Академический Документы
Профессиональный Документы
Культура Документы
Product Description
Copyright 2018 TEOCO Corporation - All rights reserved. No part of this work, which is protected by copyright, may be reproduced in any form or
by any means - graphic, electronic or mechanical, including photocopying, recording, taping or storage in an information retrieval system - without
the written permission of the copyright owner.
Mentor Product Description
Contents
1 Solution Overview ............................................................................................................................ 4
2 Advantages of Mentor ...................................................................................................................... 5
3 Key Features & Benefits .................................................................................................................. 6
3.1 A Powerful Optimization Engine based on Mobile Measurements ......................................... 6
3.1.1 Optimal Balancing & Utilization of Existing Infrastructure to Reduce Capacity and
Throughput Issues ........................................................................................................................... 6
3.2 Delivering Accurate Views and Analysis of the Network's Performance ................................ 6
3.2.1 A single point of access offering an integrative view at any time, from anywhere .............. 6
3.2.3 Virtual Drive Test (VDT) for prompt visualization of call flow .............................................. 7
3.3 Streamline & Automate the Identification and Resolution of Network Problems .................... 8
3.3.1 Scalable, fully redundant, distributed and configurable mediation engine .......................... 8
3.3.2 Powerful data model for more accuracy and efficiency ...................................................... 8
5.1.4 Overshooting and Undershooting UMTS & LTE E-Tilt Optimization ................................ 16
1 Solution Overview
The growing complexity of Networks, aligned with the exponential demands of customers for seamless
quality of service in the face of increasing competition, leave CSPs (Communication Service Providers)
battling against churn rates, decreasing ARPUs and skyrocketing CAPEX and OPEX.
The management of a radio network and its optimization to provide better services is the domain of
skilled and trained engineers whose work is endless. Besides, engaging with the constraints of the
network is a never ending challenge that is not expected to be easier with the arrival of new technologies
and the deployment of LTE systems.
TEOCO's Mentor solution provides a fully integrated tool allowing CSPs to face the challenges
of Radio Access Network, by maximizing multi-technology network (2G, 3G, 4G) quality,
capacity and coverage automatically, using mobile measurements collected from the network.
Mobile measurements are specialized Quality of Service reports generated by handsets as they interact
with the network, and indicating the subscribers’ experience. Mentor's methodology is based on these
measurements to deliver a comprehensive and accurate view of the network's performance & service
quality.
Endowed with sophisticated proprietary algorithms, the system automatically generates optimization
recommendations that improve radio network quality, capacity, and coverage, while monitoring network
experience and reporting on degradations. These recommendations help lower the amount of
calculations and activities to be manually performed by engineers and ultimately reduce time-to-problem
resolution and expenses thanks to a higher cell to engineer ratio.
Additionally, for faster visualization and localization of activities, the solution offers robust Geo-
location capacities.
Overall the key benefits and features of the solution are described at length in this document. They
include:
A powerful optimization engine based on mobile measurements, ensuring a continually
optimized network
The capability to visualize the network, through the geo-location tool and virtual drive tests.
Strong Network Analytics to obtain valuable knowledge for an enhanced user experience,
assisting the engineers in operating performance measurements and automating optimizations
reports and scripts.
Ability to streamline & automate the identification and resolution of customer-affecting problems,
with a scalable mediation engine and powerful data models. All this, integrated within one tool
that is fully extensible and multi-vendor.
TEOCO's Mentor transforms radio network data into valuable knowledge to achieve advanced
visualization, optimization and analytics focusing on network as well as customers.
2 Advantages of Mentor
TEOCO’s Mentor system has a proven track record of successfully accompanying its customers
through radio network optimization, and achieving their performance goals with reduced expenditures,
by providing a multi-vendor support to enable a single point of analysis and optimization of
heterogeneous networks.
The solution is already installed and functioning at dozens of operators (mainly Tier 1 and Tier
2), with hundreds/thousands of users.
The Mentor Optimization engine builds on advanced genetic algorithms to scan the multiple possible
configurations for each sector, and recommends a viable optimized configuration. The optimization
engine continuously identifies better configurations as it progresses.
Mentor delivers an effective solution for the periodic analysis and tuning of the cluster/wide
neighbor list quality, and antenna parameter optimization.
