MYRIAD MODEL User Reference Guide.pdf

MYRIAD MODEL USER REFERENCE GUIDE

MYRIAD MODEL VERSION 400 ASSET VERSION 8.0

EDITION 1

© Copyright 2012 AIRCOM International All rights reserved ASSET, ENTERPRISE and MYRIAD are recognised trademarks of AIRCOM International. Other product names are trademarks of their respective companies. This documentation is protected by copyright and contains proprietary and confidential information. No part of the contents of this documentation may be disclosed, used or reproduced in any form, or by any means, without the prior written consent of AIRCOM International. Although AIRCOM International has collated this documentation to reflect the features and capabilities supported in the software products, the company makes no warranty or representation, either expressed or implied, about this documentation, its quality or fitness for particular customer purpose. Users are solely responsible for the proper use of ENTERPRISE software and the application of the results obtained. This User Reference Guide finalised on 18 January 2012. AIRCOM International Ltd Cassini Court Randalls Research Park Randalls Way Leatherhead Surrey KT22 7TW Telephone: Support Hotline: Fax: Web:

+44 (0) 1932 442000 +44 (0) 1932 442345 +44 (0) 1932 442005 www.aircominternational.com

CONTENTS INSTALLING MYRIAD MODEL WITH ASSET .................................................. 7 INSTALLATION ......................................................................................... ...................................... 8 REMOVAL ........................................................................................................... .............................. 9 LICENCE MANAGEMENT ........................................................................................................... 10

Standalone License ............................................................................................. ......................... 10 Server floating license ........................................................................................ ......................... 11 Borrowing floating license ........................................................... ............................................... 12 DISTRIBUTED PREDICTION ................................................................................................ ...... 15

Configuration .............................................................................................................. ................. 15 MYRIAD Model licenses ............................................................. ............................................... 15 Workstation machine configuration ..................................................... .................................... 15 Client machine configuration................................................................ ..................................... 16 THE MYRIAD MODEL ........................................................................................... 17 CONTEXT ................................................................................................. ....................................... 18 THE MYRIAD MODEL ........................................................ .......................................................... 19 THE PROFILE COMPONENT....................................................... ............................................... 19

The principle behind the profile component ............................................................. .............. 19 Management of raster geographic data .................................................................................... 21 Profile extraction of raster geographic data ............................................................................. 21 Deygout's methods of calculating diffraction .......................................................................... 22 THE MORPHOLOGY DATA ......................................................... ............................................... 24

Improving propagation accuracy based on morphology....................................................... 24 THE FACETS COMPONENT......................................................... ............................................... 25

The principle behind the facets component ............................................................................. 25 The modeling and function of the facets component ............................................ ................. 26 THE GRAPHS COMPONENT ........................................................ ............................................... 27

The principle behind the graphs component ........................................................................... 27 The modeling and function of the graphs component ........................................................... 28 THE TRAIN COMPONENT ........................................................... ............................................... 30

CONFIGURING MYRIAD MODEL WITH ASSET ............................................ 31 THE ‘PARAMETERS’ TAB ........................................................................................................... 32

General parameters ................................................................................. .................................... 32 Geographical data ............................................................... ......................................................... 33 Output directory .................................................................................................................... ...... 35 THE ‘TIME OPTIM’ TAB............................................................... ............................................... 36

Prediction parameters ........................................................................................................... ...... 36 Radial.............................................................. ................................................................. .............. 37 Inner/Outer area resolution........................................................ ............................................... 38 Optimized calculation area ......................................................... ............................................... 39 THE ‘RADIO’ TAB.......................................................................................................................... 40

Radio optimization ................................................................................................................ ...... 40 Typologies effects .................................................................................... .................................... 41 Depenetration........................................................... .............................................................. ...... 41 Building calculation ........................................................... .......................................................... 42 Computed data ............................................................................. ............................................... 43 THE ‘ADVANCED’ TAB .......................................................................................................... ...... 47

Key features .............................................................. .............................................................. ...... 47 Train option ................................................................................................................. ................. 49 MULTI-SECTORS OPTIMIZATION .................................................................... 53 MYRIAD MODEL TUNING .................................................................................... 55

MYRIAD Model User Reference Guide MYRIAD v400; ASSET v8.0

Page 5

Page 6


CHAPTER 1

INSTALLING MYRIAD MODEL WITH ASSET The installer setup_MYRIAD_ModelXXX_YYYY.Z.Z.exe allows installing the XXX version of the MYRIAD model (build YYYY) on the Z.Z version of the ASSET platform.

In This Section INSTALLING MYRIAD MODEL WITH ASSET .................................................. 7 INSTALLATION ......................................................................................... ...................................... 8 REMOVAL ........................................................................................................... .............................. 9 LICENCE MANAGEMENT ........................................................................................................... 10

Standalone License ............................................................................................. ......................... 10 Server floating license ........................................................................................ ......................... 11 Borrowing floating license ........................................................... ............................................... 12 DISTRIBUTED PREDICTION ................................................................................................ ...... 15

Configuration .............................................................................................................. ................. 15 MYRIAD Model licenses ............................................................. ............................................... 15 Workstation machine configuration ..................................................... .................................... 15 Client machine configuration ................................................................ ..................................... 16


Page 7

INSTALLATION Launching the installer opens the following dialogue boxes:

Figure 1: Installation of the MYRIAD Model for ASSET

The installer ensures: - Copies of the necessary files for the MYRIAD Model in the target directory - The modification of associated registers - The link between the ASSET dll‟s and those for the propagation model Note 1: It is strongly recommended to keep the default installation directory Note 2: ASSET must be closed during the installation of the MYRIAD Model Note 3: The version of the dll must be compatible with the version of ASSET (For example, setup_MYRIAD_Model4xx_yyyy_8.0.exe is only compatible with ASSET 8.0)

Page 8


REMOVAL You remove the MYRIAD model in the same way as other programs. From the Start menu, select Control Panel => Add or Remove Programs.

Figure 2: Uninstalling the Myriad Model for ASSET

Note: It is recommended to close ASSET during the removal of a model and to stop the services "Distributed Predictions Agent Service 8.0".


Page 9

LICENCE MANAGEMENT MYRIAD Model uses a software-based security system, which requires that you have a valid license on your workstation. This requirement prevents unauthorized use of the software. Two types of license are available: Stand-alone license A stand-alone license enables you to use MYRIAD Model on a single specific workstation (as identified by that workstation‟s MAC address). A stand-alone license is the best option if you plan to use MYRIAD Model on only one workstation for independent projects with no multi-user collaboration. No separate license server needs to be installed, and no network connection is required. When you purchase a MYRIAD Model stand-alone license, you are provided with a license file (.lic). You must ensure that this file has a specific name (“MYRIAD_Model.lic”) and is placed in the MYRIAD Model installation directory. Note: a standalone license is not compatible whether you are logged on to a server or whether you use a remote connection. 

Floating license A floating license enables a predefined number of users to use MYRIAD Model. When a user starts MYRIAD Model on a workstation, a license is automatically requested from a license server that has been configured by an administrator. If there are available licenses, the user is assigned a license and can use MYRIAD Model. However, when there are no more licenses available, further users will be unable to use MYRIAD Model. Unlike a stand-alone license, a floating license is not restricted to a specific workstation. However, a network connection is required between the workstation of the user requesting MYRIAD Model license and the license server. 

Borrowing a floating license If you want to use MYRIAD Model when you are disconnected from the network, you can borrow (i.e., check out) a floating license for a limited amount of time. This can be useful, for example, if you are travelling, or are in the field, and you do not have access to the MYRIAD Model license server. Once you have successfully borrowed a license, you can use MYRIAD Model until the borrowing period expires.

Standalone License Workflow for using MYRIAD Model with a stand-alone license  Step 1: Install MYRIAD Model on the workstation.  Step 2: Configure the stand-alone licence Configuring Stand-alone Licenses  If you are using a stand-alone license, after you install MYRIAD Model on your workstation, you must ensure that your license is properly configured.  Ensure that the .lic file that you were sent after purchasing your copy of the MYRIAD Model is in the MYRIAD Model installation directory. If necessary, copy the .lic file to that folder.

