In 1948, information-theory pioneer Claude Shannon showed how to calculate the channel capacity, or Shannon limit, which is the theoretical maximum data rate over a “noisy” communications medium before incurring transmission errors. Over 70 years later, Professor Xiaohu You of Southeast University is exploring techniques that might enable 6G to effectively ignore the Shannon limit as detailed in a paper published in Scientia Sinica Informationis. More than that, he has proposed a number of techniques to improve not only data rates but reliability and spectrum re-use that can help meet 6G’s stringent demands… by breaking down the concept of cellular communications itself.

Background

The Shannon limit, or channel capacity of a communication channel, refers to the theoretical maximum rate of error-free data that can be transferred over the channel if the link is subject to random data-transmission errors – for example, due to an imperfect transmission medium or interference.

To use an analogy, if all that a thirsty person cared about was receiving the maximum possible flow of water, you could take a river and divert it through their house. However, the water will pick up any dirt, insects, and movable items it flows over along the way. By contrast, good drinking water needs to be filtered clean of aquatic life, silt and other contaminants. This process limits the flow of water.

In data transmission, error-correcting processes and codes effectively act as a ‘filter’ on the flow of digits, ensuring what is received is usable information and requesting that any corrupted data be re-sent. Claude Shannon demonstrated a mathematical proof that, due to the need for error-correction, there is a maximum limit to the flow of accurate information in any given data channel. A data channel in a mobile context is defined by a particular frequency and time.

As Professor You points out, implicit in Shannon’s work is the idea of a relationship between one sending antenna and one receiver. That relationship essentially creates the cellular structure of mobile telecoms that we know today.

Since then, there have been several advances that have taken us beyond Shannon’s original premise:

• MIMO systems increase a cell’s capacity by adding directionality to the data channel. That is, you can now use the same channel to send different information in different directions. As You explains, “If the number of antennas at both ends of the wireless link is increased, the MIMO channel capacity can theoretically increase linearly. Unfortunately, due to the limited physical size of the user terminal, the number of antennas that can be installed is limited.”

• Multi-user MIMO (MU-MIMO). “Its principle is based on the following basic facts”, says You, “Although the number of antennas of a single user terminal is limited, the combination of multiple users can form a point-to-multipoint (P2MP) joint processing MU-MIMO system between a base station and multiple users… But it should be noted that the number of antennas on the user side is the sum of the number of antennas of each user, and the channel capacity is the sum-rate of each user.”

In other words, MU-MIMO overcomes some of the limitations from the end-user device standpoint. However, practical limitations now start to appear at the base station instead. “When the physical size of the base station antenna array deployment is limited, there will be a serious mutual coupling effect in too dense antenna deployment, and there is a certain theoretical limit to the related MIMO channel capacity.”

Beyond Cells

Both MIMO and MU-MIMO improve the spectrum utilisation of a cell. However, they still adopt a cellular approach to communications by default.

Professor You’s paper goes on to outline how the principles underlying MU-MIMO can be extended if processing is performed jointly not just across multiple users, but across multiple base stations, “thereby forming a multi-point-to-multipoint (MP2MP) form of cell-free mobile communication system… For a non-cellular system, due to the introduction of multi-cell joint processing, multiple users form MP2MP distributed MU-MIMO in the coverage of multiple cells, and all users and base stations can work at the same frequency.”

Professor You claims this brings a number of advantages:

• It eliminates the limitation of the traditional cellular architecture in terms of cell frequency re-use, and its cell frequency re-use factor is equal to one. “This means that the non-cellular system is no longer restricted by the static allocation of frequency resources.”

• It can also be used to assist in delivering the requirements of Ultra reliable Low Latency Communications(URLLC). You’s argument is that a cellular model of error-correction relies on requests to re-send incorrect data packets, increasing latency. If, instead, multiple sources send the same data stream to a URLLC device then it both increases the probability that accurate data reaches the device – enhancing reliability – and reduces the need for re-sending packets because the error correction can be performed on-device. “The advantage of this type of technical approach is that there is no need to change the architecture of the MIMO wireless transmission system, you only need to adjust the data stream and precoding at the sending end, which can easily find a flexible compromise between the transmission rate and reliability, and can be directly extended to MU-MIMO.”

Professor You admits the essential technical and business challenge lies with the joint processing of data across multiple base stations: “Further improvement of 6G core technical indicators is all accompanied by the increase in system deployment costs and complexity, so seeking an effective balance between system performance and implementation costs will be the main challenge for the future development of 6G mobile communications.”

Note: Artemis Networks has, for some years now, been advocating a somewhat similar “beyond cells” concept applied to LTE and 5G in a consumer small cell/ high base station density context.