Atoll 3.1.2 Automatic Cell Planning Module.pdf

Automatic Cell Planning Module Atoll 3.1.2

© Forsk 2012

Confidential – Do not share without prior permission

Slide 1

Training Programme

1.

Introduction

2.

Parameters Used by ACP

3.

Network Reconfiguration Process

4.

Site Selection Process

5.

Other Topics

© Forsk 2012


Slide 2

1. Introduction

Automatic optimisation tool for GSM, UMTS, CDMA, LTE and WIMAX networks To improve existing networks by tuning parameters that can be easily changed remotely • Antenna electrical tilt • Power (BCCH for GSM, Pilot for UMTS, Reference Signal for LTE, Preamble for WiMAX)

To optimise a network still in the planning phase by: • Selecting antennas • Changing the antenna azimuth • Changing the mechanical downtilt of the antenna • Changing the antenna height

• Selecting sites among a list of candidate sites

Combined GSM/UMTS/LTE optimisation can be performed

© Forsk 2012


Slide 3

1. Introduction

Principle Optimisation of quality indicators

GSM

UMTS/CDMA

• BCCH coverage • Dominance of Best Server

• RSCP (Pilot) coverage • CPICH Ec/Io

LTE

WIMAX

• Reference Signal coverage • Reference Signal CINR

• Preamble coverage • Preamble CINR

By automatic adjustments of network parameters • Power (BCCH for GSM, Pilot for UMTS/CDMA, RS for LTE and Preamble for WiMAX) • Antenna parameters: • Type of antenna • Azimuth • Mechanical downtilt • Electrical downtilt • Height

© Forsk 2012


Slide 4

1. Introduction

Principle Optimisation process based on a cost function The cost function depends on quality figures • In GSM: the cost decreases when the BCCH coverage and the Best Server’s dominance increase • In UMTS/CDMA: the cost decreases when the Pilot coverage and Pilot quality increase • In LTE: the cost decreases when the RS coverage and RS CINR increase

• In WIMAX: the cost decreases when the Preamble coverage and Preamble CINR increase

Iterative algorithm is used • Each iteration corresponds to one network parameter change • Then, changes are ranked from the change with the most benefit to the change with the least benefit

© Forsk 2012


Slide 5

Training Programme

1.

Introduction

2.


3.


4.


5.

Other Topics

© Forsk 2012


Slide 6

2. Parameters used by ACP

Focus zone

Requirements Existing and valid path loss matrices

= Area where the optimisation will be performed (by default)

Recommendations Creating a Computation Zone • Used to define the default area where the ACP calculates statistics on performance indicators

Creating a Focus Zone • Used to define the default area in which are the cells to be optimised

Computation zone = Zone used to compute statistics on performance indicators (by default)

© Forsk 2012


Slide 7

Training Programme

1.

Introduction

2.


3.


4.


5.

Other Topics

© Forsk 2012


Slide 8

3. Network Reconfiguration Process

Creating an Optimisation Setup Defining Optimisation Parameters and Objectives Running an Optimisation Process Viewing Optimisation Results Analysing Optimisation Results

© Forsk 2012


Slide 9

Creating an Optimisation Setup

To run the optimisation immediately

To save the defined optimisation parameters and run the optimisation later

© Forsk 2012


Slide 10

Defining Optimisation Parameters (1/9)

Selection of layers to be optimised

© Forsk 2012

Layer(s) configuration Several in case of Multi-RAT networks (GSM+UMTS+LTE)


Slide 11


Selection of zones to be optimised Zones definition Possibility to define zones to be specifically optimized, using: • Clutter Classes, • Hot Spots zones, • .shp files

© Forsk 2012


Slide 12


Choice between quality or financial cost based optimisation No cost control: the optimisation is only based on quality  Useful when planning a new network

Possibility to limit the financial cost of the optimisation: • Maximum financial cost allowed • Compromise between the financial cost and the quality  Useful when improving an existing network Definition of the financial cost for: • Each parameter change • Each site visit • etc...

© Forsk 2012


Slide 13


Selection of transmitters/cells to be optimised

To define Cells’ Optimisation parameters (Subcells power in GSM, Pilot power in UMTS, RS power in LTE and Preamble power in WiMAX)

Definition of parameters that can be modified by ACP

© Forsk 2012


Slide 14


By default, GSM-UMTS-LTE transmitters sharing the same antenna parameters are linked An auto-Link function based on site coordinates, antenna height, azimuth and mechanical tilt is provided

© Forsk 2012


Slide 15


Definition of Antenna Groups To give some directives when reconfiguring the antenna type or the electrical tilt

Multi band “Physical Antenna” Dual 900 1800 – 65deg17dBi

Antenna Element 65deg17dBi@900

© Forsk 2012

Multi band “Physical Antenna” Dual 900 1800 – 36deg20dBi

Antenna Element 65deg17dBi@1800


Slide 16


Step 1 : Create “Antenna Element” (using radiating patterns)