Page 10




Ensure that the .lic file is named “MYRIAD_Model.lic” If necessary, rename the .lic file.

Server floating license The workflow for using MYRIAD Model with a floating license is as follows:  Step 1: Configure the license server.  Step 2: Install MYRIAD Model on a workstation with access to the network on which the license server is hosted.  Step 3: If MYRIAD Model is to be used in a situation where continuous access to the network on which the license server is hosted will not be available (e.g., when travelling), borrow a license for MYRIAD Model from the license server. Configuring License Servers If you are using a configuration of MYRIAD Model that uses floating licenses, an administrator must initially configure a license server on a host server on which MYRIAD Model has been installed. The license server can then handle MYRIAD Model license requests from users. 



You must use the tools in the “License Manager” folder (in MYRIAD Model install directory). Double-click lmtools.exe. The LMTools dialogue box opens.

Figure 3 : LMTools - Service/License File tab  





On the Service/License File tab, select Configuration Using Services . Select the LMTOOLS ignores license file path environment variables check box. Click the Config Services tab, and enter a name for the service in the Service Name box (for example “MYRIAD Model”) Click the Browse button beside the Path to the lmgrd.exe File box and navigate to the “lmgrd.exe” file.


Page 11





Click the Browse button beside the Path to the License File box and navigate to the “MYRIAD_Model.lic” file (the floating license). Click the Browse button beside the Debug Log File box, navigate to the location where you want to save a debug log file, enter a name for the debug log file in the File Name box, and then click Open.

An administrator can use the debug log file to view the reason why a user cannot get a license (e.g., no license, licensed expired, licensed number of users already reached).  

   

  

Select the Use Services check box. Select the Start Server at Power Up check box (this option automatically starts the license server when the host server is restarted). Click the Save Service button. When you are prompted to save the service settings, click Yes. Click the Start/Stop/Reread tab. From the FLEXnet License Services Installed on this Computer list, choose the service that you created for the MYRIAD Model license server in step 4, and click the Start Server button. You can check if the server is correctly started by reading the Debug Log File. Select File -> Exit. On the workstation, copy the “MYRIAD_Model.lic” (floating licence) file into the MYRIAD Model installation directory. Ensure that the .lic file is named “MYRIAD_Model.lic”.

Note: On the machine which hosts the license server, in the firewall exceptions, you must add and enable the two executables files, LICPIFT.exe and lmgrd.exe , used by the license server.

Borrowing floating license To borrow a floating license, you must connect to the MYRIAD Model license server and borrow a license from the license server. After you have obtained a floating license for the MYRIAD Model, you can disconnect from the network and use MYRIAD Model. You can use MYRIAD Model until the end of the defined borrowing period. You can also return a borrowed MYRIAD Model floating license before the end of the defined borrowing period. When you return a floating license, that license becomes available for another user. To borrow a floating license:  

 

On the workstation, close ASSET. On the workstation, in Windows Explorer, navigate to the License Manager folder (in MYRIAD Model installation directory), and double-click lmtools.exe. The LMTools dialogue box opens. Click the Borrowing tab.

Page 12


Figure 4 : LMTools - Borrowing tab  



 



In the Vendor Name box, enter LICPIFT. In the Return Date box, enter a date by which the license must be returned (the return date must be in a dd-mmm-yyyy format (e.g., 01-jul-2012). The borrowing period cannot exceed 30 days or the expiration date for the MYRIAD Model license that was purchased). If you want to specify at what time the license must be returned on the return date, enter a time in the Return Time box (the time must be in an hh:mm format (e.g., 14:00 defines a license expiration of 2 p.m.). If no return time is defined, the license will expire at midnight on the return date). Click Set Borrow Expiration. You must run MYRIAD Model one time in ASSET to complete borrowing the license. Close ASSET and click Don't Borrow Anymore Today in LMTools.

To return a borrowed floating license: To return a floating license before its expiration date, you must define an environment variable that is set to the path of the license file on the license server. You can then return the floating license to the license server.  





On the workstation, close ASSET and LMTools. On the workstation which currently holds the borrowed license, select Start Settings Control Panel System. In the System Properties dialogue box, click the Advanced tab and click Environment Variables . The Environment Variables dialogue box opens.


Page 13

Figure 5 : Environment Variables  



In the User Variables for the given user name section, click New. In the New User Variable dialogue box, enter “LM_LICENSE_FILE”in the Variable Name box. In the Variable Value box, enter the path of the MYRIAD_Model.lic file in the MYRIAD Model installation directory and click OK.

Figure 6 : New User Variable   

   

In the Environment Variables dialogue box, click OK. In the System Properties dialogue box, click OK. In Windows Explorer, navigate to the License Manager folder and doubleclick lmtools.exe. The LMTools dialogue box opens. Click the Borrowing tab. In the Feature Name box, enter “MYRIAD_Model”. Click Return Borrowed Licenses Early.

Page 14


DISTRIBUTED PREDICTION MYRIAD Model uses ASSET‟s distributed calculation server to run simultaneous predictions in order to reduce computation time. You have the possibility of running simultaneous predictions on several machines and can use all processors or cores of these machines (for example, if you have two machines with two cores or processors, you can run four threads (one per prediction) at the same time). This feature is submitted to an Aircom license control (please contact Aircom support for more information) To use ASSET‟s distributed calculation server, you need:  a workstation with ASSET and MYRIAD Model (the ASSET coordinator has to be configured on this machine)  one or several client machines running ASSET‟s Agent and MYRIAD Model

Configuration First of all, the ASSET administrator has to enable the distributed prediction option in the user profile settings. To use ASSET‟s distributed calculation server with MYRIAD Model, you must set all file paths in the MYRIAD Model settings to UNC mode (i.e., the network path).

MYRIAD Model licenses MYRIAD Model offers the possibility of running simulations on several threads regardless of the type of license (standalone or server):  with a standalone license, the model is limited to 4 simultaneous threads (idem for ASSET and for agent)  with a server license, the model is limited to 4 simultaneous threads per token (idem for ASSET and for agent)  For example, if the user runs a set of predictions on ASSET (6 threads) and on two agents (2 threads by agent), the models will consume 4 licenses (2 for ASSET (4 threads + 2 threads)) and one for each agent).

Workstation machine configuration To configure the coordinator on the workstation machine, user has to click right on the 'Coordinator Tray Icon', to select 'Coordinator Options…' and to set all the information in the 'Agent' tab.  The 'Agent Log File Path' has to be a valid path on each client machine  The 'Agent Policy' has to be set to 'Server'


Page 15

Figure 7 : Distribution Coordinator Options

Client machine configuration On each client machine, after the installation of an ASSET agent (same version of ASSET as the one installed on workstation) and of MYRIAD Model (same version of MYRIAD Model as the one installed on workstation), user has to set the coordinator hostname and to give the name of the workstation ASSET.

Figure 8 : Coordinator Hostname

More information is available in the ASSET online help and/or in the ASSET documentation. Note: due to the MYRIAD Model licence system, it is not possible to run simultaneously two different versions in a multithreads/multi-machines context.

