Frame Error Rate Testing - WESCO International

3
132 White Oak Road | New Holland, PA 17557 | www.berktek.com | 800-237-5835 September 2015 Frame Error Rate Testing Frame error rate testing characterizes data packet transmission performance in real-world conditions. Increasingly important metric with so many IP devices connected and powered via the network, that the additional heat can lead to more errors. More stringent metric than BER because a single frame includes many bits The results can be expressed as a total number of errors, as a percentage between 0% (no errors) and 100% (all errors), or as 10 -XX (lab use). There are multiple ways in which the quality of networking products can be assessed. Two clearly defined techniques are the measurement of bit error rate (BER) and frame error rate (FER). Many networking products are specified with the bit error rate that they achieve or support. BER is simply the ratio of bits that have been incorrectly received to the total number of bits sent. The test is performed by sending strings of pseudo-randomly generated bits through the device. Frame error rate (FER) is a better test for two reasons: 1. In real-world Ethernet networks, the basic data unit is a frame, a collection of bits arranged in a format consistent with the Ethernet standard. A frame can contain between 512 and 12,144 bits per the Ethernet standard. If a switch or a NIC receives a frame with one or many incorrectly transmitted bits, the complete frame will be discarded. The degradation of network performance due to dropped frames is the same for one or many bit errors within a frame. BER does not account for this fact. 2. FER testing takes longer to complete. Berk-Tek performs FER testing for a minimum of one hour per test, whereas a BER test would take about two (2) minutes. This increased test time is significant because it allows the hardware and cabling to reach operating temperatures. The performance of networking products decreases as the temperature increases. Robust performance is therefore better verified during longer tests that allow for transceivers, switches, cabling, etc., to reach their true operating temperature. Berk-Tek’s team of scientists and engineers characterize all of Berk-Tek’s copper and fiber optic solutions with FER using state-of-the-art test equipment and proprietary methodologies. This provides customers with robust networking solutions that perform in real-world conditions. Application: Converging IP traffic Challenge: Quantifying converged network performance under real-world conditions. Solution: Measure the Frame Error Rate of Berk-Tek LANmark TM -1000, LANmark TM -2000, and LANmark TM -XTP cabling using state-of-the-art test equipment in real-world conditions to ensure robust network performance today, and into the future. This stamp certifies that all testing was performed, reviewed, and approved by highly trained, experienced engineers dedicated to studying and developing solutions for future network infrastructures.

Transcript of Frame Error Rate Testing - WESCO International

Page 1: Frame Error Rate Testing - WESCO International

132 White Oak Road | New Holland, PA 17557 | www.berktek.com | 800-237-5835

September 2015

Frame Error Rate Testing

• Frame error rate testing characterizes data packet transmission performance inreal-world conditions.

• Increasingly important metric with so many IP devices connected and powered viathe network, that the additional heat can lead to more errors.

• More stringent metric than BER because a single frame includes many bits

• The results can be expressed as a total number of errors, as a percentage between0% (no errors) and 100% (all errors), or as 10-XX (lab use).

There are multiple ways in which the quality of networking products can be assessed. Two clearly defined techniques are the measurement of bit error rate (BER) and frame error rate (FER). Many networking products are specified with the bit error rate that they achieve or support. BER is simply the ratio of bits that have been incorrectly received to the total number of bits sent. The test is performed by sending strings of pseudo-randomly generated bits through the device.

Frame error rate (FER) is a better test for two reasons:

1. In real-world Ethernet networks, the basic data unit is a frame, a collection of bitsarranged in a format consistent with the Ethernet standard. A frame can containbetween 512 and 12,144 bits per the Ethernet standard. If a switch or a NIC receivesa frame with one or many incorrectly transmitted bits, the complete frame will bediscarded. The degradation of network performance due to dropped frames is the samefor one or many bit errors within a frame. BER does not account for this fact.

2. FER testing takes longer to complete. Berk-Tek performs FER testing for a minimum of one hour per test, whereas a BER test would take about two (2) minutes. This increased test time is significant because it allows the hardware and cabling to reach operating temperatures. The performance of networking products decreases as the temperature increases. Robust performance is therefore better verified during longer tests that allow for transceivers, switches, cabling,etc., to reach their true operating temperature.

Berk-Tek’s team of scientists and engineers characterize all of Berk-Tek’s copper and fiber optic solutions with FER using state-of-the-art test equipment and proprietary methodologies. This provides customers with robust networking solutions that perform in real-world conditions.

Application:

Converging IP traffic

Challenge:

Quantifying converged network performance under real-world conditions.

Solution:

Measure the Frame Error Rate of Berk-Tek LANmarkTM-1000, LANmarkTM-2000, and LANmarkTM-XTP cabling using state-of-the-art test equipment in real-world conditions to ensure robust network performance today, and into the future.

This stamp certifies that all testing was performed, reviewed, and approved by highly trained, experienced engineers dedicated to studying and developing solutions for future network infrastructures.

Page 2: Frame Error Rate Testing - WESCO International

132 White Oak Road | New Holland, PA 17557 | www.berktek.com | 800-237-5835

September 2015

IPTV Quality

• IPTV provides multi-media services delivered over IP networks.

• IPTV quality is measured using Media Loss Rate (MLR).

• MLR is a measure of lost or out-of-order packets (frame errors) over a specified time period.

• A MLR > 0, will equate to degraded video quality.

IPTV is an emerging IP network application, increasing service options and flexibility for IP customers. IPTV offers significant advantages over a traditional cable television distribution network by allowing organizations to control and manage content in a central location and distribute that content via an IP network. To achieve consistent high-quality video transmission, very tight timing is required, therefore IPTV uses Real Time Protocol (RTP). Unlike TCP, (the Ethernet protocol most commonly associated with Ethernet data transmission) RTP does not request retransmission of packets that contain errors. Whatever was in those packets is lost forever to the end user. This loss has noticeable effects.