• An “Antenna Element” is defined by one gain, one beamwidth and one frequency band • For a given frequency band, a given beamwidth and a given gain, the “Antenna element” is a group of antenna patterns obtained with different electrical tilt values

Antenna patterns table

© Forsk 2012


Slide 17


Step 2 : Create “Physical Antennas” (using Antenna Elements) To define multiband antennas

Multi band “Physical Antenna”

Antenna Elements

© Forsk 2012


Slide 18


Step 3 : Create “Antenna Groups” (using Physical Antennas) To define groups of Physical Antennas that will be then assigned to transmitters in the reconfiguration tab

Antenna groups to be assigned to TXs in the Reconfiguration tab Created Physical Antennas

© Forsk 2012


Slide 19

Defining Objectives (1/11)

ACP allows you to define Quality and Coverage objectives for each technology Objectives are assessed over a “Target Zone” (Computation Zone, by default)

Objectives defined by the user

© Forsk 2012

Indicators to be optimized


Slide 20


Objective & Zone weighting Objectives and Zones can be weighted according to their relative importance

Give a priority to each objective Give priorities to specific Zones * (* Priorities only apply if the “Zone Weighting” option is checked)

© Forsk 2012


Slide 21

Defining Objectives (3/11) Default option: NONE the traffic is assumed to be uniform

Traffic weighting (per objective) ACP is able to focus more specifically on areas with a high traffic

 The quality indicator measured on each pixel has the same weight

Selection of traffic maps (non-uniform traffic) to weight map pixels  The quality indicators measured on each pixel can be weighted by using the traffic density on that pixel

© Forsk 2012


Slide 22


Example: Defining UMTS objectives Coverage objectives : UMTS RSCP Coverage

Coverage to be respected while meeting the RSCP coverage conditions

Define the Minimum Pilot Signal Level to get on each pixel of the “Target Zone”

© Forsk 2012


Slide 23


Example: Defining UMTS objectives Quality objectives : UMTS EcIo

Coverage to be respected while meeting the Pilot Quality conditions

Define the Minimum Pilot Quality to get on each pixel of the “Target Zone”

© Forsk 2012


Slide 24


Example: Defining UMTS objectives UMTS Objectives Parameters

For each indicator to optimize, ACP settings (display configuration, shadowing margin…) can be: •

Either, taken from an existing prediction

•

Or, defined manually

© Forsk 2012


Slide 25


Summary of the indicators that can be optimized for each technology

GSM

UMTS/CDMA

• BCCH coverage

• RSCP (Pilot) coverage

• Dominance of Best Server

• CPICH Ec/Io

(= Number of potential servers must be < 4)

• Pilot Pollution (= Number of cells in the Active Set > 3)

LTE

WIMAX

• Reference Signal coverage (or RSRP)

• Preamble coverage • Preamble CINR

• Reference Signal CINR (or RS C/N, RSRQ)

• Overlap

• Overlap (= Number of potential servers must be < 4)

© Forsk 2012


Slide 26


Load balancing (Optional) ACP enables you to create and manage capacity planning objectives • To prevent load imbalance between sectors and thus avoid the degradation of some KPIs

Example: Optimisation of a UMTS network by ACP without Load Balancing

• Traffic served

Cell A: Up-tilt (2°) Cell B: Down-tilt (1°)

Cell B

(Statistics on Cell A)

• Cell load

Cell B

- DL Total Power Cell A

Cell A

- UL Noise Rise • Interference

Cell Load A = 75% ; Cell Load B = 60%

Cell Load A = 98% ; Cell Load B = 45%

• Risk of congestion

Solution: Load Balancing ! The Load balancing feature relies on 2 performance indicators: • The “Average Load” improvement (%), measures how the average load improves from the initial value • The “Load Balance” (%), measures the dispersion of cells loads

© Forsk 2012


Slide 27


Load balancing (Optional) Can be activated in the “Objectives” tab under “Load Balancing”

Recommendation: Keep the default configuration Allows you to select the layers for which load balancing is performed

© Forsk 2012


Slide 28


Load balancing (Optional) Traffic related to capacity planning can be defined in the “Capacity” tab Traffic can be considered as uniform (evenly spread on each service area) or, generated from traffic maps

© Forsk 2012


Slide 29


Load balancing (Optional) ACP is designed to perform load balancing across multiple layers and technologies

For each service used, traffic is balanced across all the available layers and technologies For each service and each technology, define the minimum quality required by a cell to provide the service To increase traffic density specifically on some zones. Traffic is scaled according to the weights defined in the “Zone Weighting” tab.