3.1.1 Optimal Balancing & Utilization of Existing Infrastructure to Reduce Capacity and
Throughput Issues
Advanced genetic algorithms and geo-positioning of mobile measurements yield optimized antenna
configurations that result in better balanced traffic among the radio network sectors, enabling the same
infrastructure to accommodate more traffic.
By using mobile measurements collected from the network, Mentor ensures the use of the latest and
ongoing traffic and user-experience information, and thus the possibility to run short optimization cycles,
week after week, assuring constant improvement in the network quality and adaptation to
changing traffic patterns. As such, the process of tedious and expensive prediction creation and
calibration become obsolete.
3.2.1 A single point of access offering an integrative view at any time, from anywhere
Mentor is equipped with groundbreaking optimization and visualization tools displaying geographical
information and user experience parameters, thus enabling operators and engineers to obtain a full
picture of the network and its activities from a single point, at any time, and from anywhere. Indeed,
users can access the system through their personal computers via rich client installations
and/or a Citrix terminal. Sectors, sites and antenna details are always available, as well as different
information layers and methods, including a geographical view of neighbor lists, cluster-wide analysis
information, radio network quality data and configuration reports.
TEOCO is the leading company in Geo-location solutions, and has proven its solution quality,
delivering accurate positioning for moving and stationary calls, long and short sessions, voice
& data calls, indoor/outdoor transactions. The solution is currently installed at over 25 different
operators, and has been successfully and accurately geo-locating data from hundreds of
RNCs/BSCs/MMEs on a daily basis.
3.2.3 Virtual Drive Test (VDT) for prompt visualization of call flow
The Virtual Drive Test map provides near real-time visualization of call flow events associated with
specific sectors and terminals, according to a specific time
Mentor’s applications include a highly intuitive web-based analysis and investigation tool designed for
easy user adoption, and ensuring that the information is available at any time, from anywhere.
The dashboards users can easily drill-down, up and sideways to any hierarchy level and
dimension, to quickly analyze problematic scenarios and identify best tactics, resolve dilemmas
and/or mine critical knowledge.
In addition, the analyses delivered are customizable and can be personalized for each user in order to
focus information for specific use cases, including correlated cross-domain overall analysis.
An add-on supported by Mentor, the dashboards are platform-independent and offer mobile exposure
to a variety of critical data.
Mentor introduces ready to use implementation scripts to fully automate the optimization reports and
plans. In order to streamline the optimization process and minimize the chance of human error through
manual data processing, Mentor automatically generates scripts for optimization
recommendation implementation which can be used to deploy the suggested changes into the
network easily and efficiently, without user intervention.
Mentor is equipped with a graphical user interface enabling engineers to promptly perform their
activities. A quick toolbar has been defined providing shortcuts for the activities defined below:
Viewing KPIs and other parameters as a graphical layer and color coding
Defining and saving sector selections
Displaying relationship of map entities
Pivoting analysis for advanced correlation and cross-analysis
Exporting reports to Excel
Statistics on sector and areas
Sector properties viewers on the fly, interactive with the map
Comparing different maps/tables/
Generating Templates
The Mediation server automates the data collection and processing of subscribers’ call traces, counters,
configurations.
By automating all data collection and processing, the Mentor mediation ensures that zero
engineering time is spent on data collection, processing and conversion.
The Mediation engine interfaces directly with the network, and supports gigantic data-traffic volume,
hundreds of distributed interfaces, multiple protocols, technologies, and formats.
The architecture and technology infrastructure supports the processing of tens of Terabytes of data
every day, managing, monitoring, and efficiently utilizing tens of CPUs and cores in parallel, in order to
enable rapid processing response and increased data availability with high system reliability
and availability. The Mediation server has proven its efficiency, interfacing with, and collecting raw
data from hundreds of RNCs/BSCs/MMEs in parallel.
During the modeling process, Mentor parses the vast amount of data available and samples the
information in order to build a statistical-significant data structure. Network modeling is performed on
multiple levels: carriers, subscribers, sessions, RABs, time of day, etc.…
Data is also coupled between the different sources: mobile terminal session logs and sector
performance KPIs statistics, for more accuracy and efficiency.
When using the system, the physical network environment (BSCs, RNCs, NodeBs, MME, eNodeBs
etc.) presented is automatically updated and encompasses the latest network configuration and
database models for analysis and optimization.
