Rapidly increasing growth in data traffic across mobile infrastructure continues to stretch networks to their limits. A critical part of the network design is backhaul – namely taking the traffic from the cell site back to the core. Typically, one of three technologies is used for backhaul:
- Copper, with its limitations in capacity and reach;
- Fiber, which can be prohibitively expensive to deploy; and
- Microwave. To date, microwave has been the technology of choice – an excellent combination of reliability, cost and rollout speed has given microwave the dominant position in mobile infrastructure backhaul.
Think about the technical challenges facing our industry. The key issues that stand out revolve around spectrum availability. This is as true for backhaul as it is for access. Spectrum for backhaul is a finite resource that’s going to be increasingly stressed as LTE and 4G networks roll out. Such issues include:
- Potential conflicts between increasingly congested networks
- Availability of spectrum in the 6.0 – 42.0 frequency bands
- Need for higher modulation schemes in radios to provide capacity
- The ability of antennas to protect those radios from interference
There’s the possibility of going to higher frequency bands, however, these pose their own challenges in cost and installation complexity.
This then begs the question – can spectrum be used more efficiently?
Throughout Europe, the planning of microwave networks has been driven by the various ETSI radiation pattern masks. Antenna manufacturers’ compliance to these masks ensures that when the network is properly designed, microwave links don’t cause interference with nearby links operating on the same frequency band. Current standards specify Class 2, 3, and 4 with the latter providing the tightest performance.
Most networks are planned with Class 3 antennas. That’s because historically microwave antennas can be designed to meet this specification at little or no cost penalty to a Class 2, providing a meaningful advantage in spectrum utilisation without driving up network roll out costs.
Class 4 antennas have been manufactured in the past, but usually at a high price vs. performance ratio that rendered them uneconomical in all but a few exceptional circumstances. With increasing network congestion, the time has come to re-examine the potential performance trade-offs between Class 3 and Class 4 antennas.
Looking at a typical 23 GHz, 0.6 m (2 ft) antenna nodal site, the ETSI Class 3 pattern masks would allow for around eight antennas to be deployed. Using the Class 4 masks, this jumps to around 22 antennas.
CommScope’s latest generation ofSentinel™ microwave antennas meets Class 4 specifications without the cost overhang of the original models. Their performance characteristics and enhanced immunity to interference also allow links to be planned using smaller antennas than would be previously possible with Class 3 models, likely delivering a direct capital expense saving to the operator. Further operating expense savings can be generated both in spectrum costs and in tower real estate costs. With space on towers costing upwards of $100 per foot per month, the ability to use a 2 foot antenna, instead of a 3 foot or 4 foot antenna, has the potential to save substantial amounts of money on an ongoing basis.
As we move forward, I would suggest that frequency is not so much down to spectrum availability, but more in the way in which it is used. With good link design and the latest generation of antennas, such as Sentinel, there’s still a lot of scope for link deployment in the 6 – 42 GHz spectrum bands.
Are you rethinking microwave backhaul?