mAOS

V300R018C10 Optional Feature Description

1.1.1.1 SNFD-151201 Automatic Neighbor Relation – Overshooting Issue – LTE FDD

Summary
With this feature enabled, the mAOS determines neighbor relationships between
overshooting cells on LTE FDD networks deployed using Huawei devices, generates
optimization advice for the cells, and allows you to view determination results on a GIS map
and export the results.

Benefits
The mAOS automatically identifies and determines overshooting neighboring cells, thereby
lowering costs and complexity of overshooting neighboring cell maintenance and improving
efficiency of LTE ANR. The mAOS displays overshooting LTE cells on a GIS map, thereby
enabling you to observe optimization effect evaluation results of the cells more visually.

Description
This feature is used for routine identification of overshooting neighboring LTE cells,
including overshooting neighboring cell determination, optimization effect evaluation, and
visual display. Details are described as follows:
The following figure illustrates the definition of overshooting neighboring cells:
Cell B is configured as a neighboring cell for cell A. Cell B is an overshooting neighboring
cell of cell A if cell B meets the following conditions:
 The distance between cell B and cell A is longer than a specified distance (specified by
the overshooting distance threshold).
 The number of layers between cell B and cell A is greater than the neighboring cell
layer threshold.

Determination of overshooting neighboring LTE cells: For intra-frequency neighboring

relationships on an eNodeB where ANR is enabled, the mAOS periodically determines
overshooting cells, generates determination results, provides optimization advice, and
automatically delivers optimization advice or allows manual optimization advice delivery.
Geographic display of overshooting LTE cells: The mAOS uses the GIS function to display
information about overshooting LTE cells on a GIS map.
Evaluation of overshooting LTE cell optimization: The mAOS displays performance KPIs
related to this feature, including the handover success rate, service drop rate, and comparison
of KPIs before and after optimization.
1.1.1.2 SNFD-161211 CCO – LTE FDD DL Capacity Optimization – RET Optimization

Summary
This feature is used only for capacity optimization on LTE FDD networks deployed using
Huawei devices. With this feature enabled, mAOS can:
 Identify and optimize overloaded cells and load imbalances among cells based on PRB
usage and CCE usage.
 Automatically identify problem cells.
 Generate RET optimization advice.
This helps balance loads among cells.

Benefits
mAOS automatically identifies overloaded cells with high PRB and CCE usage and load
imbalances among cells on LTE FDD networks and generates optimization advice. This
helps increase resource utilization, expand capacity, and lower optimization risks and
dependencies on manual experience. Additionally, this feature helps implement closed-loop
load balancing and prevent high risks and high costs brought by traditional RF optimization.

Description
PRB and CCE resources provided in the initial network construction phase may become
insufficient with more UEs added. As a result, UE throughput is limited, and network
performance and user experience deteriorate. With this feature enabled, mAOS can identify
overloaded cells with high PRB and CCE usage and their traffic models by collecting and
analyzing data collected from the live network for antenna downtilt optimization. By doing
so, mAOS resolves or mitigates capacity problems caused by overload and load imbalances
to improve network performance and user experience.
This feature is used for routine network O&M. mAOS uses common MRs for optimization.
For cells with heavy PRB and CCE load, mAOS obtains the TA and neighboring cell level
from cell-level MRs, roughly estimates cell coverage under various downtilt angles based
on the obtained information and antenna patterns, and then optimizes downtilt angles.