Page 16


CHAPTER 2

THE MYRIAD MODEL In This Section THE MYRIAD MODEL ........................................................................................... 17 CONTEXT ................................................................................................. ....................................... 18 THE MYRIAD MODEL ........................................................ .......................................................... 19 THE PROFILE COMPONENT....................................................... ............................................... 19

The principle behind the profile component ............................................................. .............. 19 Management of raster geographic data .................................................................................... 21 Profile extraction of raster geographic data ............................................................................. 21 Deygout's methods of calculating diffraction .......................................................................... 22 THE MORPHOLOGY DATA ......................................................... ............................................... 24

Improving propagation accuracy based on morphology....................................................... 24 THE FACETS COMPONENT......................................................... ............................................... 25

The principle behind the facets component ............................................................................. 25 The modeling and function of the facets component ............................................ ................. 26 THE GRAPHS COMPONENT ........................................................ ............................................... 27

The principle behind the graphs component ........................................................................... 27 The modeling and function of the graphs component ........................................................... 28 THE TRAIN COMPONENT ........................................................... ............................................... 30


Page 17

CONTEXT In the current context of the growth of mobile telecommunication networks, operators face innumerable technical and economic problems during the planning, deployment, and optimization phases of mobile radio networks. To address these issues, operators use dedicated engineering tools that offer solutions to problems such as the calculation of coverage, determining hand-over zones, interference, frequency allocation, antenna optimization, traffic localization, and so on. Among the most indispensable tools for mobile radio engineering are the field prediction models. Also called wave propagation models, they are incorporated in engineering tools and allow mobile radio network creators to determine the coverage zones of emitting stations and, in turn, infer the interference zones. Providing the indispensable information to numerous tools for further treatment and exploiting the predicted fields, they occupy a privileged place in cellular engineering. Strongly dependent on geographic data that allow the extraction of a vertical crosssection of the terrain (called a “profile”), a propagation model is a mathematical simulation of a physical phenomenon between two points. Performing preliminary processing of the cross-section of the terrain in order to check the calculation hypotheses, the propagation model combines theory with profile information before statistically adjusting the results using field measurements. Programmable according to different needs, a propagation model is developed with for objective to always respect significant constraints in domains such as: speed, precision, strength, and versatility.

Page 18


THE MYRIAD MODEL Capping several years of research in domains such as: propagation, modelling, optimization and algorithms, the MYRIAD propagation model is able to automatically adapt itself to all engineering (micro, mini, small, and macro cells), to all environments (dense urban, urban, suburban, mountainous, maritime, and open), and to all systems (DVB-H, GSM, GPRS, EDGE, UMTS, WiFi, WiMAX, LTE) in a frequency range starting from 200MHz to 5GHz. Relying on very realistic modelling of the channel, it achieves the synergy of the three physical elementary contributions: diffraction in the vertical plane, guided propagation in the horizontal plane, and reflection on the relief.

Reflection on mountainous areas

Diffraction (vertical plane)

Guided propagation (horizontal plane)

Figure 9: Reflection + diffraction + guided propagation = the MYRIAD Model trilogy

THE PROFILE COMPONENT The principle behind the profile component The profile component is divided in two parts:  The first one is dedicated to the environments for which the vectors‟ geographical data are not available throughout the whole calculation zone or for occasions where the user does not want to work with them.  The second one combines the vectors‟ geographic data (when available) and the raster geographic data representing the surface with raster geographic data of the relief. The calculation of the loss of propagation is almost entirely determined by the relief analysis in the vertical plane passing through the transmitter and the receiver. This hypothesis allows to assimilate each relief obstacle to a 2D theoretically thin and infinite horizontal plane, and to reduce it to a problem of wave diffraction calculation on a succession of thin ridges that can be treated with Fresnel formulas. The first operation, then, consists of elaborating the profile from the vectors‟ geographical data or raster geographic data representing the surface (clutter and/or building raster) and the raster geographic data of the relief (height). MYRIAD Model User Reference Guide MYRIAD v400; ASSET v8.0

Page 19

Figure 10: Profile example (without vectors)

Figure 11: Profile example (with vectors)

The solution chosen to construct the diffraction profile with the vectors‟ geographic data is to add the profile of the heights of the vectors to the altitude profile taken from the height as follows:  For a building contour, the diffraction profile consists of ridges of equal heights. All the height ridges supporting the buildings, modified so that the building‟s rooftop stays flat, are increased to the height of the building. The profile is then completed with the two ridges framing the building.  For a forest contour, the diffraction profile is formed from the height ridges under the forest contour to which the height of the contour is added. The profile is then completed with the two frame ridges of the contour where the height portion is determined by interpolation.

Figure 12: Profile construction for vectors

Once the profile is obtained so that the diffraction loss is not over-estimated, it is best to delete certain diffraction edges; therefore every diffraction edge less than one hundred meters away from a positive or negative diffraction edge is deleted. This Page 20


means that a succession of diffraction edges that are so close to each other that they in fact represent only one ridge is not taken into consideration. Deygout‟s method is chosen to calculate the diffraction loss. The model also calculates different variables related to the profile. The loss of the profile component results in a linear combination of these variables for which the coefficients are determined by adjusting the smallest squares with the calibration tool.

Management of raster geographic data Raster geographical data can be of 3 different types: Height (Digital terrain model): Description of altitudes above ground of the points at the centre of the pixel. This is a single point and not a calculated average on the various altitudes encountered on the pixel. Clutter (Digital surface model): Statistical description of the surface or principal theme on the pixel. Building raster (Digital elevation model): Descriptions of height above the surface of the points at the centre of the pixel.

The profile component adapts itself to all types of raster data. The management of raster geographic data is based on the construction of a “height” representation and a “surface” representation. A data representation is a zone of raster data that results from the fusion of various raster files (or from the data of one file if only one file is available) and whose construction unfolds according to the following stages:  Inventory of the files whose intersection with the calculation area is not zero  Creation of a result area and positioning of each pixel to “indefinite value”.  Reading every file and fusion in the result area. During the fusion, a pixel from the resulting area is only affected if its previous value was “indefinite value”. The “height” representation is a 2D matrix that contains the description of the relief while the “surface” representation is a 2D matrix that contains the surface height (for clutter data, each type is associated with a height by the user). The resolution of the “height” representation can be inferior, equal, or superior to that of the “clutter” representation.

Profile extraction of raster geographic data Taking into consideration the fact that propagation is principally done through diffraction on relief, the model constructs, in a vertical transmitter-receiver plane, a profile called “knife edge” comprised of ridges determined from the raster geographic data. For that, it extracts the profile ridges that are at the intersection points of the vertical planes (or the horizontal planes if α >=45) passing through the centre of the pixels with the transmitter-receiver segment. The height of each ridge is equal to the height of the bin containing the point of intersection. MYRIAD Model User Reference Guide MYRIAD v400; ASSET v8.0

Page 21

Pixels R

Y

Verticals and horizontals plane at the centre of the bins

Profile extract E

X

X

Z

Figure 13: Extraction of a profile

The profile is extracted by adding a value estimated by interpolating the distance starting from the surface profile to the value of each point of the height profile. At each point of the height profile, the point that provides the best frame in distance in the surface profile is chosen, and the height of the surface ridge is estimated with a linear interpolation starting from the heights and the distances from the transmitter of the “framed” ridges.

Deygout's methods of calculating diffraction The phenomenon of diffraction is one of the most important factors contributing to the propagation of electro-magnetic waves. Deygout‟s method has been chosen for the profile component. It uses three fundamental concepts to arrive at the calculation of diffraction losses: a) The first Fresnel zone: It is generally understood that diffraction phenomena due to all obstacles situated outside this zone are negligible. The first Fresnel zone is the volume limited by the ellipsoid with starting points E and R, such as EM + MR – ER ≤ /2 (with M points belonging to the Fresnel ellipsoid). elli psoid). b) The Huyghens principle: principle : The fundamental idea behind Deygout‟s method is to understand that, to go from transmitter to receiver, the wave must, by diffraction, avoid a limited number of obstacles (thin ridges) taking each one into consideration after the other in function to their importance regarding the method of the Fresnel calculation, that is to say in relation to the Fresnel zones defined in consequence.

Page 22


E

R

Figure 14: Profile example of obstacles in thin ridges

c) The superposition principle: As with other methods, the problem of diffraction on the multiple ridges is treated as a succession of diffractions on a single ridge, for which the Fresnel calculation is applicable. This approach is both empirical and intuitive.