Users are neither required to perform any manual textual data collection and post processing,
nor do they need to directly access the OSS to retrieve files, define profiles, schedule collections
etc. All these tasks are automatically performed, managed, and monitored by Mentor.
Mentor makes use of standard, open and state-of-the-art infrastructure components enabling the easy
integration of functional modules into the network, prompt interface creation and adaptation and cost-
effective operation & maintenance.
4 Mentor Platform
The solution is deployed with a central system, and distributed data collection & parsing agents. It uses
database servers, who run and manage the applications, and handle the respective user-client
requests, while scheduling automation tasks. The distributed agents are responsible for collecting and
parsing RNCs/BSCs/MMEs-level logging through direct OSS connectivity or through Iub probes.
The central system, which is composed of the databases, the application and the front-end servers, is
responsible for aggregating all processed data models from all the RNCs/BSCs/MMEs, interfacing with
database servers, running and managing the application servers, handling respective user-client
requests, and schedule automation tasks.
The workflow defined by Mentor can be divided into three core steps:
System Flow:
Radio Network Environment Construction: Mentor supports the import of any physical
network configuration (i.e.: site locations, antenna profiles, azimuth, tilts…), thanks to its
compatibility with existing network planning and configuration databases.
Data Log Collection and Modeling: Mentor seamlessly and automatically collects network
mobile-measurement logs and performance data without manual intervention. The collected
data is stored in a centralized, shared storage area to be used by all Mentor users.
The Mentor data-modeling module analyzes the collected information and generates a compact,
efficient database, including all the required RF calculations to be used for analysis and
optimization purposes.
Radio Network Environment Update: Mentor's Automatic Updates feature can automatically
download network updates such as new models, switch dump and physical configuration
updates from the server, and are then copied to the user's computer, without requesting manual
intervention.
The Mediation servers offload any storage requirements from the OSS.
Thanks to the server's modular architecture, Mentor can accommodate the different needs of operators,
and perform the tasks in a costs effective manner while reducing time-to-value.
The agents are also responsible for the automatic collection of network configuration files, counter data,
and other types of information.
The mediation platform supports native formats of the major equipment vendors, and each collection &
parsing agent may be configured to collect, and model a pre-defined set of eNodeBs / RNCs / BSCs /
MMEs.
The system supports grouping of collection and parsing agents. They offer a scalable, highly
performant, fully redundant, distributed, and configurable parsing engine that is guaranteed to
maintain and develop support for new vendor releases. The solution supports vendor-specific
formats (counters, KPIs, call traces, configurations, etc.), 3rd party formats.
Front-End Servers
The Front-End server is the "central nervous system" of the entire solution. As such, it is deployed in
full hot redundancy with no single point of failure. It functions as the central control: monitoring and
administrating all collection and parsing agents. It also acts as the entry point for all user requests, and
schedules application tasks.
File Storage
The system makes use of massive data storage to offload external data from OSS and other network-
level storage. This feature minimizes the impact of deploying the system on the network resources, and
provides more independence to the system as it is decoupled. It also enables higher flexibility in
configuring data to be collected, maintained, and analyzed, and to dynamically or retroactively change
these definitions.
The mediation sub-system is fully configurable with regards to the data volume and the retention period.
The users can therefore easily define and change the duration of raw data storage, and its level of
detail.
Once the appropriate definitions are made via the Admin console, the configuration is automatically
propagated to all system agents, and data is automatically retrieved and processed. Using the Server
Admin console, system administrators have a comprehensive end-to-end detailed view of the data
collection, mediation, and processing, as well as many other server and user activities.
Following the mediation sub-system processing of the Call Traces (CT), Configuration Data (CM), and
Performance Management (PM) information, the data is aggregated into its proprietary database
models.
In addition to Oracle which has been the Mentor database for many years, Mentor can now also run on
Hadoop with a Cloudera Impala database. This approach delivers a number of advantages, especially
for larger systems:
Scalability - Ensures future support for the ever expanding 2G/3G/4G/5G RAN networks
including the largest Tier 1 networks
Application performance - Improved performance of queries for faster creation of maps and
histograms. Binned map queries are up to 40x faster
Reduced costs - Open-source software and commodity hardware results in reduced costs.
