Viewers are making it clear that they want to watch live events from their mobile devices as well as from the comfort of their own homes. Although internet streaming has given us a hint of what is achievable, its inability to scale to meet viewing demands is its Achilles’ heel leaving viewers frustrated and in need of a better solution.
Other articles from this series.
Internet streaming is fundamentally flawed when we try to scale. Every device requiring streamed media demands chunks of data from media servers resulting in a resource-heavy delivery system. Adding a viewer’s device to a streamed event makes a specific demand on the internet infrastructure, and when this reaches a certain tipping point, all users are affected and suffer a significant degradation in the quality of their signal and hence experience. 5G-NR solves these challenges to greatly improve the viewer experience and provide higher quality services such as 4K and 8K streaming. This in part is due to the broadcast capabilities that allows 5G-NR to send media directly to a mobile device regardless of the number of viewers watching the service.
Mobile data has traditionally used the unicast method of delivery. This works particularly well for web page type applications but is flawed when we consider large file downloads and media streaming. To work with the internet, traffic must be TCP/IP compliant, and when web pages are used, they must be HTTP/TCP/IP compliant. This is one of the fundamental challenges we have with streaming media and downloading large files over the internet. Although the TCP/IP provides guaranteed delivery for IP packets, it does so at the great expense of variable latency.
Latency has been gaining a lot of interest in recent years as broadcasters are seeing massive delays when streaming media over the internet, sometimes more than 60 seconds. The unicast restrictions that are manifested by TCP/IP, combined with the multitude of video and audio chunking buffers that are required, all contribute to the latency.
Broadcasters have historically not had to contend with this type of latency as there has always been sufficient RF bandwidth available for true one-to-many transmission over the airwaves. However, as the popularity of internet streaming continues to grow, new methods of delivering large volumes of content to viewers is required. To achieve this, the 3GPP consortium created the MBMS (Multimedia Broadcast Multicast Service), the eMBMS (Evolved MBMS) and then the FeMBMS (Further Evolved
Figure 1 – traditional broadcasting workflows kept latency to below 500 milliseconds, mainly due to the one-to-many transmission that the uncontested RF allowed. However, delivery over the internet greatly increases latency and makes it unpredictable due to the influences of TCP/IP and video and audio chunking, and the effects this has on buffers.
The MBMS system is a one-to-many efficient broadcasting feature that is an alternative to the unicast method of operation and was originally provisioned by the 3GPP consortium in release 5 back in 2015. Multiple international tests and technology demonstrations have taken place, but the fixed allocation of bandwidth that could be provided for both 3G and 4G meant its take-up was limited.
5G-NR is a combination of four technologies that are defined as: eMBB (Enhanced Mobile Broadband, mMTC (massive Machine Type Communications), URLLC (Ultra-Reliable Low-Latency), and eMBMS (Evolved Multimedia Broadcast and Multicast Services). 5G implementation is advancing as more parts of the specification are deployed to the network.
eMBB is designed for high data rate applications and is seen as an advance on the 4G LTE mobile services, but with much higher bandwidth, data throughput, and reduced latency. The key objectives of eMBB are to provide a seamless user experience with greater coverage and faster data speeds, and it provides three main use cases: allow data access to a dense collection of users, provide data for highly mobile users, and deliver to users spread over wide areas.
Peak download data rate is 20Gbs with upload data rate of 10Gbps, and for 95% of the time, 100Mbps should be achieved. The data capacity of the network is 10,000 times that of 4G with coverage at approximately 10 Mbps/m2, and high-speed mobility is available with mobile devices moving up to 500Kmph.
As an example, in a sports stadium we would have tens of thousands of stationary supporters, but at the opposite extreme, in a train for example, there would be much fewer viewers watching events travelling very fast. The ultimate goal for eMBB is that users can get connected and stay connected as they move within the eMBB network.
Connectivity is improved so that users do not have to consistently hop from the network to private WiFi as is often the case with current 4G technology.
mMTC caters for high volumes of connectivity as expected with IoT (Internet of Things) but a relatively low data rates. Device connection density is in the order of 1,000,000 devices per square kilometre, but with a low data rate in the order of 1 to 100Kbps. The device density is approximately ten times more than 4G. Long range is supported with long device battery life so that large numbers of IoT devices can be deployed that regularly transmit small amounts of data. The expectation is that the battery life of a 5G connected IoT device will be up to ten years.
URLLC is provisioned for mission critical applications where the short scheduling intervals of the packets will deliver latencies in the order of 62.5uS at the physical layer. It provides an air interface latency of less than 1ms and a 5ms end to end latency between the user equipment and 5G-NR the base station. Medium to low data rates of approximately 50 kbps to 10 Mbps are provisioned.
Although 5G-NR may have some analogies to 4G LTE, it’s unfair to suggest 5G-NR is merely faster 4G as it provides more use cases than the LTE standard. The eMBB, mMTC, and URLLC all combine to provide a system that is future proof and will meet our needs for many decades to come. Other differences include the variable subcarrier spacing and use of bandwidth parts in the frequency allocation, both of which deliver higher data throughput with lower latency.
Two frequency ranges are currently defined for 5G – FR1 and FR2. FR1 covers the range below 6GHz, and FR2 has a range up to 60GHz, although some countries have the option of using UHF frequencies that are being released from broadcast television. The FR2 range makes provision for the millimetre wavelength spectrum to deliver the high data rates with ultra-low latency, however, this is at the cost of much reduced range of a few hundred meters and the inability to penetrate walls in buildings.
To achieve the lower latencies of less than 1mS, it’s fair to say that we will need to use the higher FR2 millimetre frequency ranges. However, this does not preclude our ability to use URLCC at the lower FR1 frequencies. 5G is a completely managed and dynamic network that automatically allocates frequencies, bandwidth, and latency on demand, depending on the user demand.
You might also like...
GatesAir continues its TV (and Radio) transmission supremacy.
Hitachi Kokusai Electric Comark marks half century of serving the broadcast community.
In this part we look at the most elegant way of finding and defining a sine wave.
We discuss the roll out of ATSC 3.0 in the USA with Jerald Fritz, Executive Vice President for Strategic and Legal Affairs at ONE Media 3.0 - part of Sinclair Broadcast Group.
In television, ‘talent’ isn’t just the people in front of the camera. Everyone working at a station needs talent, dedication, initiative, and team spirit to succeed.