Shannon Theorem and 5G Capacity
The central premise of most science fiction is “Everything Connected and Communicating wirelessly”, and any wired connection is part of time travel to the 21st century. The telecom industry is moving on this path for some time. The launch of the iPhone in 2007 can be considered a fundamental alteration in this path, with content consumption, eventually moving to the mobile world that also is opening up the entire industry to the creation of new services with a focus on the “Mobile first” domain.
Now we are entering the 5th Generation of the mobile world, and paraphrasing Andreessen Horowitz means “Mobile has started becoming a fabric of every aspect of the society as well as the industry”.
The 5G is fundamentally different from the earlier generations with the core network catching up with new software services architecture as well as defining the standards — in contrast to the previous generation focused on more providing high-speed internet.
All the services, applications use cases as well as plans to make 5G part of the socio-economic fabric are because of the ability to provide high capacity due to the advances ushered by 5G technology.
We can say unambiguously,
“Capacity is the killer app of 5G”
ITU defined the IMT-2020 standard with a vision of increase in area capacity by 100 times, peak data rate by 20 times, and User experienced data rate by 10 times.
These are huge jumps from earlier wireless systems. How did the industry achieve these numbers? I am going to take a different approach and will solely focus on Shannon Capacity Theorem, how each decision point is related to Shanon Theorem.
Shannon Capacity Theorem, also known as Noisy-channel coding theorem as well as Shanon’s limit, establishes that:
“For any given degree of noise contamination of a communication channel, it is possible to communicate discrete data (digital information) nearly error-free up to a computable maximum rate through the channel.”
In simplified words, It defines the maximum amount of information, or data capacity, which can be sent over any channel or medium (wireless, coax, copper, fiber, etc.), and this limit is a function of available bandwidth and signal-to-noise ratio of the link.
In the wireless world with MIMO** (multiple input multiple outputs) antenna systems deployed, bandwidth is dependent upon the beams created by the antenna systems. The noise in the field (deployed cellular system) has the component of interference from other users’ signals too along with the white noise. In a scenario of multiple users accessing the same base station/cell, the bandwidth per user is reduced by proportion to the total users.
Example -Considering the total capacity of the channel, the number of users can be taken as “1”.
In 5G world, we are talking about the high capacity of the system, focussing on high-speed network availability to the consumer. This article will focus on the increase in the capacity of the 5G network from the Shanon Theorem’s perspective. Along with technology, there are Systems (architectural) changes made by 3GPP in the 5G end-to-end networking which are helping in achieving the capacity increase.
The below figure provides a holistic view of how 3GPP is able to achieve high capacity by adopting new technologies as well as putting changes in architecture. As you can see, we can connect these changes to the Shannon Theorem and how these changes/parameters impact the capacity.
On the technology front, the capacity targets are achieved by supporting the
wider spectrum bandwidth as well as increasing spectrum enhancements. At systems levels, it is achieved with more efficient 5G planning and networking.
The below fig provides the relationship of these changes and the impact on channel bandwidth (as defined by Shanon’s Theorem).
Spectrum Bandwidth: Support of mobile phone service in mmWave (24Ghz to 52GHz, and support till 100GHz is coming in future releases) is the biggest change in 5G service. At this frequency range, it’s possible to provide high bandwidth of 100MHz per channel (per endpoint).
Spectrum Efficiency: Along with mmWave support, a massive-MIMO antenna system has changed the way cellular networks will be planned, designed, and deployed. With the massive-MIMO antenna system, the increased spectral efficiency as well as high bit rate is achieved with antenna gain, diversity gain as well as opened the new domain of spatial multiplexing.
3GPP adopted the same multiplexing and modulation techniques as of 4G for 5G NR (New Radio). To support a high capacity symmetrical network (both high capacity UL as well as DL), 5G supports 256 QAM modulation. In the future, 5G will support up to 1024 QAM modulation methods. Using Filtered OFDM (F-OFDM), the reduction of guard band between consecutive waveforms from 10% to 2–3% is achieved.
One newly adopted coding technique will have a long-lasting impact on future wireless standards and technology evolution. 3GPP has adopted the “Polar Codes” for the control plane. Polar codes, invented by Erdal Arikan in 2008, are the first code invented in the last 60-years and these are meeting the Shannon limit of the capacity. Though currently, they are still in their first iteration, we may see their adoption in the user plane too in the future. Just adopted new DECT-2000 NR by ITU for IMT-2020 (5G NR), is using Polar codes.
Systems Engineering: The standard bodies and industry have updated the end-to-end architecture as well as network deployment methods. To get the maximum advantage of mmWave band, small cell deployment methodologies are used. Traffic offloading to other access technologies as well as Dynamic Spectrum sharing/Carrier aggregation help in minimizing the hot spot in dense areas (as well as a dense spectrum). Small Cell architecture helps in maximizing the spectral efficiency in the mmWave spectrum.
We can see the above-discussed technology enhancement, as well as Systems changes, to meet the goal of high capacity of 5G system, are be connected with Shanon’s Theorem.
In future blog entries, I will cover the evolution of the core network from legacy point-to-point network to API-based Cloud Native world.
5G iteration can be considered a tipping point in mobile phone evolution and we may be entering the “Democratization of the Mobile Technology”, similar to what happened with the Web world in the last 2 decades.