~30% saving on 3rd party HW and SW for a 100 RNC network system
Higher retention periods - Enables longer data retention periods at a much lower storage cost
giving engineers more insights into historical data when troubleshooting issues
The optimization applications support a multitude of automatic activities including: scrambling code/ PCI
conflict resolution, neighbor list optimizations, parameters, antenna power and tilt, and overshooting
cell identification and fixing. Each optimization module uses a specialized wizard to guide the engineers
in running the optimization, and supports reports and graphic layers to analyze the current and
optimized plan.
Many of the optimization features also provide automatic provisioning capabilities. For example, when
a new Neighbor List optimization is generated, the application will also automatically generate a
provisioning script for the specific network vendor, such as Ericsson, Nokia, Huawei, ZTE etc. These
scripts can be executed by the operation department directly through the OSS, without the need for
manual data entry.
The system allows its users to plan ahead for new sites or sectors. Indeed, using Mentor,
engineers can find and analyze areas with poor coverage, high traffic areas, etc., and plan a new site
on that location for improved coverage, traffic, or user experience.
Currently, Mentor supports automatic SC and NL planning for new UMTS sites and NL, RSI and PCI
planning for new LTE sites. Two methods are available to create and display planned sites:
using a predefined file
manually adding the sites or sectors directly on the Mentor GUI
The Neighbor List Summary report displays measurements for each sector, and Mentor generates a
complete list of the newly updated neighbor list which can be analyzed by the engineers.
Easy modification and tweaking of Mentor’s automatically generated plans, to reflect operational and
local constraints. These modifications are maintained as server-side constraints, and will be used in
future automatic optimization plans.
Mentor’s power and tilt optimization engines use small changes over a cluster of sites to generate
evident improvement in the overall cluster KPIs. Mentor analyzes network quality considering traffic,
coverage and capacity, balancing the network resources according to predefined goals set by the
engineers.
The optimization process is driven by a Wizard thanks to which it is possible to define the carrier for
optimization, timeframe of logs to use or exclude global constraints and optimization goals. Once the
wizard's steps are completed, Mentor generates an optimization calculation. The system evaluates the
outcome of each optimization calculation by generating a set of reports and layers for each calculation.
The reports compare the different key aspects of the network before and after the optimization: traffic
load, power distribution, handover factor changes, RSCP distribution, Ec/Io, pilot dominance etc. After
doing so, the tool selects the best possible solution. Mentor also displays a sector layer showing which
sector's power/tilts are changed and the direction of change.
When implementing the new power and tilt plan into the network, the engineer can use the automatic
script creation in Mentor. This reduces the risk of data-entry errors while speeding up the
implementation process.
The optimization identifies overshooting UMTS or LTE cells as well as undershooting sectors, and
suggests appropriate tilt changes to improve their RF footprint. The results are presented in tables,
reports and map layers divided into three types of reporting: before/after/change.
The optimization identifies scrambling code clashes, calculates and suggests a new scrambling code
planning with minimal impact on the surrounding cells. The optimization can be run to solve severe
scrambling code impacts or replan the distribution of scrambling codes using a broad selection or the
entire network. The results are presented in tables, reports, and map layers and include three types of
reporting: before/after/change.
The scrambling code plan is based on the physical array of cell sites and antennas, the neighbor list
definition, and various constraints and planning rules. Since Mentor v9.5, the algorithm has been
enhanced to include a neighbor-relation matrix that is calculated from the call traces.
The paging area optimization uses the actual network statistics of call traces and cell level KPIs as
input, and provides recommendations for Paging Area parameters including:
The optimization results are targeted to improve the balance of the paging load between different area
codes, reduce HO ping pongs and excessive HOs in the area code borders. The optimization uses the
existing codes, recommending moving cells from one code area to another.
The optimization identifies PCI clashes, calculates and suggests a new PCI planning with minimal
impact on the surrounding cells. The optimization can be run to solve severe PCI impacts or replan the
distribution of PCI using a broad selection or the entire network. The results are presented in tables,
reports, and map layers and include three types of reporting: before/after/change.
LTE RSI Optimization enables the detection of existing clashes in RSI (Root Sequence Index) and CS
(Cyclic Shift) combinations in the current allocation of these parameters in the network.