Since even very low data loss has noticeable, visible impact on the user experience of video, evaluating IPTV performance requires unique quality of service metrics. The Media Loss Rate (MLR) is a measure of video quality that measures lost or out-of-order media packets used within the IPTV application.

Negligible MLR is desired, because any packet loss will cause a visible distortion to the picture quality. The table below provides MLR values that would generally be considered acceptable, highlighting the different quality levels needed to ensure either High Definition or Standard Definition transmissions. Also provided are the equivalent average frame errors per hour. It is important to recognize that these MLR values equate to acceptable performance for instances of ongoing viewing of media, as opposed to changing between streams (i.e., channel surfing) which demands even lower MLR values to ensure performance.

Berk-Tek uses sophisticated analytical tools to measure the MLR of networks using industry leading Berk-Tek LANmark™-1000, LANmark™-2000, and LANmark™-XTP solutions. Berk-Tek solutions are subjected to real world environmental conditions and demonstrate reliable and robust IPTV performance. This provides IT managers with the confidence to leverage IPTV in their converging networks without worrying about degradation to video quality.

Application:

Ensuring that users experience clear and reliable IPTV performance.

Challenge:

Simple measurements of bit error rate (BER) do not sufficiently characterize IPTV performance in real world scenarios. MLR was specifically designed to measure IPTV video application quality.

Solution:

Berk-Tek LANmark™-1000, LANmark™-2000, and LANmark™-XTPare tested for IPTV transmission, ensuring that the best quality video can be delivered.

MLR Calculation:

MLR is technically expressed as a media loss rate per second and the following formula is used to calculate MLR value to Equivalent Max Avg:

This stamp certifies that all testing was performed, reviewed, and approved by highly trained, experienced engineers dedicated to studying and developing solutions for future network infrastructures.

IPTV MLR Value Equivalent Max Avg Frame Errors per Hr Quality Level

0 0.00 Ideal - Best Quality

0.0005 or less ≈ 0.25 Acceptable for HDTV Service

0.004 or less ≈ 2.00 Acceptable for SDTV Service

2 IPpackets

60 minutes

1minute

60 seconds

7 mediapackets

1 IPpacket

X X = 0.004

Formula from Agilent Technoloies White Paper: IPTV QoE, Undersanding and Interpreting MDI Values). Assume 7 media packets per IP packet.

Page 3: Frame Error Rate Testing - WESCO International

132 White Oak Road | New Holland, PA 17557 | www.berktek.com | 800-237-5835

September 2015

Measuring VoIP Call Quality

• VoIP call quality can be estimated using network performance metrics.

• MOS is an objective, quantitative measurement capable of classifying user experience as “Very Satisfied,” “Satisfied,” “Some Users Satisfied,” “Many Users Dissatisfied” or “Nearly All Users Dissatisfied.”

• Berk-Tek tests network infrastructure in real world conditions to validate VoIP performance.

Historically, analog telephone call quality was measured by asking a group of people to rate the auditory clarity of the sound by assigning a score or ranking based on their perceptions of the sound. These subjective scores were then averaged and reported as the Mean Opinion Score. Using this type of methodology can produce inconsistent results.

Unlike analog telephony, voice over internet protocol (VoIP) converts sound into packets of data. This makes it possible for software to be used to analyze the data transmission, which then results in repeatable, quantitative measurement of call quality that is correlated to user experience.

The International Telecommunication Union (ITU) is the branch of the United Nations that handles issues related to information and communications technologies (ICT). Its document, G.107, defines a model for rating the quality of a ‘mouth-to-ear’ transmission so that planners can design and install systems that provide users with satisfactory end-to-end performance. One of the metrics from this document is Mean Opinion Score Call Quality Estimated (MOScqe). For VoIP, this metric takes into account parameters specific to data networks such as one-way delay, latency and packet loss. Capturing these metrics is important because VoIP protocols send data packets only once (typically UDP), so any error causes a part of the audio stream to be permanently lost.

As IP networks converge, and more devices are connected, the stresses placed on the network’s structured cabling infrastructure increase. Many IP devices, including VoIP phones are also powered via PoE or PoE+. As power increases, so does attenuation caused by heat, and the chance that errors occur. Power is also inherently “noisy,” and this noise can interfere with voice and data packet transmission as well.

Berk-Tek uses sophisticated analytical tools to measure the MOS of network infrastructure using Berk-Tek’s premium solutions. When these networks are subjected to real world environmental conditions, they exhibit reliable, robust VoIP performance. With these results, IT managers can confidently leverage VoIP in their networks without worrying about degradation to call quality.

Application:

Real World VoIP

Challenge:

Simple measurements of bit error rate (BER) do not sufficiently characterize VoIP performance in real world scenarios.

Solution:

Berk-Tek’s LANmark™-1000, LANmark™-2000, and LANmark™-XTPare characterized using MOS, an objective, quantitative measurement of call quality. This ensures that Berk-Tek products enable robust and reliable VoIP performance.

The MOS is divided into five distinct categories that are described in the table below:

This stamp certifies that all testing was performed, reviewed, and approved by highly trained, experienced engineers dedicated to studying and developing solutions for future network infrastructures.

MOS Range User Satisfaction

4.34-5.00 Very Satisfied

4.03-4.33 Satisfied

3.60-4.02 Some Users Dissatisfied

3.10-3.59 Many Users Dissatisfied

2.58-3.09 Nearly All User Dissatisfied