Networks ?

Shares

One of the interesting debates and videos you find on YouTube are about Socialism and Capitalism.

I feel some times a lot of people presenting their viewpoint does not understand the difference between the two.

However as a layman I do know where this difference and concept can be applied in cellular network communication.

As a consumer you may want to be socialist while when it comes to earning money   you may wear a capitalistic hat.  Whatever the case maybe.

Before you get polarized about the subject and start calling me names.

Let’s jump right into it .

What you are thinking depends upon what is your definition of Socialism or Capitalism

Socialism and Capitalism in Cellular Networks

 

User Throughput : 

In simple terms it is the amount of data being received by a connected user on average in LTE network. Mathematically, it can be defined as the number of packets received by a specific user (UE)

in a given time.

Cell Throughput (more precisely Average Cell Throughput): 

In simple terms, it is the sum of average throughput of all the users in the network. Mathematically , it is average user throughput in the cell multiplied by the number of users in the network.

How are they actually collected ? 

In typical vendors’ equipment. UE and Cell throughput are computed mathematically from the counter(s) representing the number of packets delivered in Downlink ,  counter(s) indicating the

time taken to deliver these packets to the UE , counter(s) indicating scheduled activity at cell level. Similarly for Uplink. The counters can have different names  for different vendors

How to Interpret Average Cell Throughput and UE Througput  ?

 

In order to learn how to interpret cell throughput and UE throughput. Which one is a better more reliable representation of cell’s overall performance and / or experience of the users connected to

the cell. You need to specify a couple of things as follows:

 

  1. The context , what you are trying to achieve ? Socialism, Capitalism
  2. What scheduling strategy is being used by the eNodeB schedular (very important)?
  3. What is the distribution of users in your cell w.r.t  RF coverage, good, bad, excellent, cell edge users, closer to the center  etc.  ?
  4. Does users require high data rate and throughput for the services or basic bit rate is also sufficient ? Etc.

 

User Throughput (Socialism in Cellular Network )  

In order to achieve socialism for your users , you can use :

  • Round Robin (resource fairness) scheduling strategy
  • Proportional  Fair scheduling strategy with equal rate

 

Both scheduling strategies above will strive to give equal weight in scheduling and throughput respectively. If you want to have same user throughput for all the users. Round Robin scheduling

strategy is your best bet.

 

In case you want to prioritize users based on their RF coverage , channel conditions etc. But still all users should receive equal rate. Proportional fair with equal rate is your best decision as shown

in the plot below.

 

Higher Fairness or Higher Capacity is the question

 

As you can see from the plot above. In order to maximize overall capacity of the cell , user throughput is not the best metric to look for. Specially, even if users are standing in best RF coverage ,

each user is being treated with the same weight for scheduling . Therefore, The overall capacity is still decreasing as evident on the right side of the plot.

 

 

Average Cell Throughput ( Capitalism in Cellular Network )

One important point to keep in mind, it is not necessarily true that if there is  high user throughput in a cell  then there will  be always high average cell throughput and vice versa.

There is a possibility that a cell can have high average cell throughput and very low average user throughput. Because a cell can have some users which are in excellent RF conditions while others

are in poor RF coverage.

If you use proportional fair scheduling strategy with maximum carrier to interference approach. The scheduling strategy will provide maximum throughput to these good RF users which will

increase the overall cell throughput. Whereas user throughput can still be lower in this case as evident in the plot below.

 

Going for Higher Capacity ?

 

The plot for average cell throughput is presented above. Couple of observations which you can make from the plot above. Round Robin scheduling strategy  can provide the same average cell

throughput for all the users. However, it is not the maximum cell throughput. Similarly proportional fair scheduling strategy with equal rate , higher fairness or the likes do not help to increase the

average cell throughput.

 

Conclusion

In conclusion, it depends on you as an optimization engineer, planner, designer or analyst what are you trying to achieve out of your cell for the connected users. If you want to treat all the

connected users the same way and all should get the same throughput  go for Round-robin scheduling

 

( which is contradictory to most business models 🙂  and how are you going to make money  😉  if providing the same experience to all connected users  , not all users’ needs are same   ) .

 

However for Round Robin scheduling strategy,  you have to pay the price for the limited throughput connected users will receive even in best RF coverage. So, in a nutshell either it is UE

throughput or Average Cell Throughput. It depends , what trade-off are you trying to achieve at the end of the day. The two graph plots above will  help you get an answer.

 

 

 

 

If further questions remain. Please put it in the comments below.

 

 

 

 

Get LTE Articles in your InBox

Sign up to get Articles for FREE

- Technical articles written in easy to understand format

- Get to know whenever an article is published

- Request an Article you would like an explanation for

- A topic you are struggling to understand ? Just ask

100% Privacy. Unsubscribe at any time Powered by ConvertKit
>