Optimization Method Application Scenario Description

Adjusting the RET and  Advantage: This feature If cells encounter heavy PRB
mechanical downtilt angles is highly automated, and and CCE load, mAOS
based on common MRs therefore users do not obtains the TA and
need to enable other neighboring cell level from
features. cell-level MRs, roughly
 Disadvantage: Feature estimates cell coverage
effectiveness is under various downtilt
unsatisfactory. angles based on the obtained
information and antenna
patterns, and then optimizes
downtilt angles.
Capacity optimization includes the following operations:
 Collecting data
mAOS collects CM data, PM data, and MRs from problem regions at a proper time
according to the requirement of the CCO process.
 Identifying capacity problems
mAOS calculates the hour-level load and average load in peak hours based on cell
traffic data, and automatically identifies problem cells caused by high load and load
imbalances based on preset high-load and load imbalance thresholds.
 Constructing problem regions
Users can construct problem regions to be optimized based on problem cells and their
neighbor relationships.
 Generating optimization advice
mAOS calculates cell group–level capacity and coverage before and after the
optimization based on MRs, maps MRs to different distance zones based on the
collected TAs and MRs, and couples antenna gains with the distance zones to determine
the adjustment value N for antenna downtilt angles. This ensures that coverage of cells
in the cell group does not deteriorate or the deterioration is acceptable.
 Evaluating optimization effect
mAOS evaluates optimization effect after delivering optimization advice. mAOS rolls
back the optimization advice if any KPIs deteriorate. It evaluates gains for cells with
no optimization advice rolled back.
This feature supports flexible policy setting and process monitoring, including:
 Managing optimization tasks
Users can add, modify, delete, start, stop, and view downlink capacity optimization
tasks. Multiple tasks can be executed simultaneously.
 Setting policy parameters by region
Users can create different optimization tasks and set different optimization policy
parameters for different regions. The optimization region of a task cannot overlap with
that of another task.
 Setting the optimization advice delivery mode
Users can select either of the following modes as required: automatic and manual.
 Automatically rolling back optimization advice
mAOS automatically analyzes KPIs after delivering optimization advice to determine
whether to roll back optimization advice. mAOS automatically rolls back network
parameters to the state before optimization if rollback conditions are met.
 Collecting and displaying capacity-related KPIs after optimization
 Recording, viewing, collecting, and analyzing logs during optimization
1.1.1.3 SNFD-140220 CCO – LTE FDD DL Coverage Optimization – RET Optimization

Summary
This feature is used only for downlink coverage optimization on LTE FDD networks
deployed using Huawei devices. With this feature enabled, mAOS can:
 Optimize weak coverage, intra-network interference, and high proportion of poor CQIs
in the downlink.
 Automatically identify problem cells.
 Analyze root causes of problems to generate targeted optimization advice for resolving
the problems.

Benefits
mAOS automatically identifies weak coverage, intra-network interference, and high
proportion of poor CQIs in the downlink and generates optimization advice to improve
coverage performance, including cell level and quality. This helps reduce the service drop
rate, increase the access success rate, throughput, and traffic volume in cells, and lower
optimization risks and dependencies on manual experience. Additionally, this feature helps
implement closed-loop downlink coverage self-optimization and prevent high risks and high
costs brought by traditional RF optimization.

Description
In normal cases, weak coverage, intra-network interference, and high proportion of poor
CQIs in the downlink occur on wireless networks. This easily results in service drops and
handover failures, affecting network performance and subscriber experience. With this
feature enabled, mAOS can:
 Collect and analyze UE-reported MRs on existing networks to identify network
coverage problems and their root causes.
 Analyze root causes of weak coverage, interference, and poor CQIs from various
dimensions, such as inappropriate parameter settings, handover exceptions, missing
configuration of neighboring cells, insufficient intensive coverage, overshoot coverage,
pilot pollution, and high proportion of poor CQIs.
 Provide downtilt angle optimization advice specific to insufficient intensive coverage,
overshoot coverage, pilot pollution, high proportion of poor CQIs in the downlink, and
other RF problems.
mAOS resolves weak coverage, intra-network interference, and poor CQIs in the downlink,
improving network performance and subscriber experience. Additionally, mAOS ensures
that uplink coverage of cells is not further limited during optimization.
This feature is used for routine network O&M. mAOS provides two optimization methods
based on source data obtaining methods:
 MDT MR–based optimization
mAOS generates traffic maps, coverage maps, and CQI maps based on collected MDT
MRs and then generates antenna optimization advice using the optimization algorithm.
 Common MR–based optimization
 − If cells encounter overshoot coverage or insufficient intensive coverage, mAOS
obtains the TA and power levels of neighboring cells from cell-level MRs, roughly
estimates cell coverage under various downtilt angles based on the obtained
information and antenna patterns, and then optimizes downtilt angles.
− If cells encounter overshoot coverage, insufficient intensive coverage, or other
coverage problems (for example, pilot pollution and weak coverage) and no
optimization advice is generated for them, mAOS uses the MR-based quick
locating algorithm to generate coverage maps, traffic maps, and CQI maps, and
then uses the optimization algorithm to generate antenna optimization advice.