P2

P1

P2

P0

P2

P1

P2

R

E 3

2

4

1

6

5

7

Figure 15: Order of obtaining the positive edges committed

The algorithm calculates recursively by sets of three (thus with a maximum of 15 edges of diffraction), between the transmitter and the receiver, the interference coefficient of each ridge in the Fresnel ellipsis. It conserves the ridges that have the largest interference coefficients, as well as the number of engaged ridges, and diffraction losses are calculated starting from these ridges. The loss represented by the profile component, created by the accumulation of the different diffraction losses, is then corrected by adding the weighting of the calculated variables along the profile whose coefficients are determined by adjusting the least squares method by the calibration cal ibration tool.


Page 23

THE MORPHOLOGY DATA Improving propagation accuracy based on morphology Morphological data is a representation of the ground specific to MYRIAD Model. Each geographic area is stored as a classified grid, in which each point is dependant on both the elevation and the clutter. The data point value is calculated in a continuous way: For example, a point representing “DENSE_URBAN_RELIEF” morphology cannot have a point representing “FOREST_FLAT” morphology as a neighbouring point. It is strongly recommended that you use data representing the morphology of the area with the MYRIAD Model in order to ensure accurate results, especially for the tuning mechanism. This data must be generated from both the clutter and elevation data. This can be done using the MYRIAD Model interface. There are ten different types of morphology:  FOREST_FLAT  FOREST_RELIEF  OPEN_FLAT  OPEN_RELIEF  SUBURBAN_FLAT  SUBURBAN_RELIEF  URBAN_FLAT  URBAN_RELIEF  DENSE_URBAN_FLAT  DENSE_URBAN_RELIEF

Page 24


THE FACETS COMPONENT The principle behind the facets component The facets component of the MYRIAD Model allows you to evaluate the contribution of the multi–path while the electro-magnetic connection takes place in an irregular terrain. It allows you to predict the signal strength received by the mobile taking into account the reflections produced on the mountain slopes. This improves the precision in relation to the classic components that only take into consideration the propagation phenomena in the vertical cross-section of the terrain between the transmitter and the receiver. The consideration of reflection is very complex and depends on numerous factors. It is suitable then to make simplified hypotheses so that the component is usable by the operator of radio mobile networks.  The first simplification is the consideration by the model of the paths comprising only the reflection on a mountain. It can be assumed that this hypothesis introduces a slight error because the received paths by multiple reflections on the topography are strongly mitigated and not significant at the receiver.  The second simplification is that only the zones on the mountain slopes that are in “height” visibility of both the transmitter and receiver are considered. Note that there is visibility when the different obstacles do not protrude from the axis of the Fresnel ellipsoid.  The third hypothesis is that the determination of the reflected paths is only performed if the direct transmitter-receiver path is obstructed by the terrain.  The fourth hypothesis is to limit the extent of the search area for possible reflectors by fixing a maximum delay for the consideration of the echoes. The measurement campaigns of impulse response have allowed validation to a value of 50μs. Because of calculation time, the value by default is fixed at 25μs, which corresponds to a maximum distance C .∆tmax = 300 x25 = 7500 m (where C equals the speed of light).  Finally, the fifth hypothesis is the understanding that there is a reflection, or diffusion, (as opposed to discrete reflection on plane surface) given the large irregularities of mountainous slopes in relation to wavelengths. The received signal is uniformly shone in the half-space above the slope of the mountain, and the diffused power is inconsistent. Nevertheless, by analogy with the radar equation, it is assumed that all the contributions of the diffused power could be added together by the receiver. This hypothesis has been verified on the terrain and explains why a large quantity of energy nonetheless reaches the mobile even though there are rarely discrete reflections in the direction of the mobile.


Page 25

The modeling and function of the facets component Activating the facets component is only effective if the propagation context is hilly enough. The search of reflection zones is performed using the facets index parameter. The files generated with the aid of an algorithm based on the Delaunay triangulation method are defined in this parameter. These files contain the information about the modelling of the relief (reflection facets, visibility in relation to the reflection facets, etc.) Once the user has obtained the reflection zones, the facets component calculates the loss of the reflected paths to determine the total strength received by the receiver. The profile loss is calculated and is added to the facets component loss to form the total loss of the MYRIAD Model.

Page 26


THE GRAPHS COMPONENT The principle behind the graphs component The graphs component is only effective for transmitters for which the vectors geographic data are available over the entire calculation zone. In a micro cellular context, the emitting and receiving antennas are positioned below the average level of the rooftops. In this case, the signal strength received by diffraction above the rooftops is significantly inferior to that of the signal strength received by propagation in the street and diffraction on the edges. The propagation hypothesis above the rooftops is therefore no longer sufficient, especially when the number of edges between the transmitter and receiver is not too great.

Figure 16: Profile example for the graphs component

When the antenna is well under the rooftops, the electro-magnetic propagation occurs principally along the streets which act as street canyons. There is a “guiding” phenomenon of the electro-magnetic waves by the street which implies that the major propagation phenomena are situated in the horizontal plane and no longer the vertical plane, as is the case with the profile models. Thus, we benefit from the canalization effect to increase propagation. To model this guiding phenomenon of the electro-magnetic waves by the streets, the model uses a graph of the street axes. A graph is formed by arcs representing the axes of the streets capable of acting as a wave guide.


Page 27

Figure 17: Example of a graph (PARIS)

Searching the graphs is done using the graphs index parameter in which are defined the files that contain the relative information used to model the route maintenance.

The modeling and function of the graphs component This component is responsible for calculating all the contributions coming from the receiver by the different paths illustrated by the graph. It calculates the loss for each path as a result of distance, of the diffraction on the vertical ridges formed by street corners, and of multiple reflections along the street axes. All the elementary contributions are then added to constitute the total loss of the graphs component representing the electro-magnetic energy that propagates along the streets. The number of paths going from the transmitter to the receiver is not defined. Only those paths for which the loss is less than a defined threshold, that is to say sufficiently significant, are selected.

Page 28


Transmitter Receiver Best path considered for the graphs component Paths considered by the graphs component Figure 18: Examples of paths for the graphs component

The profile loss is calculated and is added to the graphs component loss to form the total loss of the MYRIAD Model.


Page 29

THE TRAIN COMPONENT The train component provides the corrections of additional losses while the receiver is in a train. When this option is used, the losses calculated for receivers on a railway track or very close to one are stronger than if no train were found. These additional losses are highly dependent on the characteristics of the railway tracks (i.e., excavation, tunnel, embankment, viaduct, or soundproof). The losses are also different if the train is single, double-decker-low or double-decker-high. The Use railway tracks option must be selected for the train component to be taken into consideration.

Page 30


CHAPTER 3

CONFIGURING MYRIAD MODEL WITH ASSET It is very easy and fast to configure the MYRIAD Model in ASSET. When you add a new model (Configuration => Propagation Models… => Add), you have to choose the MYRIAD Model and the first specific window appears.

In This Section CONFIGURING MYRIAD MODEL WITH ASSET ............................................ 31 THE ‘PARAMETERS’ TAB ........................................................................................................... 32

General parameters ................................................................................. .................................... 32 Geographical data ............................................................... ......................................................... 33 Output directory .................................................................................................................... ...... 35 THE ‘TIME OPTIM’ TAB............................................................... ............................................... 36

Prediction parameters ........................................................................................................... ...... 36 Radial.............................................................. ................................................................. .............. 37 Inner/Outer area resolution........................................................ ............................................... 38 Optimized calculation area ......................................................... ............................................... 39 THE ‘RADIO’ TAB.......................................................................................................................... 40

Radio optimization ................................................................................................................ ...... 40 Typologies effects .................................................................................... .................................... 41 Depenetration........................................................... .............................................................. ...... 41 Building calculation........................................................... .......................................................... 42 Computed data ............................................................................. ............................................... 43 THE ‘ADVANCED’ TAB .......................................................................................................... ...... 47

Key features .............................................................. .............................................................. ...... 47 Train option ................................................................................................................. ................. 49


Page 31

THE ‘PARAMETERS’ TAB

Figure 19: MYRIAD Model - Parameters tab

In this window, you can check the MYRIAD Model parameters which are:

