Dynamic Spectrum Sharing Paving the Way to Fast 5G Roll Out

Initial 5G deployments are currently focusing on enhanced Mobile Broadband (eMBB) while mainstream deployment frequencies are predominantly on C-band, i.e., 3.4 GHz to 4.2 GHz. According to ABI Research latest work on tracking 5G deployments, more than 75% of deployments across the world are in C-band. This band offers adequate network capacity improvements, but its drawback lies in its relatively limited transmission coverage compared to current 4G networks, especially for the uplink, where the handset transmit power is limited. In an ideal scenario, Mobile Network Operators (MNOs) prefer lower band spectrum, e.g., sub 3 GHz, to boost their 5G coverage. However, most mid- and low- spectrum bands have already been allocated to other terrestrial services, like 2G/3G/4G, which means that re-farming this spectrum for 5G use will require the sunsetting of these existing services, which is typically a multi-year process. In the meantime, due to COVID-19, auctions for 5G spectrum have been postponed in many countries. In the race to complete their 5G deployments, MNOs are in dire need of working out more efficient methods to solve the above spectrum problems with their partners.

Dynamic Spectrum Sharing (DSS), the development of which has been led by Ericsson as part of 3rd Generation Partnership Project (3GPP) Release 15, offers the opportunity to allow operation of 5G in 4G Frequency Division Duplex (FDD) band simultaneously, thereby improving 5G network coverage and also avoiding a potential multi-year process of spectrum re-farming. As of February 2020, Ericsson has commercialized its DSS solution, and many MNOs testing the solution plan to deploy it in the coming months. Examples of these MNOs are AT&T and Verizon from the United States, Ooredoo from Qatar, Telia from Norway, Play from Poland, Swisscom from Switzerland, and Telstra from Australia. Huawei, Nokia, and ZTE have also developed their own DSS solutions and are currently commercialising them. Although DSS does allow for a more rapid rollout of 5G, it also has drawbacks For example, there is an efficiency loss when introducing both 4G and 5G in the same frequency, as the base station scheduler will need to switch between different generations for the same frequency band. Moreover, there are tangible differences between vendor implementations, which may lead to interoperability challenges in multi-vendor environments.

To help MNOs and potential investors to remove doubts about this new technology and drive momentum, there are three major concerns about DSS that should be clarified:

• What is the difference between Dual Connectivity (DC) and DSS?

DC was originally proposed in 3GPP Release 12 to enable two LTE eNBs supporting one mobile user simultaneously in a synergistic way. DC between 4G LTE and 5G New Radio (NR) allows 5G-enabled mobile users to dynamically switch a radio connection between the two generations for better communication quality. In contrast, DSS mainly works in a single base station and, assisted by carrier aggregation technology, it enables 5G radio transmissions operate in 4G frequency band, so that the limited transmission coverage issue of C-band can be compensated efficiently, especially for uplink scenarios. For example, DSS can couple a 3.5GHz band with a 2.1GHz band to offer a boosted 5G uplink transmission

• Does a modern 4G BS need to be upgraded to 5G NR to enable DSS?

DSS is a complicated scheduling algorithm which allows 5G users to share lower band FDD spectrum resources with 4G users, thus enhancing the spectrum utilization for both sets of users. 3GPP in Release 15 has defined new standards for 5G baseband functionality. In this case, to allow 5G users to work in the 4G FDD band, 5G baseband units must be connected to lower band FDD radio remote units and jointly work with 4G baseband to carry out DSS functionality. In other words, a modern 4G base station needs to have the ability to accommodate both 5G and 4G baseband functionalities to enable DSS, no matter if via a software or hardware-based upgrade.

• Can DSS work in a multi-vendor interoperable environment?

For DSS to be deployed effectively and to improve network capacity, base station needs to dynamically schedule Physical Resource Blocks (PRBs) for both 4G and 5G users at the 1 millisecond level. Thus, this needs to be a very tight coordination process, implementation dependent, and normally single vendor-specific solution. However, as we explained before, a modern 4G base station needs to have the capability to accommodate both 5G and 4G basebands to enable DSS. In theory, if the interface between 5G and 4G basebands is open, and low-latency requirements can be met, multi-vendor interoperability is applicable. But, considering current network development and deployment conditions, this does not seem to be realistic and is unlikely to happen in the short term. It is also unlikely that Tier-1 vendors will open interfaces within their base station platforms, making upgrades of existing 4G base stations single-vendor.

Recalling the comment that DSS would reduce network efficiency, ABI Research believes that the overall network throughput will inevitably be affected if both 4G and 5G contend for the same spectrum band, especially when the data traffic intensity from either side is high. ABI Research also believes that DSS brings value by allowing 5G users to share the same spectrum with 4G users, especially when spectrum is underutilized. Under such circumstances, to guarantee a high quality of experience for both 4G and 5G users, MNOs should consider 4G lifecycle before DSS is deployed. In addition, MNOs need to implement policiesand set up hierarchical levels to protect 4G traffic. After all, this is 4G spectrum predominantly, and MNOs must avoid harming the user experience of their loyal 4G users, who have a great potential to upgrade to 5G. To solve the spectrum shortage for 5G use, MNOs can either exploit other underutilized spectrum, e.g. 2G or 3G, via DSS, or cooperate with partner vendors to promote more advanced transmission techniques, such as massive Multiple Input, Multiple Output (MIMO) and/or coordinated multi-point (CoMP) transmissions and receptions. However, none of these offer the practical efficiencies DSS offers, and this is why several MNOs, particularly in the US, are expected to rely heavily on DSS for their nationwide deployments.

As discussed above, DSS is a complicated scheduling algorithm that has considerably high requirements for baseband hardware processing capability. If the existing chipset installed in the network infrastructure cannot meet such stringent processing requirements, MNOs have to replace or extend their existing hardware to guarantee DSS functionality. To shorten the time of deployment and save potential capital expenditure, MNOs should choose radio equipment that can support modularized and scalable architectures. In this way, by lightweight network reconstruction or software upgrades, the new features, like DSS, can be easily implemented. Apart from that, mobile devices are another important concern, where substantial derives that are ready for 5G will foster the deployment of DSS.

ABI Research sees DSS as a very balanced initiative to facilitate a smoother transition from current 4G network to 5G NR network; earnestly giving full play to its role will pave the way to fast 5G rollout.