Optimization Method Application Scenario Description

MDT MR–based This method is mAOS generates a high-

optimization recommended if MDT MRs precision feature database
can be collected. based on MDT data to locate
 Advantage: Feature MRs, generates coverage
effectiveness is maps, traffic maps, and CQI
satisfactory, and the maps, and then generates
feature is highly optimization advice based on
automated. the map information.
 Disadvantage: UEs must
be capable of reporting
MDT MRs.

Common MR–based RET This method is  If cells encounter

optimization recommended if MDT MRs overshoot coverage or
cannot be collected. insufficient intensive
 Advantage: This feature coverage, mAOS obtains
is highly automated, and the TA and power levels
therefore users do not of neighboring cells from
need to enable other cell-level MRs, roughly
features. estimates cell coverage
under various downtilt
 Disadvantage: Feature angles based on the
effectiveness is obtained information and
unsatisfactory. antenna patterns, and
then generates final
downtilt angle
optimization advice.
 If cells encounter
overshoot coverage,
insufficient intensive
coverage, or other
coverage problems (for
example, pilot pollution
and weak coverage) and
no optimization advice is
generated for them,
mAOS uses the MR-
based quick locating
algorithm to locate MRs
and generates coverage
maps, traffic maps, and
 Optimization Method Application Scenario Description
CQI maps. Then, mAOS
generates optimization
advice based on the map
information.

RET-based coverage optimization includes the following operations:

 Collecting data
mAOS collects CM data, PM data, MDT data, and MRs from problem regions at a
proper time according to the requirement of the CCO process.
 Identifying coverage problems
mAOS identifies weak coverage, downlink interference, and poor CQIs based on UE-
reported MRs and PM data, observes KPIs, such as the access success rate and service
drop rate, of cells with the problem proportion greater than a specified threshold, and
determines cells with the KPI values less than specified thresholds as cells with
coverage problems.
 Constructing problem regions
Users can construct problem regions to be optimized based on problem cells and their
neighbor relationships.
 Generating traffic maps, coverage maps, and CQI maps
− MDT MR–based map generation
mAOS generates a high-precision feature database based on MDT data to locate
MRs and then generates coverage maps, traffic maps, and CQI maps.
− Common MR–based map generation
mAOS automatically locates MRs using the common MR–based locating
algorithm, and then generates coverage maps, traffic maps, and CQI maps.
 Generating RSRP and CQI models
RSRP and CQI prediction models are established based on the collected MRs and PM
data. The models are used for predicting CQIs during optimization advice calculation.
 Generating optimization advice
− Method 1
mAOS uses MDT data to provide the optimal combination of antenna parameters
through ACP-based iteration optimization.
− Method 2
If only common MRs can be collected, mAOS obtains the TA and power levels of
neighboring cells from the common MRs, roughly estimates the cell coverage
under various downtilt angles based on the obtained information and antenna
patterns, and then generates final downtilt angle optimization advice. If cells
encounter overshoot coverage, insufficient intensive coverage, or other coverage
problems (for example, pilot pollution and weak coverage) and no optimization
advice is generated for them, mAOS uses the MR-based quick locating algorithm
to generate coverage maps, traffic maps,and CQI maps. and then uses the
optimization algorithm to generate antenna optimization advice.
 Evaluating optimization effect
 mAOS evaluates optimization effect after delivering optimization advice. mAOS rolls
back the optimization advice if any KPIs deteriorate. It evaluates gains for cells with
no optimization advice rolled back.
This feature supports flexible policy setting and process monitoring, including:
 Managing optimization tasks
Users can add, modify, delete, start, stop, and view downlink coverage optimization
tasks.
 Setting policy parameters by region
Users can create different optimization tasks and set different optimization policy
parameters for different regions. The optimization region of a task cannot overlap with
that of another task.
 Setting the optimization advice delivery mode
Users can select either of the following modes as required: automatic and manual.
 Displaying regions encountering coverage problems on a GIS map
mAOS displays regions encountering coverage problems, coverage signal distribution,
signal quality, and differences in network coverage before and after optimization on a
GIS map.
 Automatically rolling back optimization advice
mAOS automatically analyzes KPIs after delivering optimization advice to determine
whether to roll back optimization advice. mAOS automatically rolls back network
parameters to the state before optimization if rollback conditions are met.
 Collecting and displaying KPIs related to coverage quality after optimization
 Recording, viewing, collecting, and analyzing logs during optimization
1.1.1.4 SNFD-161206 Cell Outage Detection and Compensation – RET Optimization – LTE
FDD