General parameters In this section, you can define the frequency and the mobile height.  The frequency, two possibilities: If you select the Use antenna frequency checkbox, the model uses the frequency defined for the antenna (except if the antenna is not defined or if the antenna frequency is set to 0) The default frequency is used when Use antenna frequency is not selected and/or when the antenna is not defined (unassociated measurements for example) 

The mobile height (1.7 meter by default)

Page 32


Geographical data In this section, you select the geographical data you want to use. There are four options:  Heights: this type of data is compulsory for the model.  Building raster  Clutter: If you select this type of data, the Clutter height assignment window appears:

Figure 20: MYRIAD Model - Clutter height assignment window

In this window, the list of clutter classes appears. You must determine whether the clutter is a Water type (this information is used for the maritime optimization), a Forest type (this information is used for the forest optimization) or an Undefined type (meaning there is no data). You also have to indicate the height information. In most cases, the clutters have a default value which is an example; it is recommended to check that each value corresponds to the project reality. Note: All the themes contained in the clutter file(s) have to be referenced in the clutter‟ description file (including “No_data”) For raster geographical data, you have other options

Figure 21 : MYRIAD Model - Raster options


Page 33

If you did not select the Use advanced accuracy checkbox, the model uses less detailed calculations in order to decrease processing time. Otherwise, the model exploits the accuracy of the geographical data with the goal of generating more exact output. If you have selected Use highest resolution checkbox, the model uses the highest resolution available. Otherwise, the model uses the resolution closest to the prediction resolution. If there is no surface data (neither vectors, nor clutter and nor building raster), you can set an average surface height using the Clutter height parameter. 

Vectors: If you select this type of data, the Vectors type assignment window appears:

Figure 22: MYRIAD Model - Vectors type assignment window

In this window, the list of vectors names appears. You must determine whether the vector is a BUILDING type, a FOREST type, a WATER type, a BRIDGE type or an UNDEFINED type. This configuration creates or modifies a file, named 'menu.xml', containing vectors type specifications in the same directory as the vectors index file. This feature use the attribute NAME of tag VECTOR in the " index.xml" file to determine the vector name. The vectors files which are under the same tag VECTOR must have a single type of vector. This feature also generates a file " indexOptim.xml" which optimizes the vectors access and so improves the calculation time (this file is not compulsory for the model). If MYRIAD Model encounters a vector description that is not defined in the menu.xml file while loading the vectors index file, you are informed of the issue and the UNDEFINED type is assigned to the corresponding geographic data file.

Page 34


If the menu.xml file contains non standard MYRIAD Model vector type, you are informed of the issue and geographic data file relating to that unknown vector type is assigned an UNDEFINED type. If the vectors index file was updated, you are informed of the issue and you have to update files "menu.xml" (and "indexOptim.xml") with the settings.

Output directory In this section, you have to select a writing directory (required by the model). It is used to store temporary files and specific propagation files which can be re-used by the model in multi sectors context (under certain conditions) to decrease the computation time. The number of temporary files is limited to 100. This directory is automatically emptied during the closing of the platform.


Page 35

THE ‘TIME OPTIM’ TAB On this tab, you can choose the calculation parameters (each of which has a default value so that you can run the model without defining any values).

Figure 23: MYRIAD Model - Time Optim tab

You can define the following MYRIAD Model parameters:

Prediction parameters In this section, you have the possibility to set the resolution of the calculation. If you select the radio button Use pathloss prediction manager resolution, the model uses the prediction resolution If you select the radio button Userdefined resolution, the model always uses the resolution defined in the associated field (100 meters in this example). Note: the calculation resolution of the model has to be superior to the one of the platform. The Circular calculation areas parameter allows you to indicate whether the model has to consider the calculation areas as circular areas.

Page 36


Radial The Use Radial mode parameter allows you to activate radial calculation. During a coverage calculation, this mode allows you to reuse a profile extracted on the perimeter of the calculation area, approximating for the distance. The radial mode is approximately three times as efficient. In fact, in a square calculation zone of n pixels long, in normal mode n² profiles are extracted, while in radial mode only 4n-4 profiles are extracted. Radial mode works as follows: - MYRIAD Model associates a profile at several points (depending of the resolution) of the calculation area perimeter, while determining the closest profile by orthogonal projection - MYRIAD Model extracts the first profile of the perimeter and uses it for each calculation point along the profile, and so on In the diagram below, the resolution corresponds to the transmitter resolution. Profile A

dA dB

Profile B

In this case: dA < dB so profile A is kept

Transmitter Extraction point of the profiles Profiles of the perimeter of the calculation area Calculation point Orthogonal projections Figure 24: MYRIAD Model - Radial mode

You can define the extraction resolution (in meters) of the profiles on the perimeter of the calculation area when you use radial mode. If you select Use calculation radial resolution, the model uses the calculation resolution If you select User-defined radial resolution is activated, the model uses resolution you define (100 meters in this example) Note: If the User-defined radial resolution is greater than the transmitter resolution, the transmitter resolution is used.


Page 37

Inner/Outer area resolution The Use second mesh size parameter lets you activate or deactivate the Inner/outer area resolution. The Distance Threshold parameter enables you to specify the transmitter-receiver distance (in meters, 10,000 m by default) starting from which, during a calculation, the model changes from a grid using the transmitter‟s resolution to a larger one defined according to the following method: If you select Use 2*calculation resolution, the model uses the double of the transmitter resolution If you select User-defined distant resolution, the model uses resolution you define (100 meters in this example). Note: If the User-defined distant resolution is less than the transmitter resolution, the transmitter resolution is used. The model takes into account two different resolutions, the transmitter resolution (R1) and the resolution assigned by the user (R2). The model checks whether the ”transmitter/receiver” distance is greater than distance threshold (the red circle in the diagram below). If so, it calculates the considered point and transfers the results in the neighbouring pixels. Otherwise, the model calculates each calculation point. For example, in the diagram below, R2 is four times greater than R1. Transmitter Effective calculation points Calculation points

Figure 25: MYRIAD Model - Inner/outer area resolution mode

Page 38


Optimized calculation area The Use optimized calculation area parameter allows you to determine whether the model reduces the calculation zone behind the antenna or not (reduction is done orthogonally at the azimuth). If so, you have to set the recoil distance (in meters) starting from which the model no longer calculates behind the transmitter.

If you select Automatic backwards distance, the model automatically adjusts the value of the recoil distance, starting from which the model no longer calculates behind the transmitter. Limited to 30% of the distance along the azimuth, it is used by the model as described in the following equation: Recoil distance = e

 20  log( DIST _ AZIM)  ADIR_ MIN   ln(10)   20  

Where: - Recoil distance: The maximal distance calculated behind the station at 30% of the distance along the azimuth - DIST_AZIM: The distance between the transmitter and the edge of the influence area according to the direction of the azimuth - ADIR_MIN: The smallest directional loss of the antenna behind the station.

Calculated pixels Non calculated pixels Boundary of the zone to be calculated Azimuth DIST_AZIM Recoil distance

Figure 26: MYRIAD Model - Optimized calculation area mode

If you select User-defined backwards distance, the model uses the defined recoil distance (50 meters in this example) starting from which it no longer calculates behind the transmitter. Note: This distance is limited to 30% of the distance along the azimuth


Page 39

THE ‘RADIO’ TAB

Figure 27: MYRIAD Model - Radio tab

You can define the following MYRIAD Model radio parameters:

Radio optimization The Loss correction parameter indicates the value of the correction (in dB, 0 by default) to add to the total loss of the model. For example, Loss correction = 3dB means 3dB more loss. Note: this parameter does not affect the tuning. The Additional extraction distance parameter indicates the additional extraction distance (in meters, 0 by default, the value must be contained in the interval [0-300]) of vectors above the receiver (to the right of the receiver on the cross-section of terrain). This parameter is only available if vectors are available. The Graphs extraction radius parameter indicates the distance (in meters, 1500 by default, the value must be contained in the interval [0-2000]) beyond which the graphs component is no longer used. This parameter is only available if graphs are available. The Free space in line of sight parameter allows computing free space attenuation for pixels that are in Line Of Sight of the transmitter without taking into account any model calibration. This can be useful when calculation resolution is low and especially when receivers are fixed like in Machine to Machine.

