Sprint Opts for Microwave Radio over Fiber for Backhaul

Sprint-opts-for-microwave-over-fiber-in-backhaul-Aviat-Networks-March-10-2016In late January and into February 2016, a big tumult ensued when Sprint announced that it would begin to move its mobile backhaul strategy from one based on leased fiber to another based on owned microwave radio. The story first ran in technology news site Re/code and quickly got reposted with additional commentary by FierceWireless, Wireless Week and others, and which was reiterated this week in RCR Wireless.

While the breathtaking headlines about reducing costs by $1 billion caught most people’s attention—primarily through reducing tower leasing costs and not using competitors’ networks—lower down in the copy came a potent reminder from Sprint about the economic benefits of microwave radio. It also highlighted the fact that backhaul has entered a transitional period (see article end for more on that).

Most of that $1 billion that Sprint seeks to save comes by way of moving away from AT&T and Verizon fiber backhaul networks. You might think that Sprint would build its own fiber network instead. But that would take too long and still have an exorbitant price tag associated with it. It’s a function of both out-of-pocket capital costs and embedded lost opportunity costs. Bottom line: laying fiber connections is expensive and slow. Putting up a network of high-speed, broadband microwave relay towers is quicker and easier.

Fiber cost per mile: high in the country, higher in the city

As everyone already knows and previously mentioned here, fiber optic technology has long enjoyed a lofty price tag with costs coming in at $30,000 to $70,000 per mile for rural communities depending on topography, climate and other general deployment conditions. But if you consider the costs for fiber backhaul in the urban cores of cities, the cost per mile can become astronomical with total cost per mile topping out at a $250,000 per mile, historically.

Granted, like other technologies, the cost of fiber has come down since those halcyon days when DWDM first appeared. But even today, for equipment and fiber alone the projected 2016 urban backhaul cost per mile baselines at $50,000 according to an industry filing with the California ISO* (California Independent System Operator). However, this figure does not include costs for installing fiber in underground conduits. Those costs need to be calculated on a per project basis. But in a separate industry filing* it states that total urban fiber backhaul cost per mile ranges from $99,000 to $134,000 (*Excel spreadsheet opens in separate window).

Backhaul in transition

Even while the perennial debate about microwave vs. fiber rages on, the backhaul market has undergone change and transition. In the small cell space for urban settings, E-band radios have taken root for backhaul. For example, it is being considered for microcells, educational institutions and large enterprise campuses. This white paper explores these urban backhaul ideas more in-depth.

Others have proposed exotic solutions such as bonding E-band and traditional microwave radios to expand capacity and ensure transmission distance. That’s an interesting proposal but something that would need closer investigation and verification as to its feasibility.

Many more have talked about Software Defined Networking (SDN) for network forwarding and control functions in the backhaul. However, SDN remains a work in progress. For example, one industry research survey found that only about 30 percent of operators would deploy SDN by 2018 or later with a shade over 60 percent of operators saying SDN deployment was only in the evaluation stage or that they were unclear about the timing.

Surely, SDN will exist in the backhaul at some point, but wireless network operators need to leverage existing equipment and protocols now to get to that future state. Inasmuch, IP/MPLS can provide the solutions to today’s backhaul challenges as well as provide a transitional technology for graceful migration to SDN.

Impetus for IP/MPLS

Truly, IP/MPLS remains the keystone technology offering for mobile backhaul, ready and available for backhaul-to-access applications. Mobile cell sites are transitioning from simple base stations for voice and low-speed mobile data access to multi-functional hubs for delivery of new services, multimedia-rich content and broadband data access.

IP/MPLS provides backhaul networks the ability to handle data with particular characteristics in a smart manner—such as video over low latency routes—and can therefore deliver a wide variety of services from multi-functional hubs very efficiently. This white paper can tell you more about the case for more intelligent Layer 3 functionality at the cell site.

-Derek Handova
Corporate Marketing
Aviat Networks

Comments

    Hugo van Wyk March 25, 2016/ Reply

    I’ve come across this article on LinkedIn, and have commented on it – much in the same way. Specifically E-Band availability as well as the tremendous advances in microwave backhaul technologies such as MIMO, XPIC and high modulation capabilities – the latter referring to the E-Band in particular. These are the obvious “flags” to consider when evaluating a network plan. My recent exposure to FTTH/FTTB implementation has re-confirmed my personal opinions – those stated above. There is, in my mind, no reason to consider fiber for FTTB/FTTH from all aspects – time, cost, application and in some cases reliability and security. From a reliability and security point of view South Africa faces the reality of cable theft and poor infrastructure planning that leads to disruption of services. And yes, fiber gets mistaken for copper. From an application point of view I often ask myself how much bandwidth do the service providers what to deploy to the individual FTTB/FTTH customer. A well planned wireless network that utilises a combination of new technologies will certainly allow satisfactory performance for these customers, from both a backhual and last mile perspective. Time and cost factors do not even be mentioned – anyone will be able to understand the implications. Finally, the wireless option all boils down to proper network planning to ensure performance – obviously. The problem is that more often than not it doesn’t happen. Fiber, as I have stated in my LinkedIn comment, will always have its place in a network. Find that place and implement it appropriately – stop digging up our pavements, roads and gardens for no reason at all!!!

    Aviat Networks March 29, 2016/ Reply

    Thanks Hugo,

    We obviously agree!

    Best regards,
    Aviat Networks


Leave a Comment

Subscribe to our newsletter