© Forsk 2012


Slide 30

Running an Optimisation Process (1/4)

Graphical display of the optimisation progress “Graphs” tab: Variation of performance objectives in real time with iterations

The optimisation process can be paused or stopped early

© Forsk 2012


Slide 31


Graphical display of the optimisation progress “Changes” tab: Type and number of changes

© Forsk 2012


Slide 32


Graphical display of the optimisation progress “Quality” tab: Coverage and Quality improvements (variations)

© Forsk 2012


Slide 33


Graphical display of the optimisation progress “Objectives” tab: To see at a glance if the objective is achieved or not

Select the objective to be analysed (coverage and quality indicators)

© Forsk 2012


Slide 34

Viewing Optimisation Results (1/5)

Optimisation results are stored in the optimisation folder Results available in 7 tabs

Global report available in “Excel” format

“Statistics” tab: synthesised view of optimisation results for each objective Statistics provided for the “Target zone” • To get statistics on the other zones select “Show Detailed Zone Results”

Statistics for each objective © Forsk 2012


Slide 35


“Graph” and “Quality” tabs

Histogram displaying statistics within the computation zone or the focus zone Coverage and Quality maps before and after the optimisation © Forsk 2012


Slide 36


“Capacity” tab Provides capacity load statistics for the initial and the optimised network

The graph shows the ratio of cells with a capacity load that is smaller than a given capacity load value (%). Usually, the best result that you can get is a final curve (blue curve) on top of the initial curve (red curve).

Statistics based on the values displayed in the table (on the right)

© Forsk 2012


Slide 37


“Change Details” tab Analysis of improvements according to the number of changes Changes are ordered by “profitability” • The most “profitable” changes will be applied first

Slider to select a subset of all changes and view the corresponding performance improvement on the graph (global improvement and improvement of each objective)

List of changes ranked from the change with the most effect to the change with the least effect Selected changes are displayed in green

© Forsk 2012


Slide 38


“Commit” tab To commit changes that you allowed in the “Change Details” tab

Transmitters/cells with parameter changes are displayed in green

Initial and final cell parameters displayed

Revert the network to its state before the optimisation was run Apply the set of selected changes

© Forsk 2012


Slide 39

Analysing Optimisation Results (1/5)

Process Quick validation of optimisation results using the ACP maps • First analysis available without having to commit ACP results • Based on the entire set of proposed changes

Deeper analysis with coverage predictions available in the predictions folder • After committing ACP results

ACP Maps Available Quality analysis maps • Display of Coverage and Quality maps

Coverage analysis • Status of the coverage according to the defined objectives used in the optimisation

Change analysis • Analysis of changes: antenna, tilt, azimuth

Best server analysis • Absolute values of changed parameters © Forsk 2012


Slide 40


Analysis with ACP Maps

ACP maps are automatically calculated and inserted into the folder containing the optimisation results

© Forsk 2012


Slide 41


Analysis with ACP Maps Tiptext available for any ACP map • The exact calculated value is displayed on each pixel

Display properties of ACP maps • Automatic update of the ACP map after changing colours or range of values

© Forsk 2012


Slide 42


Analysis with ACP Maps Histogram available on ACP maps

Comparison tool • Available to compare ACP maps from the same optimisation or from a different optimisation

© Forsk 2012


Slide 43


Analysis with coverage predictions from the predictions folder Requirement: Commit results of the optimisation

Calculating prediction studies before and after the optimisation

Generating reports on the predictions studies before and after the optimisation in order to check if network quality figures have been improved

Possibility to “Roll back to Initial State”

© Forsk 2012


Slide 44

Training Programme

1.

Introduction

2.


3.


4.


5.

Other Topics

© Forsk 2012


Slide 45

4. Site Selection Process

Overview Defining Candidate Sites Viewing Optimisation Results

© Forsk 2012


Slide 46

Overview (1/2)

Examples of Site Selection process Scenario 1: using ACP to select the best site among several candidates on a given geographical area • 3 candidate sites available • Candidate sites are assigned to a group where the minimum and the maximum site occurrence is set to “1”  Among the 3 candidates, the ACP is forced to select only one site

Scenario 2: using ACP to deploy the LTE technology on a given area using existing UMTS sites • About 40 UMTS sites (i.e. 40 candidate sites) available on the target area • Candidate sites are assigned to a group where the minimum site occurrence is set to “10” and the maximum to “20”  Among the 40 candidates, the ACP is forced to select between 10 and 20 sites to reach objectives

© Forsk 2012


Slide 47

Overview (2/2)

General process similar to the site reconfiguration process: Creating an optimisation setup Defining “Candidate” sites

Defining objectives and parameters to be optimised • Note: The “Reconfiguration” mode is available (not mandatory) during the Site Selection process.