Page 40


The Calculate on water parameter allows you to indicate whether the model should calculate on water. The Horizontal propagation parameter allows you to activate or deactivate horizontal propagation. If activated, when the models finds an obstacle between the transmitter and the receiver, if the height of this obstacle is greater than the smallest apparent width of it, then the height is replaced by the apparent width on the profile. The purpose of this is to avoid taking into account the diffraction on the top of the obstacle in the calculations carried out when the electro-magnetic field is mainly diffracted on the side of the obstacle.

"Left" apparent w idth

R

E

"Right" apparent width

Figure 28: MYRIAD Model - Horizontal propagation

In the above example, if the height is greater than the ”right” apparent width (to the right of the Transmitter-Receiver axis when looking at the receiver), then this height is replaced by the “right” apparent width.

Typologies effects The Use clutter type parameter allows you, in case of forest/maritime optimization, to indicate to the model to retrieve the clutter classes‟ information in case of building raster The Forest optimization parameter allows you to activate the forest optimization. If activated, the vegetation is not consider as an obstacle but as a transmission phenomenon which will be associated with specifics losses properties Note: The using this parameter slightly increases calculation time. The Maritime optimization parameter allows you to activate maritime optimization. If activated, a statistic optimization will be applied in the case of propagation over water. Note: The using this parameter slightly increases calculation time.

Depenetration The Depenetration option allows you to activate the depenetration feature. MYRIAD Model User Reference Guide MYRIAD v400; ASSET v8.0

Page 41

Depenetration works as follows: for an antenna located in a building, the model applies a specific treatment to take into account the propagation inside this building. The model identifies several virtual transmitters and considers the one which generates the lowest losses.

Figure 29: MYRIAD Model - Depenetration

In addition to the propagation phenomena between the virtual transmitter and the receiver, the model estimates the losses from the real transmitter (RT) to the virtual transmitter (VT) with the following formula: Loss (from RT to VT) =

d * loss ( Inside  Inside)

  Maxh  3;0   3 *

loss ( Inside Outside)

3

Where: - d: the distance between the real transmitter and the virtual transmitter - h: the difference of height between the real transmitter and the virtual transmitter - loss(Inside->Outside): corresponds to the Inside Outside parameter which allows you to set the value in dB (the default is 10 dB) of the assigned additional loss to the transmitter inside a building while the signal passes through the building. - loss(Inside->Inside) : corresponds to the Inside Inside parameter which allows you to set the value in dB/m (the default is 0.66 dB/m) of the assigned additional loss to the transmitter inside a building, while the signal is propagated inside the building. Note: the „transmitter-receiver‟ angle is considered as the incidence angle

Building calculation The Do not calculate in building parameter allows you to deactivate the penetration feature. The Model-defined penetration parameter allows you to activate the penetration feature. If activated, the model calculates only the outdoor pixels in a first pass, and then it calculates indoor pixels that border on at least one pixel calculated during the previous pass. Among the bordering pixels, the model selects the one with the smallest loss and adds a correction to get the loss on the new pixels. It carries on until there are no more uncalculated pixels left. The correction is automatically calculated by the model according to the frequency, the penetration angle and the distance with regard to the outside of the building. Page 42


Outdoor calculation ixels Indoor calculation ixels second Indoor calculation ixels third Building

First pass

Second pass

Third pass

Figure 30: MYRIAD Model - Building penetration

The User-defined penetration parameter also allows you to activate penetration, but for penetration, you manually set the correction. The Outside Inside parameter allows you to set the value (in dB, 10 by default) of the additional loss to a pixel situated inside a building, while at least one of the eight surrounding pixels is found on the street. The Inside Inside parameter allows you to set the value (in dB/m, 0.66 by default) of the additional loss to a pixel situated inside a building, while all of the eight surrounding pixels is found inside a building. Note: Penetration calculation can appreciably increase calculation time for very high resolutions. The Mobile on top of buildings parameter allows you to enable MYRIAD Model to consider the height of the receiver from the top of the building (or from the ground if there is no building). No penetration is available with this parameter.

Computed data The Tuning index parameter allows you to access a specific optimization for the different components of the model. This option is only available if you have already tuned the model using survey data (See chapter for more information about the MYRIAD Model automatic tuning). If the parameter is not assigned, MYRIAD Model uses its default tuning. The Morphology index parameter allows you to access the morphology geographic data type. If the parameter is not assigned, the model uses the default morphology. To generate these data, you click on the Morphologies Index tool ( ) button. The morphologies generator creates the morphologies on the intersection of the height and the clutter data. The morphologies are used by the model to determine in which geographical context the prediction is performed, and depending on context, special optimizations are applied. You have to set the name of the index file that is generated and map each clutter class to one of 17 basic clutter classes. A morphologies file and morphologies index file are created. Note: All the themes contained in the clutter file(s) have to be referenced in the clutter classes‟ description file (including “No_data”)


Page 43

Figure 31: MYRIAD Model - Morphologies generator window

Themes used by MYRIAD Model:  No data: This must be used for each theme for which there is no information (it means area where no data is available)  Other (default value): This is a neutral theme similar to the open environment (area with little or no vegetation). This theme can be used with a special theme, such as Pylon, etc.  Dense Collective Building : This must be used for a high density of collective buildings. Areas within urban perimeter. The included dense urban areas with dense development where built-up features do not appear distinct from each other. It also includes built-up features of the downtown district with height below 40m. This is also the heart of the city.  Mean Collective Building : This must be used for a mean density of collective buildings. Areas within urban perimeter. The mean urban should have mean street density with no pattern, the major streets are visible, the built-up features appear distinct from each other. Some small vegetation could be included. Average height is below 40m.  Sparse High Building : This must be used for skyscrapers or towers. This theme is used for low construction density. Cluster of high towers or skyscrapers higher than 40m.  Mean high Building : This must be used for skyscrapers or towers. This theme is used for mean construction density. Cluster of grouped high towers or skyscrapers higher than 40m (more than 3 buildings)  Dense High Building : This must be used for skyscrapers or towers. This theme is used for high construction density. Cluster of grouped high towers or skyscrapers higher than 40m (more than 10 buildings)  Industrial : This theme must be used for industrial clutter. Area including buildings with large footprints (greater than or equal to 20m by 40m) with height below 20m, separated by streets wider than 20m.  Intra Urban Open: This theme must be used for open areas surrounded by urban areas. Small open land area with no vegetation surrounded by mean urban, dense urban or residential.

Page 44














 

Intra Urban Forest: This theme must be used for forest areas surrounded by urban areas. Any vegetation land in any urban environment. Golf courses, municipal parks, extensive cemeteries or recreational areas are included in this category. Dense Individual Building : This theme must be used for areas with small constructions, such as residential areas. Houses in suburban environment. Suburban density typically involves laid out street patterns in which streets are visible. Lots must be as small as 30m by 30m but are typically larger and without vegetation. Individual houses are frequently visible. Average height is below 15m. Mean Individual Building : This theme must be used for areas with small constructions, such as residential areas. Houses in suburban environment. Suburban density typically involves laid out street patterns in which streets are visible. Lots must be as small as 30mm by 30m but are typically larger and include vegetation cover. Individual houses are frequently visible. Average height is below 12m. Sparse Individual Building : This theme must be used for areas with small constructions, such as residential areas. Scattered housing in open or open in urban surrounding. This is also buildings in parks. Forest: This theme must be used for each area related to forest or high vegetation. Forested lands with closed tree canopy. No distinction is made between deciduous and coniferous. Open: This theme must be used for areas with no habitation or vegetation, such as agricultural fields, open areas, motorways, etc. Water: This theme must be used for rivers, lakes or canals Sea: This theme must be used for oceans and seas.

