Nearly 33% of Video Sources in Hong Kong Have a Mobile vMOS of 4 or Higher, Heralding Experience 4.0 by MBB 2020
At the beginning of November 2015, the SpeedVideo software was used in Hong Kong to test the mobile vMOSs of sampled YouTube video sources on networks of three leading operators: P, C, and S. The tests covered various coverage scenarios, such as downtown business districts, scenic spots, and hotels. Key findings are as follows:
1. Operator P has the highest mobile vMOS (3.52) for 2K video sources, followed by operators C (3.44) and S (3.21).
2. The proportion of mobile vMOSs higher than 4 for 2K video sources is 34.6% for operator P, 26.2% for operator C, and 32.5% for operator S. The objective of a network-wide mobile vMOS of 4 or higher is challenging but achievable.
3. Operator P has the highest video download speed (8.97 Mbit/s), preceding operators C (5.93 Mbit/s) and S (4.22 Mbit/s).
4. Operator S has the highest RSRP (–74.8 dBm), followed by operators P (–78.6 dBm) and C (–84 dBm).
5. Operator S has the shortest RTT (32 ms), followed by operators C (33 ms) and P (40 ms).
6. In certain areas covered by operator S's network, video freeze occurred frequently for 2K video sources due to a low download speed during specific time periods. As a result, the mobile vMOS for 2K video sources is lower than that for 1080p video sources.
7. For a video source whose sQuality remains unchanged, the mobile vMOS, sStalling, and sLoading decrease as the RSRP decreases.
8. For a single video source sample, the vMOS on an LTE network is higher than that on a UMTS network. In areas with poor LTE network coverage, however, the average vMOS on the LTE network may be lower than that on the UMTS network. Operators are advised to optimize their LTE network promptly to expand coverage and lead more users into watching HD videos, thereby obtaining a higher mobile vMOS.
9. On a lightly-loaded LTE network with a high bandwidth, 2K videos are provided in typical cases. Differences in mobile vMOS between UMTS and LTE networks can be detected only in heavily-loaded areas during specific time periods.
10. The preceding test results are for reference only. They do not represent the actual conditions of each network due to a limited number of video source samples and the network status uncertainty in terms of time and space.
The test scenario and method are as follows:
Scenario: Comparison of user experience with YouTube video services among operators P, C, and S
Test case: 720p, 1080p, and 2K YouTube videos
Test period: 09:00–22:00 from November 5, 2015 to November 7, 2015
Test UE: Huawei Mate7 (with the SpeedVideo software installed)
Test method: drive tests (DTs) in a selected area
The following figure shows the comparison of mobile vMOS distribution for YouTube 2K videos among the three operators.
The following figure shows the comparison of download speed distribution among the three operators.
The following figure shows the comparison of RTT distribution among the three operators.
The following figure shows the comparison of RSRP distribution among the three operators.
Data analysis further indicates a weak positive correlation between the mobile vMOS and RSRP:
1. The total downlink throughput decreases as the RSRP decreases.
2. For a video source with fluctuating RSRP but sQuality remaining unchanged, the mobile vMOS, sStalling, and sLoading decrease as the RSRP decreases.
The SpeedVideo software used for mobile vMOS evaluation can be downloaded by using any of the following methods:
1. Visit //mlabvmos.com.
2. Search and download the SpeedVideo from the Huawei App Market.
3. Scan the following quick response (QR) code to download the SpeedVideo: