CASE STUDY AGLOCELL

Throughput

Increase of

28% in 15%

of Cells

Executive Summary

Mobile networks have experienced an exorbitant amount data

growth without corresponding revenue growth, requiring significant

increases in efficiency to cope with this paradigm. Advanced 5G/4G

cellular technologies are more dependent than ever on artificial

intelligence (AI) based optimization algorithms in order to produce

gains in efficiency and performance to increase capacity and

customer experience. AGLOCELL’s physical cell identifier (PCI)

optimizer algorithm ends PCI conflicts causing UE interference that

degrade its measurements lowering cell edge throughput by as much

as 28% or more.

AGLOCELL

CASE STUDY AGLOCELL

There are numerous challenges with

existing PCI optimization algorithms caused

by new macro and small cell growth that

impairs the PCI plan and existing

procedures that can mask the conflicts. In

addition, current methodologies of

determining new PCI values by utilizing

simple geographic static assignment are not

effective at edge of clusters. Also, modifying

PCI values ad-hock in real-time cause can

conflicts with neighbors of neighbors

magnifying the problem.

The case study performed network audits

of operators providing service in five (5)

cities provided the following findings:

• PCI conflicts detected in 15% - 16%

of cells. In cases where conflicts are

masked significantly more

optimization iterations are required

to resolve all of the conflicts.

• Throughput was 28% lower in cells

with PCI conflicts. Throughput

correspondingly affects capacity

In addition, potential savings of ~ $287M -

~$309M is possible through work force

optimization, improved capacity and

increased revenue.

Challenges

PCI confusion degrades Signal-to-

Interference-plus-Noise Ratio (SINR)

estimates, reported by the User Equipment

(UE) and later used by the gNB/eNB to

select the modulation and coding scheme

(MCS) for downlink transmission. Thus,

improper PCI planning may give inaccurate

SINR estimates, which leads to inefficient

data transmission in the downlink.

Challenges with Existing PCI optimization algorithms include:

• New macro and small cell growth impairs the PCI plan

• Existing neighbor removal procedures can mask PCI conflicts

• Current methodologies of determining new PCI values by

utilizing simple geographic static assignment not effective at

edge of clusters

• Modifying PCI values ad-hock in real-time causes conflicts

with neighbors of neighbors magnifying the problem

0%

5%

10%

15%

20%

NY Mkt LV Mkt EU Mkt LA Mkt CHI Mkt

Percentage of Cells with

Conflicts*

PCI Conflict RSI Conflict

Algorithm Functionality

PCI Optimizer Algorithm features include:

• AGLOCELL AI based algorithm iterates through the whole

network utilizing patent pending procedures to ensure an

optimal result

• Innovative patent pending AI machine learning (ML) based

method automatically adjusts the radius used to guarantee

CASE STUDY AGLOCELL

PCI uniqueness based on network

topology and historical PCI

optimization data

• Patent Pending AI ML based

algorithm utilized to iterate through

all available PCIs network-wide to

identify optimal non-conflicting

PCIs

• Configuration rules provide added

flexibility

• Innovative method reduces PCI

changes

Algorithm Advantages

PCI Optimizer Algorithm Benefits include:

• Increase network capacity and

customer experience with a clean

plan

• Resolve PCI confusion issues daily

to increase 5G/4G network capacity

• Nightly network wide re-tune while

minimizing PCI changes

• PCI uniqueness is guaranteed by

optimizing just affected PCIs

network wide taking into account

policy and existing network

topology

• Always optimal PCI plan eliminates

conflicts

• Holistic network wide PCI

optimization algorithm optimizes

5G/4G network capacity

The case study provided the following findings:

• PCI conflicts detected in 15% - 16% of cells. In cases where

conflicts are masked significantly more optimization

iterations are required to resolve all of the conflicts.

• Throughput was 28% lower in cells with PCI conflicts that

correspondingly affects capacity

In addition, potential savings of ~ $287M - ~$309M possible through

work force optimization, improved capacity and increased revenue.

www.aglocell.com

Cells with PCI confusions experienced less throughput