The Facets index parameter allows you to access to the facets geographic data type. If the parameter is not assigned, the model will calculate without taking this into consideration. To generate these data, you click on the associated Facets index tool ( ) button. The facets generator creates the facets on the project. These data are useful to help the model calculate the refection for mountainous areas. You set the name of the index file which will be generated. The facets generator automatically detects the mountainous areas and decides if it has to generate the facets in each case. Note: Depending to the size of the project area, the calculation time can be significant. By default, the Height index parameter contains the height index of the project. You have the possibility of using another height index to generate the facets on another region. If the project area is not hilly enough, the model will consider that the reflection component is not significant and the facets generator will display the message: “No mountainous environment detected. Facets will not be generated.”.


Page 45

Figure 32: MYRIAD Model - Facets generator window

The Graph index parameter allows you to access to the graphs geographic data type. If the parameter is not assigned, the model will calculate without taking this into consideration. To generate these data, you click on the Graph index tool ( ) button: The graphs generator creates the graphs on the project. These data are useful to help the model to compute 'guided' propagation for micro cells. You set the name of the index file to be generated. The graphs generator automatically detects the dense urban areas and decides if it has to generate the graphs in each case. Note: Depending to the size of the project area, the calculation time can be significant. By default, the Vectors index parameter contains the vectors index of the project. You have the possibility of using another vectors index to generate the graphs on another region.

Figure 33: MYRIAD Model - Graphs generator window

If the project area is not considers as a microcell environment, the model will consider that the guided propagation component is not significant and the graphs generator will display the message: “No microcell environment detected. Graphs will not be generated .”.

Page 46


THE ‘ADVANCED’ TAB

Figure 34: MYRIAD Model - Advanced tab

You can set the following MYRIAD Model advanced parameters:

Key features The Antenna correction option allows you to determine whether the model verifies the coordinates of the antenna in an urban environment. Antenna correction works as follows: For an antenna located in a building, the model calculates the new DX, DY or the height so that the antenna is on the exterior of the building (DX and DY correspond to the displacement in the direction of the azimuth according to the axis of the Xs and the Ys in the horizontal plane in relation to the transmitter). After that, you can move the antenna to the facade by applying DX and DY to the transmitter coordinates or on the rooftop (by modifying the transmitter‟s height). Roof distance

Roof distance Facade

Facade distance

Position of the station Correction of the station

Case of a possible difference in facade

Case of an impossible difference in facade

Figure 35: MYRIAD Model - Antenna correction


Page 47

A text file (which can be opened in a spreadsheet), called “AntennaCorrectionFile.txt”, summing up the different correction parameters and the obtained results, is generated in the Result directory (the correction will appear only for poorly located transmitters). You can make an automatic update of the antenna location (roof or facade correction). For that, you have to make an xml export: GSM case: the selected xml file is "GSM/AMPS Cell" UMTS case: the selected xml file is "UMTS-NodeB" CDMA2000 case: the selected xml file is "CDMA BS" Fixed WIMAX case: the selected xml file is "Fixed WIMAX Node" Mobile WIMAX case: the selected xml file is "Mobile WIMAX Node"     

Figure 36: Antenna correction - xml export

You have to copy these xml files in the same directory than AntennaCorrectionFile.txt file, and run the automatic antenna correction tool available in the MYRIAD Model installation directory. Example: Antenna_Correction_Tool.exe

Figure 37: Automatic antenna correction tool

You need to choose between Antenna Roof Correction or Antenna Facade Correction and to select the directory where AntennaCorrectionFile.txt and xml files are located. Finally, you have to make an xml import ("Replace" has to be selected to erase older location): GSM case: the selected xml file is GSM-CELL-LIST001.xml ("GSM/AMPS Cell") UMTS case: the selected xml file is NODEB-LIST001.xml ("UMTS-NodeB") CDMA2000 case: the selected xml file is CDMA-BS-LIST001.xml ("CDMA BS") Fixed WIMAX case: the selected xml file is WiMAX-NODE-LIST001.xml ("Fixed WIMAX Node") Mobile WIMAX case: the selected xml file is MOBILE-WiMAX-NODELIST001.xml ("Mobile WIMAX Node")    



Page 48


Figure 38: Antenna correction - xml import

Note: The use of this option implies that the result files of the model are not assigned; it means that there is no calculation; the model only creates the “AntennaCorrectionFile.txt” file. The Minimal distance between antenna and façade parameter allows you to set the minimum distance (in meters) between the antenna and the building facade that supports it. The Minimal distance between antenna and roof parameter allows you to set the minimum distance (in meters) between the height of the antenna and the roof of the building that supports it. The Result directory parameter allows you to select the directory where the file AntennaCorrectionFile.txt will be generated. If it is assigned to "NULL", the results files are generated in the directory 'tmp' under the MYRIAD Model output directory.

Train option The Use railway tracks parameter allows you to indicate whether the train option is used by the model. If you select this type of data, the Railway vectors assignment window appears:


Page 49

Figure 39 : MYRIAD Model - Railway vectors assignment window

In this window, the list of vectors names appears. You must determine whether the railway is a TRAIN type, a TRAIN_EXCAVATION type, a TRAIN_TUNNEL type, a TRAIN_SOUNDPROOF, a TRAIN_EMBANKMENT type, a TRAIN_ VIADUCT type or an UNDEFINED type. This configuration creates or modifies a file, named 'menu.xml', containing building and railway types‟ specifications in the same directory as the vectors index file. The railway tracks data is vector linear data with only 2 points. The railway files, which are in vector MapInfo format (TAB file), have to be registered in the same xml index file as the vectors files (index.xml). The type of railway is stored under the attribute NAME of tag VECTOR. The railways files which are under the same tag VECTOR must have a single type of railway. If MYRIAD Model encounters a vector name that is not defined in the menu.xml file while loading the vectors (or/and linear) index file, you are informed of the issue and the UNDEFINED type is assigned to the corresponding geographic data file. If the menu.xml file contains non standard MYRIAD Model vector type, you are informed of the issue and geographic data file relating to that unknown vector type is assigned an UNDEFINED type. If the vectors (or/and linear) index file was updated, you are informed of the issue and you have to update files " menu.xml" with the settings. The Calculate only along the railway tracks parameter defines whether the model calculates only along the railway tracks (other calculation points will not be calculated). The Train type parameter indicates the type of train on the railways: single-decker, double-decker-low or double-decker-high.

Page 50


The Track width parameter defines the area where the model will apply the train correction. For example, for 20 meters, the model will consider an area of 20 meters along the railway track (i.e., 10 meters on either side) and every calculation point in this zone will be considered as being on the railway track.


Page 51

Page 52


CHAPTER 4

MULTI-SECTORS OPTIMIZATION In the case of a multi-sectors site, the model is able of re-using the results of the first sector to decrease the computation time of the others sectors. This optimization, available under certain conditions (same position, same antenna height ...) is not active in micro-cellular (graphs utilisation) and mountainous (facets utilisation) contexts and is not compatible with the parameters 'Antenna correction' and 'Optimized calculation area'.


Page 53

Page 54


CHAPTER 5

MYRIAD MODEL TUNING By using Tuning, you can calibrate the model (the aim of MYRIAD Model Automatic Tuning is to reduce the effects of imperfections and to enable MYRIAD Model to be as close to real propagation measures as possible). Note: the number of surveys file is limited to 3000. First, you have to add the measurement header files (*.hd) into the "Measurements Toolbox" window under the Tools menu in ASSET.

Figure 40: Association loading box

There are 3 different modes for loading measurement headers:  Not associated  Associated, but header left unchanged  Associated and header changed For the two first modes, header data are used for tuning (including antenna frequency and 'Connector_Loss' value). For the 3rd mode, data of the platform are used, including antenna frequency. Connector_Loss is set to 0. Tuning uses the EiRP (or similar for other technologies bar GSM).


Page 55

Figure 41: Measurement toolbox

By using the 'Analyse' button, you can run a validation of the model along further measurements. By using 'Auto Tune' button, you can calibrate the model, with specific tuning parameters dialog.