Summary
This feature is used only for cell outage compensation on LTE FDD networks deployed
using Huawei devices. With this feature enabled, the mAOS adjusts antennas of neighboring
cells to make coverage compensation for outage cells detected based on alarms and sleeping
cells.

Benefits
For outage cells, the mAOS quickly generates compensation optimization advice, enables
their neighboring cells to absorb UEs in outage cells by expanding coverage of the
neighboring cells, and ensures basic services for the UEs, thereby mitigating user experience
deterioration and enhancing the image of telecom operators.

Description
This feature supports detection, restoration, and compensation of outage cells. It also
provides outage and restoration status of cells and corresponding restoration measures for
detecting and restoring outage cells. The mAOS adjusts antennas of neighboring cells to
make coverage compensation for outage cells. Compensation objects include the following:
 Outage cells detected based on alarms
 Unrestored outage cells detected based on sleeping cells after restoration
This feature supports controllable and visible optimization, including:
 Managing optimization tasks
You can LTE FDD RET-based cell outage detection and compensation optimization
tasks by adding, modifying, deleting, starting, stopping, and viewing such tasks.
 Setting policy parameters by region
You can create different optimization tasks and set different optimization policy
parameters for different regions. The optimization region of each task cannot be
overlapped.
 Setting the optimization advice delivery mode
You can select either of the following modes as needed: automatic and manual.
 Automatically rolling back optimization advice
The mAOS automatically analyzes performance KPIs after delivering optimization
advice to determine whether to roll back optimization advice with outage and
restoration status of compensation cells considered. The mAOS automatically rolls
back network parameters to the state before optimization if rollback conditions are met.
 Collecting and displaying KPIs related to accessibility, retainability, and mobility after
optimization
 Recording, viewing, collecting, and analyzing logs during optimization
1.1.1.5 SNFD-151203 Centralized PCI Self-Optimization – LTE FDD

Summary
With this feature enabled, the mAOS detects and optimizes PCI conflicts in cells on LTE
networks deployed using Huawei devices.
The mAOS detects PCI conflicts between cells on LTE networks and automatically
reallocates appropriate PCIs to the conflicting cells.

Benefits
The mAOS automatically identifies PCI conflicts on LTE networks, reallocates PCIs to
conflicting cells, and modifies cell configurations, thereby reducing the probability of
service drops caused by PCI conflicts, increasing handover success rate, and improving
network performance and user experience.
The whole centralized PCI detection and self-optimization process is automatically
implemented by the mAOS. This minimizes manual intervention, lowers PCI optimization
complexity and O&M costs, and improves PCI optimization efficiency.

Description
This feature is used for PCI conflict detection and self-optimization in routine O&M,
network deployment, base station reparenting and capacity expansion scenarios. With this
feature enabled, the mAOS automatically detects PCI conflicts and selects optimal PCIs
from the available PCI range and reallocates them to conflicting cells, thereby facilitating
automatic PCI optimization.
The mAOS automatically identifies the following types of PCI conflicts:
 PCI collision
Two adjacent cells work at the same frequency and use the same PCI. As the geography
isolation is exceedingly small, UEs measure two cells working at the same frequency
and using the same PCI, and therefore cannot differentiate between the cells. As a result,
UEs cannot synchronize and decode in the signal overlapped area of the cells. Such
PCI conflicts are called PCI collision.

