Prs Makes Network O&M Less Art Than Science: Textstart

Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 3

PRS makes network O&M less art than science

TextStart
By Chen Yuhan & Song Ruhao

Huawei PRS is an end-to-end visibility solution for mobile broadband (MBB)


networks that makes the O&M process far simpler, both in terms of the steps involved
and the relative skill needed to carry them out.
Mobile broadband (MBB) services are bringing new challenges to network operation
and maintenance (O&M). Structural complexities are making mobile network
performance increasingly difficult, and Huawei offers a solution in the form of its
performance report system – an end-to-end (E2E) visibility platform that spans from
the radio access network (RAN) to the packet-switched (PS) core network, enabling
comprehensive analysis of network resources & performance and displaying the
results in a user-friendly format, making for a vastly more efficient O&M process,
while improving investment efficiency.
Visualized network performance
Adding multiple modes/bands to any network will increase its complexity, but
Huawei PRS can visualize and simplify the performance analysis process. Based on
predefined KPI monitoring and analysis parameters for the GSM, UMTS, and LTE
technologies, PRS detects network performance faults, generates KPI alarms
automatically, and facilitates troubleshooting through KPI trend, high-priority cell,
and fault cause analyses; all help operators to enhance network performance.
Visualized network resources
PRS can illustrate the traffic generated by smartphones and facilitate MBB network
planning in detail. According to statistics published by comScore (a digital business
analytics firm), in December 2012, the smartphone penetration rate in five European
countries (Britain, France, Germany, Spain, and Italy) hit 55%, a 13% increase from
2011. Smartphones greatly differ from their predecessors in terms of traffic and
signaling, and traditional traffic models no longer meet the needs of the modern MBB
network. Huawei PRS visualizes smartphone resource consumption and displays
traffic and signaling accordingly, greatly facilitating the network planning process.
MBB networks are constantly changing and detailed network planning in the initial
stages simply cannot prevent this. PRS detects emergent bottlenecks, while
facilitating timely MBB expansion. It also provides an integrated network capacity
KPI system, while separately illustrating capacity trends and other statistics through a
capacity map or capacity panel, depending on user preferences.
MBB network capacity is finite; high-value services should be prioritized. Huawei
PRS supports in-depth analysis of bandwidth resource distribution, displaying
bandwidth consumed by various services, helping operators differentiate service
bandwidth management and monitor bandwidth policies so that they can be tweaked
as needed.
Successful cases
As of January 2013, Huawei PRS has been in commercial use in more than 70
markets, serving over 100 operators.
KPI insight enhances network performance
A certain Chinese operator (Operator X) deployed a radio network with gear sourced
from multiple vendors, and was facing difficulties in network performance
enhancement as well as O&M efficiency. Each vendor’s systems had thousands of
performance indicators, and engineers needed to spend a lot of time defining the
relevant network KPIs and tracking down the faults when one went wrong. A large
number of problems had to be analyzed by experts, making for poor O&M efficiency.
Operator X chose to simplify its O&M process through Huawei PRS, which provides
a KPI monitoring and analysis system that automatically monitors network
performance based on different KPIs and provides root cause analysis reporting to
speed up the troubleshooting process. Through analysis of KPI trends, high-priority
cells, and fault causes, Operator X has improved its O&M efficiency significantly.
Where it once took one or two hours to track down the cause of a low GSM access
success rate, it now takes five minutes at most (a twelve-fold improvement).
What’s more, the KPI insight function integrates Huawei’s expertise in network
performance analysis. Performance faults are classified by scenario, with a standard
fault location process given for each. Thanks to PRS, a less experienced engineer can
now carry out the analysis that was once done by an expert.
Smartphone & service resource visibility
Operator Y in Southeast Asia provides GSM, UMTS and LTE services, and the
smartphone penetration rate among its customers exceeds 40%. Smartphones not only
bring excessive data flow but also a large amount of signaling traffic, and this stems
from a 15-fold increase in network interactions when compared with traditional
terminals, making for a great impact on network operations.
With Huawei PRS, Operator Y now has a view of the data and signaling traffic
generated by smartphones and can produce traffic models for each type. Operator Y
can also use other statistics such as smartphone penetration rate and user growth rates,
together with data growth trends and signaling traffic, to ensure cost-effective
network upgrade.
Network resource usage planning is another concern for Operator Y. Various data
services run through its networks, and yet the bandwidth they consume may not be
proportional to the revenue they contribute. PRS helps Operator Y visualize
bandwidth utilization for each type of service, and the operator has found that
low-value point-to-point (P2P) services occupy more than 40% of bandwidth during
busy hours, so measures must be taken to minimize this impact. With PRS, Operator
Y can also learn about how network resources and customers are affected by its traffic
management policies, so it can adjust them accordingly.
Identifying capacity bottlenecks
As services and smartphones multiply, Operator Z (also in Southeast Asia) has seen
dynamic changes to its network load, such as those seen with the iPhone 5 package it
recently launched.
Operator Z has tried to cope through weekly analysis of the RNC board CPU usage,
but this process is time consuming. First, network performance data must be exported
from each network management system (NMS) and then imported into a
preconfigured spreadsheet so that KPI curves are generated and overloaded boards are
identified. If the signal processing unit (SPU) is overloaded, the base station that
manages it must be identified through the NMS, with its load distributed among other
SPUs manually. Completion of an overall network analysis usually takes a day or
more due to a heavy manual workload. Certain key KPIs, such as Iub interface
utilization, cannot even be evaluated directly, as both network performance and
configuration data are needed. In this way, capacity expansion happens only after
network performance is affected.
Operator Z uses PRS to visualize its network capacity usage, as it provides a complete
KPI system based on network performance, configuration, and licensing, including
statistics for interface utilization, board CPU usage, license utilization, and cell power
utilization. PRS also automatically filters out capacity bottlenecks from alarm
messages/logs, and offers capacity display, capacity trends, and capacity maps for
further analysis, helping Operator Z effectively expand network capacity.
Huawei PRS takes less than a minute to filter out capacity bottlenecks, at an accuracy
rate of more than 98%. If SPUs are overloaded, PRS can identify base stations where
SPUs are either overloaded or idle, enabling Operator Z to determine resource
allocation in less than ten minutes. The operator can also view the Iub interface
utilization rate, and enhance transmission resource usage accordingly.
In 2012, a Formula One race was held in Operator Z’s home market, attracting a large
number of visitors from abroad. During the event itself, network traffic increased
substantially, but PRS successfully identified three network capacity bottlenecks in
the immediate racing area, and Operator Z was able to adjust accordingly, ensuring
smooth communication and news broadcast during the event.
TextEnd

You might also like