Mentor’s unique dynamic planning algorithm of the RSI allocation assigns the exact number of RSI
codes for each cell, according to the calculated or actual cell range, thereby reducing the number of
RSI repetitions in surrounding cells. Preventing RSI clashes will optimize network performance related
to RACH procedures in the call setup stage.
All identified sectors are highlighted on a special map layer and included in a report for further analysis.
Area code optimization (Location Area/Routing Area in UMTS and Tracking Area in LTE) enables
detection of potential problems in the current area code planning and prevents redundant area-update
messages. It specifically aims to identify isolated sectors within a foreign code area and recommends
a new area code for these sectors. Additionally, it identifies sectors on area code borders, which should
be investigated. All identified sectors are highlighted on a special map layer and included in a report for
further analysis.
Swapped sector detection is used to detect whether cables between the transmitters and the antennas
in a site were incorrectly installed. The resulting statistical data reflected in Mentor will show abnormal
behavior in the area to which the sector points, and coverage for an area that the sector is not designed
to point to. The analysis results include a list of candidate sectors that are suspected of being swapped
sectors. The results can be displayed as a report and a layer.
Mentor automatically generates scripts for optimization recommendation implementation, which can be
used to deploy the suggested changes into the network easily and efficiently, without user intervention.
Implementation scripts for power, neighbor list and PCI changes are created specifically, per vendor
technology.
Mentor monitors the provisioning of optimization changes into the network and provides a reconciliation
report, identifying any discrepancies from the original provisioning plan. As such, engineers can validate
that their optimization plan has been fully implemented, assuring its full impact on the network
Mentor offers a rich client. Each system delivers specific advantages and targets different tasks and
usage, relieving engineers from performing redundant tasks.
All analyses or combination of analyses performed by Mentor can be saved as templates for future use,
and the server can be scheduled to generate information about optimization, KPIs, Maps, or any other
topic selected by the engineers. Users can select whether email notifications are enabled for specific
analysis and optimization tasks.
KPI Analysis
KPI Analysis is quickly performed by Mentor, and accessible to its users, who can select which
indicators they want to represent.
The network analysis results are displayed by means of reports, graphic layers and positioning maps.
Mentor includes a predefined set of reports which analyze multiple network aspect, their relevant KPIs
and creates sector layers that are presented on the geographical map.
Users can easily change the selection of KPIs presented in each report, create new ones, change the
time and network element resolution, date/time filter values etc.
By combining multiple KPIs per sector it is easy to identify which ones need special attention.
Mentor reports are also available in MS Excel format, including graphs and additional tabs, providing
raw data on the sector levels and time-resolution level.
Mentor can generate a script that automatically resolves all UMTS/GSM discrepancies in the switch,
without the need to run an IRAT neighbor list plan. The UMTS-GSM discrepancy report, which is
generated after the network has been imported, displays all the discrepancies between the different
configuration databases.
For models including positioning, areas or selection can be analyzed by displaying the relevant RF
statistics for the busy hour. This can also be achieved with KPIs, thus adding the selected features
onto the report.
Figure 9: Example of Mobile Drop Statistics and associated map for roamers analysis
Performance Statistics:
Carrier Statistics
The Carrier Statistics report displays various statistics on the network, such as RSRP histogram, RSRQ,
etc.
Cell Footprint
The Cell Footprint report displays information per cell of the area coverage, number of bins, number of
events, etc. In addition, it contains a ratio that indicates the cell footprint score.
The dashboards constitute an intuitive visualization and troubleshooting solution to investigate network
and subscriber behavior for end-to-end insight & root cause solutions. They include configuration
analysis and audit dashboards delivering aggregation and easy drill down options to network element
configuration.
Easy to use, with a friendly GUI, the web based analytics dashboards enable users to drill down to
single network element or time in a single click, apply filters to customize their views. Overall, the tool
provides:
Mobility Dashboard: presenting country-wide views of Neighbor list quality, and identifying
worst network elements and geographical area.
Devices Dashboard: countrywide view of devices statistics, including KPIs and KPI distributions
per devices names, types and vendors, in the network: drop rate, drop count, throughput call
count. In the Device Count dashboard a user can view count statistics and the distribution of
distinct devices, per device name or device vendor, as well as a trend graph per date.
