Quel produit utilisez-vous actuellement ?

Managing the SIP Network

Monday, March 30, 2015 - 12:45
Managing the SIP Network

As deployment of Session Initiation Protocol (SIP)-based systems has increased, so has the need to manage these complex systems. As such, we will be reviewing the different facets of network management and how they apply to the challenge of managing a multimedia network.

Network Management for Multimedia Networks

Network performance is often an afterthought for most companies until something goes wrong.  When a network failure occurs, network management and performance then becomes a higher priority.  So, to assist companies with network management, the International Standards Organization (ISO) developed a management framework in 1989 to integrate with the Open Systems Interconnection (OSI) Reference Model.  

This model divides network management tasks into 5 specific management functions and defines each accordingly:

  1. Fault Management – encompasses fault detection, isolation, and correction of abnormal OSI environment operation.
  2. Accounting Management – enables charges to be established for the use of resources in the OSI environment, as well as user resource costs identified.
  3. Configuration Management – identifies, exercises control over, collects data from, and provides data to open systems for the purposes of preparing for, initializing, starting, providing for the continuous operation of, and terminating interconnection services.
  4. Performance Management – enables the behavior of resources in the OSI environment and the effectiveness of communication activities to be evaluated.
  5. Security Management – supports the application of security policies via functions which include the creation, deletion, control of security services and mechanisms, and the distribution of security-relevant information and reporting of security-relevant events.

To optimize the network in general, and specifically for multimedia devices, enterprise managers should evaluate the status of their network based on the 5 functional areas above and determine whether modifications should be made.

Quality of Service

An important part of network management is ensuring end users are satisfied with the way the network is operating, and the overall quality of service it provides.  Reviewing where the concept of network quality came from, and what modifications may need to be made in regards to quality of service, is especially crucial when dealing with more complex multimedia environments and the needs unique to real-time applications.

The original benchmark for quality in the telecommunications industry began with AT&T.  When AT&T monopolized the telecommunications market, one of its main performance standards was known as the Five Nines of Reliability.  At that time, the quality objective for central office switching systems specified 2 hours of downtime in 40 years of operation.  Working through all the time conversions, you arrive at what we know today as 99.999% reliability (the five 9s) factor.

Defining Quality of Service (QoS)

The International Telecommunications Union – Telecommunications Standards Sector (ITU-T)defines the subjective term Quality of Service (QoS) as: the collective effect of service performance which determines the degree of satisfaction of a user of the service.  The operative words here are end-user and satisfied.  In other words, the end users’ performance criteria for the multimedia network must be met.

As a result, much research has gone into the subject of QoS resulting in a number of algorithms and protocols that have been developed to address and meet user satisfaction for quality issues.

Measuring Toll Quality

So what is toll quality for voice systems and how is it measured?  

The ITU-T has addressed these questions in two key recommendations:

1. Methods of Subjective Determination of Transmission Quality (P.800)

  • Consists of the Conversation Opinion Test – volunteers are asked to render an opinion of the connection used based on a 5-point scale, ranging from 1 (bad) – 5 (excellent).
  • A large number of test subjects are used with an average taken of the individual scores = the Mean Opinion Score (MOS).
  • A MOS of 4 is considered a target toll quality within the telephone industry.

2. Perceptual Evaluation of Speech Quality (PESQ), an Objective Method for End-to-end Speech Quality Assessment or Narrow-band Telephone Networks and Speech Codecs (P.862)

  • Addresses the effects of filters, variable delay, and coding distortions
  • Used for both speech codec evaluation and end-to-end measurements
  • Implemented in hardware for greater accuracy
  • Using both QoS testing and evaluation methods can provide a bird’s eye view of the overall voice system quality.

Optimizing QoS on the WAN using MPLS

Multiprotocol Label Switching (MPLS) originated from Asynchronous Transfer Mode (ATM) in the mid-1990s, when the Internet Engineering Task Force (IETF) and others sought to integrate the high speed Layer 2 switching in ATM with the Internet Protocol (IP) Layer 3 routing technologies. A key reason for the interest in MPLS is how easily it can be applied and implemented over a wide variety of networking infrastructures, including ATM, Frame Relay, LANs, etc.  For instance, MPLS-capable routers called Label Switching Routers (LSRs), only have to examine the MPLS header and not the entire IP header, so the QoS mechanism can operate independently of the Network Layer protocol.

Impairments that Impact Voice Quality

In terms of factors that influence voice quality, it is important to remember that PSTN is circuit switched and connection-oriented, while VoIP networks are packet switched and connectionless. With the differences in infrastructure, varying QoS challenges can occur.  For example, several of the VoIP network QoS challenges come from using packet switched networks to perform functions that they weren’t originally designed to support, like voice transmission.  Because of the packet conversion that takes place between the technologies, there can be transmission impairments to the end-to-end network voice connection.  

Other items affecting voice QoS:

  • Packet Delay or Latency – the difference between when the signal is transmitted and when it is received
  • Packet Jitter – the variation in arrival rates between successive packets
  • Packet Loss – the measure of the number of packets from the original data stream that don’t find their way to the destination

Of these three main issues with voice QoS, packet delay is most often the focus.  In fact, the ITU-T published a benchmark for voice circuit delay that details accepted delay objectives.  In their recommendation, delays of 150-250 milliseconds per one-way transmission path are within acceptable standards.  Charts contained with the ITU-T’s G.114 standard are available to assist VoIP engineers with necessary design variable modifications, along with design tools that can assist with delay budget calculations.

Tools for VoIP Network Management

There are many challenges that correspond with converged networks.  Fortunately, there are several management tools that can be used to assist in identifying and solving these issues.

SNMP Network Management Systems

  • Run on a protocol suite developed by the IETF that first became popular in the 1990s
  • Two key elements
  • Manager (console) – oversees the network operations over UNIX or a similar platform
  • Agent – monitors the local operations and reports key data back to the manager
  • SNMP multimedia management solutions come from companies like: 

Protocol Analyzers

  • Used by network administrators and engineers
  • Attaches to the LAN/WAN and captures all data and voice packets travelling along the wire, analyzes the packets based on the protocol being used, then displays that information in various formats
  • Protocol analyzers can be found from companies like

Performance/Quality Monitors

  • Look at network operations independently from underlying protocols in use
  • Display high-level information providing a snapshot of network health containing statistics like Mean Opinion Score, packet delay, etc.
  • Performance monitors can be found from:

Optimizing SIP Trunking Network Solutions with West Unified Communications

West Unified Communications has been delivering Unified Communications (UC) services to clients over a next-generation MPLS network since 2000.  Its managed MPLS networking service called Maxxis is flexible, scalable, and reliable; it allows enterprises to replace separate data, video, and voice networks with one affordable, converged network.

Moreover, West UC utilizes industry best practices for optimum interoperability with a variety of IP-PBX systems, and maintains certified partner status with Cisco, ShoreTel, and Avaya to deploy SIP Trunking to their respective IP-PBX platforms.  In addition, these customized trunking solutions – integrated with components from industry leaders like these – come complete with sophisticated Disaster Recovery and Survivability features.

For enterprises facing the challenge of supporting a complex mix of IP, TDM, and legacy telecom infrastructures, and looking to leverage investments in MPLS networks, West UC offers MaxxConnect.  This network management solution yields a suite of IP trunking solutions designed to provide enterprise clients with the carrier–grade service they have come to expect from traditional providers – as well as the benefits of next–generation IP–based services – that allow their business to run more efficiently.

Read more SIP blog posts >

Contact a SIP Expert