A Comparison of Next-Generation FWA vs. 3GPP

3GPP standards-based radio networks (4G/5G) are designed to deliver excellent mobile services. They are not well-designed, however, to meet the unique demands of fixed wireless access (FWA). This can result in performance differences where next-generation fixed wireless access (ngFWA) delivers up to 2x downlink and 10x uplink speeds.

These differences arise from the how and why these two technologies came to be. We’ll cover a few of these differences here. For a detailed discussion, please refer to the full white paper here.

3GPP: Mobile Constraints for Broadband Delivery

The 3GPP standards were developed based on the premise of mobile devices (e.g., mobile phones) using licensed spectrum to deliver broadband. However, these mobile devices are inherently limited in power, size, and computational capacity. Over time, new 3GPP standards have added more capabilities, but, because the user equipment (UE) is relatively simple, nearly all of the cost and complexity resides in the base station.

This design choice comes with consequences. While the base station may be capable of making sophisticated decisions on one end of the link, there is little the UE can do to improve conditions on its side of the link (uplink). Making the UE more capable could improve link capacity as well as help with performance degradation due to interference but due to the drive to keep the UE simple, that rarely happens.

Mobile networks are also typically not symmetrical in terms of link budget. This significantly impacts uplink performance. As more uplink-heavy applications such as video conferencing become widespread the need for faster uplink speeds – in particular symmetric speeds – becomes necessary. Mobile broadband networks are simply not designed to meet this need.

A final consideration is data tonnage. Fixed broadband subscribers have far higher bandwidth usage than mobile subscribers. A fixed broadband subscriber typically uses 50 times more data than a mobile subscriber. This far exceeds the capacity most mobile broadband networks are architected for which limits scalability.

Tarana G1: A Different Approach

In contrast, Tarana’s Gigabit 1 (G1) was built from the ground up with the specific aim of addressing the unique requirements and challenges of fixed wireless networks. The architecture of G1 was designed such that FWA UEs, being fixed terminals, can benefit from significant power, size, and computation enhancements over handheld devices. This allows for the use of larger, more sophisticated antenna arrays, as well as higher-order, computationally intensive multi-dimensional signal processing capabilities at both ends of the link. With 2.4 Gbps of sector capacity and link speeds up to 800 Mbps (aggregate) and speeds of 1.6 Gbps coming soon, G1 answers the needs of data hungry subscribers.

In a typical deployment where there is a high density of users and limited capacity, the purpose-built architecture of G1 allows for:

  • Two times faster link speeds on average and eight times faster link speeds at the cell edge using equivalent spectrum.
  • Three times higher overall spectral efficiency (capacity per amount of spectrum allocation), implying a third the number of towers in well-loaded networks.
  • Symmetric signal strength in uplink and downlink, enabling excellent range and flexible up/down bandwidth division with no loss in capacity.
  • Self-interference cancellation allowing hyper-efficient use of spectrum (no frequency interleaving required).
  • Unique asynchronous interference cancellation, enabling reliable operation in unlicensed spectrum.
  • Consistent performance with network densification, limiting service degradation when adding new subscribers, a common issue in scaling fixed wireless broadband networks.
  • Hundreds of megabits to gigabit speeds

G1 and 3GPP: Operator Validation

After implementing both 3GPP networks and G1 in their FWA services, several Tarana customers have reported that the real-world performance of 3GPP networks does not live up to promised results. The table below, from Alyrica Networks, represents side-by-side testing in which all variables were kept identical for G1 and a leading 3GPP vendor. 





Leave a Reply

Your email address will not be published. Required fields are marked *