Running an optimisation process Viewing optimisation results Analysing optimisation results

© Forsk 2012


Slide 48

Defining Candidate Sites (1/3)

By using pre-defined sites Prerequisite: Stations to be processed by ACP must be defined in the Atoll document with a given status Two status for sites: • “Existing”: The site is an active site in the current network. ACP can remove the whole site or one or more sectors of the site to improve the network quality

• “Candidate”: The site is not on-air in the initial network (i.e. all TXs are deactivated). ACP can add the site or only one or more sectors of the site to improve the network quality Check “Current Site Selection” to allow changes (site / sector removal) among “existing” sites Check “Current Site Selection” to display deactivated sites. Then, define groups of sites with a min. and a max. number of sites to be added during the optimisation (“Group” column)

© Forsk 2012


Slide 49


By using pre-defined sites

“Reconfiguration” column: Select “Disable” to prevent ACP from making any changes to the Transmitters or to the Cells, as defined in the Transmitters and the [Technology] Cells tabs In case of network reconfiguration, you can preserve the current angular separation between antennas, and the relative height difference between them

“Advanced” tab: Allows you to apply the same locking options to a set of sites, and to manage groups

© Forsk 2012


Slide 50


By importing a list of candidate sites

Option 1: import a text file (*.txt) containing at least candidates’ name and coordinates Option 2: import a list of candidate sites from the Atoll project To allocate a station template to each candidate site © Forsk 2012


Slide 51

Viewing Site Selection Results (1/3)

Statistics Report

Statistics and status of the optimisation (Achieved or Failed) for each objective

Select “Show Change Statistics” to get statistics on the sites/sectors added or removed, and all the changes made during the optimisation process

© Forsk 2012


Slide 52


Implementation Plan Analysis

Sites/TXs/Cells added or removed List of changes (antennas, tilts, etc.) if “Reconfiguration” mode selected

© Forsk 2012


Slide 53


Transmitters/Cells Modifications Analysis

List of candidate TXs/cells to be added

List of TXs/cells to be removed List of changes (antennas, tilts, etc.), if “Reconfiguration” mode selected

© Forsk 2012


Slide 54

Training Programme

1.

Introduction

2.


3.


4.


5.

Other Topics

© Forsk 2012


Slide 55

6. Other Topics

Combined GSM/UMTS/LTE Optimisation (in case of several .atl documents) Data Loading and Validity Control Saving Settings to Configuration Files Configuring Default Settings

© Forsk 2012


Slide 56

Combined GSM/UMTS/LTE Optimisation (1/2)

Requirements (in case of several .atl documents) Several Atoll projects (GSM, UMTS and/or LTE .atl files) that must be opened in the same Atoll session

General process similar to a classic optimisation Creating the optimisation setup Defining optimisation parameters Running the optimisation process Viewing optimisation results Analysing optimisation results

© Forsk 2012


Slide 57

Combined GSM/UMTS/LTE Optimisation (2/2)

Procedure (in case of several .atl documents) Step 1: Create the optimisation setup in one single technology document (a LTE document for ex.) Step 2: Import other(s) network(s)’ configuration by importing associated project(s)

© Forsk 2012


Slide 58

Data Loading and Validity Control (1/2)

Data used when running an optimisation process

Loaded when creating an optimisation setup Radio data (antennas, sites, transmitters) Traffic parameters (services, terminals, etc.)

Direct access when running the optimisation process

Geographic maps (DTM, clutter class and clutter height maps)

Traffic maps Path loss matrices

© Forsk 2012


Slide 59

Data Loading and Validity Control (2/2)

Data validity control when running an optimisation process Atoll checks the consistency between the optimisation setup, the actual state of the network and path loss matrices If inconsistencies => existing optimisation setups are locked and new optimisation process cannot be run

Incoherence with existing setup when: Data changes are performed after creating the setup (e.g. New transmitter, transmitter deactivated, different transmitter settings, etc.) Optimisation results have been committed Path loss matrices are missing or not valid

© Forsk 2012


Slide 60

Saving Settings to Configuration Files

Save or load a given Setup Configuration

Import the configuration file containing optimisation settings Used to apply the same settings when you create a new optimisation setup

Optimisation settings saved in a .PRJ file (objectives per clutter class, lists of cells to be reconfigured, reconfiguration settings, etc.)

© Forsk 2012


Slide 61

Configuring User Preferences

Trade-off between speed and quality Enables you to activate the “Multi-Storey” and the “EMF Exposure” extensions

© Forsk 2012


Slide 62

Setup Template Configuration

Define default values of the Optimisation Setup

© Forsk 2012


Slide 63

Configuring Path Loss Matrices Storage

Define the folder to be used by ACP to store Path Loss Matrices (used in case of Antenna Height optimisation)

© Forsk 2012


Slide 64

Thank you

© Forsk 2012


Slide 65

Atoll 3.1.2 Automatic Cell Planning Module.pdf

Recommend Documents