Call Fault Dashboard: based on network recording logs, it enables both high level reports and
detailed analysis for drop calls, establishment failures, low throughput, etc., in areas where
counters are not covered. The recorded logs include KPIs per device and identifying problematic
devices and subscriber's identifications, alongside IMSI, Model and various faults distributions…
KPI Analysis Dashboard: displaying multiple vendor KPIs, and allowing drill down on elements
and time, as well as performing utilization audits to classify cells. Alarm counts and severities
are also correlated to the KPI trends in this dashboard.
Configuration Analysis & Audit Dashboard: presenting all RAN parameters, daily
configuration history, changes (delta between configuration), and audit capacities with pre-
defined alarms based on operator rules. Audit rule statistics are also included. An audit statistics
summary table summarizes the audit success and failure statistics per technology, vendor,
RNC/TAC, date and node type (RNC, Site, Cell and Relation).
Overall the dashboards provide an intuitive, straightforward yet powerful and scalable analysis, enabling
its users to identify problems in leading indicators, metrics and KPIs. Engineers can explore data on
their own, and “pull” answers, rather than following predefined reports.
Deployed over Mentor's server, the dashboards include all the capacities of Mentor. The dashboards
are sold separately. For more information, please get in touch with your TEOCO contact point.
Mentor delivers a state-of-the-art, distributed and parallel geo-location engine platform that geo-locates
all calls—moving and stationary, indoor and outdoor, short (SMS) and long calls, data, voice,
individual call events, and full call routes.
The Geo-location module uses advanced positioning techniques that consider a multitude of data within
the call and unique algorithms to generate geographical maps per bin to simplify the identification of
network hotspots for traffic, quality degradation, potential bottlenecks, and calculate as accurately as
possible locations for call events, and populate it in the relevant call database.
The solution uses call-trace data in the native vendor-specific formats in order to geo-locate individual
events within calls. The individual events geo-location process is aggregated into a call geo-
mapping including aggregated cells and Area-level geo-maps displaying a multitude of RF
parameters and events. Similar Geo-location capabilities are also available for probe-based
interfaces.
The resulting data model is then used through the specialized different applications by engineers,
managers, and customer support to identify network-traffic hotspots, network and subscriber level
quality degradation, potential bottlenecks, equipment type problems, VIP subscribers’ service level,
indoor coverage level…
Some of the derived maps (e.g. Ec/Io and RSCP) and RF data models are used in the optimization
process for the evaluation of different configurations and their impact on the network.
Based on the data generated by the Geo-location sub-system and stored with the data models, the
application sub-system generates all required mapping outputs in any necessary granularity and
resolution (IMSI//Network/UE type/time of day, etc.).
Map accuracy: TEOCO provides the highest accuracy on the market today.
A thorough engine: The Geo-location engine calculates locations based on all available and
relevant messages for positioning, (one-way, two-way, and three-way) to display reliable traffic
coverage and RF conditions, and full subscriber data. TEOCOs Geo-location is not limited only
to handover areas or areas where accuracy is the most optimal. In addition, the Geo-location
algorithms map both stationary and moving traffic as well as short (SMS) and long calls, and do
not require additional external information (e.g., mobile speed, height, or calibrated predictions).
Multi-vendor/multi-technology: The solution supports all major equipment vendors and
technologies.
Scalability: The solution provides full control of every aspect of the system deployment,
including the parameters that control the Geo-location scalability. Additional Geo-location
engines can be seamlessly integrated; aggregation parameters can be set, as well as filtering
event types.
Positioning over transportation layers: The Mentor geo-location algorithm is enriched with
data from Open Street Map (OSM) transportation layers, which enables unique and improved
techniques for geo-locating moving calls carried over roads, highways, and railways.
Positioning over topographic layers: The Mentor geo-location algorithm is enriched with data
from OSM topography layers that are used to improve the geo-location of calls carried in
buildings and parks and on rivers.
Positioning on low, medium and high building floors: The Mentor geo-location algorithm is
enriched with OSM building height data, enabling unique techniques based on machine learning
principals for geo-locating calls with heights. Indoor calls analysis is provided as classification to
lower, medium and upper floors. The user can select floor classifications in the Location Profile
filters. More accurate ground level traffic maps can be created which are used for network
planning (including small cells), optimization and analysis.