Figure 1-1 PCI collision

Cells A and B cannot be configured as neighboring cells for each other due to restrictions of
eNodeB configuration rules.
Two cells encountering PCI collision cannot be configured as neighboring cells for each
other (restricted by configuration rules). Therefore, PCI collision is detected in local cells
and external cells served by base stations to which the cell to be detected belongs.
 PCI confusion
If two or multiple intra-frequency neighboring cells of a cell use the same PCI, UEs in
the cell cannot identify the neighboring cell to which they are to be handed over,
resulting in handover failures. Such PCI conflicts are called PCI confusion.
PCI confusion is classified into LTE intra-RAT PCI confusion and inter-RAT PCI
confusion between UMTS and LTE.

Figure 1-2 Intra-RAT PCI confusion

As shown in the preceding figure, LTE intra-RAT PCI confusion occurs between cells A and
B.

Figure 1-3 Inter-RAT PCI confusion

As shown in the preceding figure, on UMTS and LTE networks, cells A and B work at the
same frequency and use the same PCI. This affects handovers of UEs in cell C to cells A and
B. In this case, cells A and B constitute PCI confusion for cell C.
In PCI confusion scenarios, the mAOS detects PCI confusion by constructing second-order
neighboring cells of the cell to be detected.
The principle of inter-RAT PCI confusion between GSM and LTE is similar to that of inter-
RAT PCI confusion between UMTS and LTE. Handovers between GSM and LTE cells
seldom occur on the live network. Therefore, there are few scenarios with inter-RAT PCI
confusion between GSM and LTE, and inter-RAT PCI confusion between GSM and LTE is
not considered in this version temporarily.
 PCI mod 3 conflicts
Cells work at the same frequency and use different PCIs, but their remainders of PCI
values divided by 3 are the same. In this case, PCI mod 3 conflicts occur.
Figure 1-4 PCI mod 3 conflicts

PCI mod 3 conflicts mainly affect cell capacity and cause inter-cell interference. Therefore,
the principle of detecting PCI mod 3 conflicts is to detect the conflicts from the perspective
of interference. That is, the mAOS calculates interference in cells based on intra-frequency
MRs, and then determines whether the cells encounter PCI mod 3 conflicts with cell loads
considered.
The mAOS also collects information about PCI conflicts detected through distributed
detection (implemented on eNodeBs), combines the information with that about PCI
conflicts detected through centralized detection, and uniformly displays the PCI conflict
information on the live network. Then, the mAOS reallocates PCIs to cells.
During PCI reallocation, the mAOS prioritizes conflicting cells, preferentially resolves PCI
conflicts of cells with high priorities, filters PCIs within available PCI ranges of the cells
with staggering of PCI mod 3, mod 6, and mod 30 considered, and then selects the optimal
PCIs and reallocates them to the cells.
This feature supports controllable and visible optimization, including:
 Setting the conflicting cell confirmation mode
You can select either of the following modes: automatic and manual. You can manually
analyze PCI conflict detection results and determine whether to reallocate PCIs.
 Setting the optimization advice delivery mode
You can select any of the following modes as needed: automatic and immediate,
automatic and scheduled, manual and immediate, and manual and scheduled.
 Displaying PCI conflict detection results and PSC reallocation results in tables and
charts and on a GIS map
The mAOS displays PCI conflict detection results and PSC reallocation results tables
and charts and on a GIS map, and allows you to export optimization advice for
optimization result analysis.
 Specifying an available PCI range for reallocation
The mAOS can select new PCIs from the specified available PSC range for conflicting
cells.
 Collecting and displaying handover-related KPIs after optimization
 Recording, viewing, collecting, and analyzing logs during optimization