Figure 42: MYRIAD Model - Auto Tune window



In the Files section, you have to select the index and the name of the optimization file. After the tuning, MYRIAD Model uses the index file that contains the optimization file.

Note: An index file can include one or several optimization files. For example, it is possible to have an optimization file for the northern part of a country and another optimization file for the Southern. The first optimization file that matches the calculation area will be taken into account. Page 56


The optimization file that you select should not exist, if it exists, you must make sure you do not use the file anymore. With respect the index file, if the selected file already exists, the new optimization file will be automatically placed at the beginning of the index. Configuration: MYRIAD Model allows you to check the index optimization files by clicking on the Index file tool ( ) button which makes it possible to check whether the different calibration files of the calibration index recover all the project area. On this window, you can visualize a summary of the index optimization files:  In blue: name and coordinates of the different optimization area of the index optimization file  In red: coordinates of the area(s) which is (are) not covered by the index optimization file

On the first column, you can select the area you want to visualize.

Figure 43: MYRIAD Model - Index Optim File Configuration

You can modify the index optimization file. The modifications can be made by changing the value of the coordinates of the calibration area on the grid (double click on the case of the grid that you want to modify) or by directly moving the calibration area on the picture. You can also change the order of the calibration files in the calibration index. You can select the line of the calibration file and move up or move down the selected calibration file with up and down button.


Page 57

In the Calibration type section, you can define the survey filtering and the calibration type: You select the Use automatic filtering checkbox to filter out surveys where inconsistent values have been found. These surveys can generate inconsistency in the tuning coefficients. Note: When this option is selected, the log displays filtered surveys and a number between 0 and 100 indicates the level of quality (i.e., a value close to 95 indicates that the survey points are of a good quality) During the tuning process, the calibration tool optimizes coefficients for each type of morphology. Note that if the morphology option is not selected, the model will only use the default morphology meaning a unique optimization for all environments. Before starting the tuning process, the calibration tool will separate the measurements points in different categories (boxes) according to the morphology they belong to. Please note that for both options, there is an initial checking based on min. number of points: a min of 1000 measurements points per box are required to start optimization otherwise the model uses default parameters for the box(es). You can choose between two types of tuning: 



When using the Standard calibration, the calibration tool performs specific tuning for each box without any control, it means that, for each box, it just tries to associate the variables calculated along the profile with the best coefficients to reduce the average and the standard deviation. Note: the calibration tool optimizes all parameters (provided enough points are available), which means the model can show very good stats against the set of measurements used, but will probably be less robust when using the same model in other type of environment. When using the Advanced calibration, the calibration tool performs specific calibration for each box but it first checks the distribution of each variable calculated along the profile, it means that the calibration tool will associate the variables calculated along the profile with the best coefficients to reduce the average and the standard deviation but only the variables considered as representative of the reality, the other variables will be associated to a default coefficient ('representative of the reality' means in adequacy with what the model learnt during all steps of its development) Note: the calibration tool optimizes certain parameters (provided enough points are available), which means the model can show very good stats, but according if the set of measurements used, the results can be a little less good than with the other option but it will be more robust when using the same model in other type of environment

Note: the transmitters and the receivers located inside a building are not considered during the MYRIAD Model‟s Analysis/Tuning process

Page 58


In the Calibration Area section, you can select the optimization area. After the tuning of MYRIAD Model, the model uses the new optimization for the stations located in this optimization area. Three cases are possible:  Surveys extent: the optimization area is formed by the rectangle including the whole of the measurement points  Project extent (default value): the optimization area is equal to the ASSET project area  User-defined extent: you define the optimization area manually Tuning the model starts after you click Start tuning . You can follow the progress of the tuning in the progress bar and in the message log. Depending on your choice, you can run several tunings and compare the differences performance displayed in the different Tuning performances tab. At the end of the tuning, you can view the global results, by morphology and by survey. You can export these result in a CSV or Excel file with the option of exporting the performance per pixel data. Finally, depending on the results, you can choose whether you want to keep this new MYRIAD Model tuning by clicking on Select (if there is more than one tuning) and Ok to apply it automatically to the next predictions.

Figure 44: CW Measurement Analysis - Tuning performances window

Warning: during the Analyse or the Auto Tune functions, the MYRIAD Model parameters are always taken into account, for example for the frequency: If you select the Use antenna frequency checkbox, the model uses the frequency defined for the antenna (database or header,


Page 59

depending on whether measurements are associated or not) The default frequency is used when Use antenna frequency is not selected and/or when the antenna is not defined (unassociated measurements for example)

Page 60


TABLE OF ILLUSTRATIONS Figure 1: Installation of the MYRIAD Model for ASSET .......................................................... ........ 8 Figure 2: Uninstalling the Myriad Model for ASSET..................................................... ................... 9 Figure 3 : LMTools - Service/License File tab ........................................................ ......................... 11 Figure 4 : LMTools - Borrowing tab ............................................................................................ ...... 13 Figure 5 : Environment Variables ....................................................... ............................................... 14 Figure 6 : New User Variable ............................................................................................................. 14 Figure 7 : Distribution Coordinator Options .......................................................... ......................... 16 Figure 8 : Coordinator Hostname ....................................................... ............................................... 16 Figure 9: Reflection + diffraction + guided propagation = the MYRIAD Model trilogy ........... 19 Figure 10: Profile example (without vectors) .......................................................... ......................... 20 Figure 11: Profile example (with vectors)................................................................ ......................... 20 Figure 12: Profile construction for vectors .............................................................. ......................... 20 Figure 13: Extraction of a profile ................................................................ ....................................... 22 Figure 14: Profile example of obstacles in thin ridges .............................................................. ...... 23 Figure 15: Order of obtaining the positive edges committed ........................................................ 23 Figure 16: Profile example for the graphs component ....................................... ............................ 27 Figure 17: Example of a graph (PARIS) ............................................................................................ 28 Figure 18: Examples of paths for the graphs component ............................................................... 29 Figure 19: MYRIAD Model - Parameters tab ................................................................................... 32 Figure 20: MYRIAD Model - Clutter height assignment window ................................................ 33 Figure 21 : MYRIAD Model - Raster options ................................................................................... 33 Figure 22: MYRIAD Model - Vectors type assignment window................................................... 34 Figure 23: MYRIAD Model - Time Optim tab ........................................................ ......................... 36 Figure 24: MYRIAD Model - Radial mode ....................................................................................... 37 Figure 25: MYRIAD Model - Inner/outer area resolution mode .................................................. 38 Figure 26: MYRIAD Model - Optimized calculation area mode ................................................... 39 Figure 27: MYRIAD Model - Radio tab ........................................................ .................................... 40 Figure 28: MYRIAD Model - Horizontal propagation.............................................................. ...... 41 Figure 29: MYRIAD Model - Depenetration .................................................................................... 42 Figure 30: MYRIAD Model - Building penetration ......................................................................... 43 Figure 31: MYRIAD Model - Morphologies generator window ................................................... 44 Figure 32: MYRIAD Model - Facets generator window ................................................................. 46 Figure 33: MYRIAD Model - Graphs generator window ............................................................... 46 Figure 34: MYRIAD Model - Advanced tab............................................................ ......................... 47 Figure 35: MYRIAD Model - Antenna correction .................................................. ......................... 47 Figure 36: Antenna correction - xml export .................................................................... ................. 48 Figure 37: Automatic antenna correction tool ........................................................ ......................... 48 Figure 38: Antenna correction - xml import............................................................ ......................... 49 Figure 39 : MYRIAD Model - Railway vectors assignment window............................................ 50 Figure 40: Association loading box ....................................................................... ............................ 55 Figure 41: Measurement toolbox ...................................................................................... ................. 56 Figure 42: MYRIAD Model - Auto Tune window............................................................. .............. 56 Figure 43: MYRIAD Model - Index Optim File Configuration...................................................... 57 Figure 44: CW Measurement Analysis - Tuning performances window ................... ................. 59


Page 61

MYRIAD MODEL User Reference Guide.pdf

Recommend Documents