Positioning in underground transportation. Mentor’s geo-location algorithm is now enriched
with data from the OSM underground transportation layers, enabling unique techniques for
automatically geo-locating underground calls and differentiating between different traffic layers
for analysis. Calls carried in the underground tunnels are provided to the user in the level of
roads, subways and railways. The user can select each one of them in the Location Profile filters.
LTE MDT support. The Mentor geo-location algorithm is enriched with MDT data, which
enables the usage of unique techniques based on machine learning principals for better geo-
location of calls. This is done by using a combination of the standard events data required for
geolocation together with the MDT data. MDT data is used not only for improving the geolocation
of calls that include MDT data, but also of calls that don’t include MDT data.
Mentor uses advanced positioning techniques to generate geographical maps partitioned in small bins.
Each bin is colored according to the KPI value which reflects users' network experience within this bin.
This simplifies the identification of network hotspots for traffic, quality degradation, potential bottlenecks
etc.
Mentor supports multiple map types for analysis and troubleshooting purposes, including, but not limited
to the following:
GSM
Performance Maps
Downlink RxLevel Best Server
Downlink RxQual Best Server
GSM MS Power
Uplink RxLevel Best Server
Uplink RxQual Best Server
Traffic Maps
GSM Traffic
GSM MS Power
GSM Best Server
GSM Candidate Sectors
The GSM Event Viewer for virtual drive test call queries includes the L3-event call flow. The
events include call flow events such as Measurement Reports, Channel Busy, Clear Request,
Handover Command, and more. The user can double-click each event to view the details of
the event.
UMTS
Traffic Maps
Busy Hour Number of Calls
Busy Hour Minutes of Use
Busy Hour Total Number of Users
Busy Hour Traffic
Total Traffic
Number of Distinct Users
Performance Maps
Downlink Throughput
Downlink Throughput Peak
Downlink Volume
Ec/Io Best Server
Ec/Io Coverage
Ec/Io Coverage Peak
IRAT Density
Number of Polluting Pilots
Penetration Loss
Pilot Dominance
Pilot Pollution
RRC Setup Time
RSCP Best Server
RSCP Coverage
RSCP Coverage Peak
Soft HO Factor
Softer HO Factor
UE TX Power
Uplink Throughput
Uplink Throughput Peak
Uplink Volume
Drop Maps
Best server cell name
Strongest Leg
Arc Maps
Ec/Io
Fixed Distance Busy Hour Traffic
Fixed Distance Drop Call
Fixed Percentage Busy Hour Traffic
Pilot Dominance
Pilot Pollution
Soft HO Factor
Softer HO Factor
Other Maps
Combined Maps
Virtual Drive Test Maps
Mobile Drop Maps
Access Failure Maps
Mobile IRAT Maps
LTE
Traffic Maps
DL Traffic (Throughput)
Number of Distinct Users
Number of Sessions
UL Traffic (Throughput)
Performance Maps
CQI Index
Downlink Throughput
User Downlink Throughput
Downlink Volume
IRAT Density
Overlapping
RRC Setup Time
PUSCH UL SINR
RSRP Best Server
RSRP Best Peak
RSRP Coverage
RSRQ Coverage
RSRQ Best Server
RSRQ Best Peak
UL Power Restriction Rate
Uplink Throughput
User Uplink Throughput
Uplink Volume
Power Head Room
HO Delay Map
Arc Maps
Traffic by Distance
Average RSRP by Distance
Average RSRQ by Distance
Overlapping by Distance
Number of Drops by Distance
Downlink Volume by Distance
Uplink Volume by Distance
Fixed Percentage of Downlink Volume
Fixed Percentage of Uplink Volume
Number of IRAT Events by Distance
- Best Server
- Serving Cell Dominance
Multi RAT
Performance Maps
Best Technology (Throughput)
Downlink Volume
LTE Volume Ratio
Figure 15: Example of a Downlink Throughput combined map for UMTS and LTE
Each map can be generated on a sector selection or area selection. While generating a map for an area
selection, the user is able to display and analyze the serving sector on that area to solve pilot pollution,
overshooting sectors, high drop sectors etc.
Mentor also supports identification and notification of multi-eRAB states, where the related analysis
such as binned map and event map generation or drop release eRABs can be generated over multi-
eRAB states.
