In the past, the majority of business phone systems relied on desk phones and landlines. Since the introduction of VoIP technology, more businesses have been rapidly switching to VoIP phone systems due to a number of advantages they offer. That becomes especially prevalent in the context of today’s changing workplace landscape as the workforce becomes more distributed. VoIP has made it possible to make cheaper calls regardless of the distance. However, since VoIP is internet-based, VoIP call quality largely depends on the reliability of the internet connection. In order to ensure a consistent user experience, voice quality metrics, like R-Factor and MOS (Mean Opinion Score) were introduced.
What is R-Factor in VoIP?
R-Factor (Rating Factor) is one of the two most commonly used measurements of overall VoIP call quality in IP networks. The R-Factor value is derived from metrics such as packet loss, packet delay variation (jitter), and latency. It allows you to quickly assess the quality of experience for calls on your VoIP network. The three main variations of R-Factors are R-Call Quality Estimate, R-Listening Quality Estimate, and R-Network Performance Estimate.
R-Factor values range from 0 to 100, with 100 being perfect. In reality, however, there is always some unavoidable degradation that may affect the R-value. It may be caused by the end equipment used (microphone impairment), room noise, losses in the network, and compression algorithms used. Typical telephone connections don’t get above 93 R-Factor, so the most common scores range from 50 (poor quality) to 90 (high quality). R-Factor with a value of under 50 is generally unacceptable.
How do you measure R-Factor in VoIP Quality?
The following equation measures the R-Factor value:
R = Ro – Is – Id – Ie-eff – Irecency + A
- Ro – Signal-to-noise ratio (SNR) value that is determined by noise levels, loudness, etc., including circuit noise, room noise, and subscriber line noise. It is the highest value of R.
- Is – All the impairments manifest simultaneously with the voice signal. The major contributing factors are loudness ratings of the phone set, side-tone loudness rating, and the number of quantization distortion units.
- Id – The impairments introduced by delay. The factors that contribute to these impairments are the amount of delay present in the network as well as the values of talker and listener echo loudness ratings. The Id value is typically small and does not significantly degrade the R-Factor. The degradation can be significant in case the delay becomes large enough.
- Ie-eff – A larger impairment component than Id that comprises impairments resulting from low bit rate CODECs, packet loss, and packet rejection.
- Irecency – The impairments caused by significant packet loss — eight or more packets lost in a row.
- A – The advantage factor, which allows for the compensation of impairment factors and represents the user tolerance to the degradation of the voice quality. For wire-bound communication, the advantage factor A component is always 0 because the test device has no prior knowledge of this component.
What is the Difference between R-Factor and MOS in VoIP?
R-Factor and MOS are both call quality measurements in VoIP, though there is a difference in the scales they use and the testing process.
The MOS score reports on three factors: listening quality, transmission quality, and conversational quality. Originally, MOS (Mean Opinion Score) was a subjective test that was based on user perception rather than measurable statistics. MOS values were derived by having a statistically large number of people listen to calls and rank them on a scale of 1 to 5, where 1 would mean bad quality (I can’t hear you), 2 – poor, 3 – fair, 4 – good and 5 – excellent (We’re sitting next to each other).
Later on, MOS testing has been adapted to VoIP in the ITU-T PESQ P.862 standard, which defines how to calculate a MOS score considering multiple factors like bandwidth, jitter, latency, packet loss, and codec version used. The modern approach of measuring MOS score is based on sophisticated algorithms (Objective Measurement Methods) that attempt to approximate human experience and predict perceived call quality by the human ear.
Currently, on average, VoIP calls fall in the 3.5 – 4.2 MOS range. A 5.0 MOS score has been compared to in-person, face-to-face conversations. However, it’s generally uncommon. The maximum MOS in VoIP for a G.711 codec (the most popular VoIP codec today) is 4.4.
Unlike a limited scale of 1 to 5 that is used to measure MOS, R-Factor scaling from 0 to 100 is somewhat more objective and allows ranking the score with more precision in order to determine the VoIP call quality and VoIP network health. The following chart provides MOS and R-Factor comparison values:
|User Satisfaction Level||MOS||R-Factor|
|Maximum using G.711||4.4<||93|
|Excellent||4.3 – 5.0||90 – 100|
|Good||4.0 – 4.3||80 – 90|
|Satisfied||3.6 – 4||70 – 80|
|Dissatisfied||3.1 – 3.6||60 – 70|
|Fully dissatisfied||2.6 – 3.1||50 – 60|
|Not recommended||1.0 – 2.6||Less than 50|
Using a reliable internet connection to make VoIP calls made it possible for people to lower the cost of their calls. As a way to ensure the quality of user experience, you should evaluate the VoIP call quality based on the R-Factor transmission rating. This guide and others on VoIP-info.org can help you with learning how to measure it.
Oliver Stasinszky is an outreach team lead at LiveAgent, with an e-commerce and customer service background. Passionate about writing, reading, and learning how to play any musical instrument he comes across.