Event Maps
Mobile Drop
Access Failure
Mobile IRAT
CSFB (Circuit Switched Fall Back)
S1 Handover Event Map
X2 Handover Event Map
UMTS Handover Event Maps for 1A, 1B, 1C, 1D, 2D, 3A and 6D
UMTS Registration Failure
Customized Event Maps - Mentor also supports customized event maps which can be defined
for any 3GPP event
For quicker drill-down investigations, Mentor links events in event maps to their corresponding VDT L3
call flow events without having to run a separate VDT query. The user can right-click any event and drill
down to the call L3 signaling events of the selected event. VDT queries covering GSM, UMTS and LTE
technologies together are supported.
Performance, Drop, and Traffic maps can now be created as either binned maps or as sector-level
maps. In sector-level maps, each sector is colored according to its specific KPI value. Event maps and
computed maps are also supported on a sector level.
The Computed Maps feature enables you to carry out various mathematical operations on the values
of up to 8 binned map layers or sector level maps in order to assess the effectiveness of optimization
or to carry out detailed performance investigations.
The analyses are carried out on the value of each bin in the map, and the results are added as a layer
under the GIS category, in the Layers pane. Computed maps can be created during the network
analysis or generated at any time from the toolbar or by right-clicking an existing binned or combined
map.
Virtual Drive Test can be viewed and analyzed on the map area, in the Event Viewer, Detailed Table,
and in Google Earth.
Figure 17: Detailed list and data of call flow event - For each point on the route, the serving sectors as well as the
Detailed Table information can be seen.
The Threshold-Crossing Alarms (TCA) module enables users to view sectors with KPI values that
violate pre-defined rules. This enables the user to focus on the problematic cells in the network. The
rich threshold-crossing rules can be defined via the web interface of TEOCO's TrafficGuard.
The Mentor server generates KPI data in hourly resolution. TrafficGuard analyzes the KPI data and
based on the user-defined rules, creates alarms that are saved for a predefined retention period.
In Mentor, the user can view all alarms on the selected sectors within a configurable time period. Alarm
information may be viewed both in the UI and as an Excel report. The user can view the number of
alarms generated for each sector. For each rule, the user can also view a layer that shows the maximal
severity of all alarms generated for it.
Note: The Threshold Crossing Alarms (TCAs) module requires the purchase of an additional license.
The Automatic Problem Detection (APD) module enables automatic detection of geo-graphical
polygons with problems such as power shortage or call faults, using rules made up of a combination of
cell KPIs and geo-located information. In some cases, the related cause of the problem is assigned to
the specific problem, such as quality or coverage issues. The results are visualized in a GIS polygon
map layer and an enriched tabular viewer, and can be further investigated using Mentor’s analysis
features. The detected problems are assigned a severity level to enable the user to focus on the most
severe service-affecting cases first.
Network Planning
ASSET is the market-leading radio network planning suite providing propagation modeling, traffic
modeling and neighbor planning to facilitate the most demanding planning needs of today and
tomorrow’s multi-vendor, multi-operator, multi-technology mobile networks. The ASSET suite also
includes automatic cell planning, transmission network planning, and network capacity planning.
RAN Optimization
Mentor is an optimization and analytics solution leveraging robust algorithms in multi Radio Access
Technologies (RAT) Networks (LTE, UMTS, GSM) for maximal network and subscriber intelligence.
Mentor supports geo-location-based multi-RAT analyses across numerous dimensions: devices,
subscribers, networks and services.
Forte enables operators to maximize GSM/GPRS/EDGE networks quality, capacity, and coverage
automatically, as well as perform spectrum refarming allowing the redeployment of spectrum from
legacy technologies to new ones, while maintaining optimal performance.
Services
TEOCO’s RAN solutions team delivers packaged service solutions, a powerful combination of tools
methodology and people, leveraging TEOCO software solutions and telecom expertise. Our success is
based on a number of key factors; unbiased independent advice, our extensive tools portfolio, global
experience, local viewpoints, and the fact we recognize the unique aspects of each network. This
ensures each customer stays relevant to their market, enhances network efficiency, maximizes ROI
and delivers real value to its customers.
“TEOCO’s market leading position in Engineering Systems reflects the success of its extensive network
planning and optimization solution set, built by some very cohesive acquisitions and integrated
effectively to form a leading portfolio” Dean Ramsay – Senior Analyst, Analysys Mason’s Telecoms
Software