You don’t get too many Herculean battles in communications, so when one comes along, it’s good to pay attention.
In an
article a few months back, I discussed some of the important infrastructure developments from upstart vendors such as Altiostar, Celona, Federated Wireless, JMA Wireless, Mavenir, Parallel Wireless, and Tango Networks being touted along with the move to 5G. Among those developments are open radio access networks (ORANs), centralized RANs (CRANs), network function virtualization, cloudification of cellular infrastructure, network slicing, and private 5G. I also raised the core question: Will the upstarts be able to convince the mobile operators, with their conservative utility mindsets, that the cost and flexibility benefits of such options will be worth the coordination effort and the risk of going with a new supplier?
In that article, I focused on the three national carriers — AT&T, Verizon, and T-Mobile — but a fourth operator, Dish Networks, is the wild card in this deck. For those who don’t recall, as part of the
agreement for T-Mobile’s acquisition of Sprint, the U.S. Department of Justice mandated that Sprint sell certain of its spectrum holdings to Dish. Dish, in turn, committed to the FCC that it would deploy a core network using its spectrum assets and offer 5G broadband service to at least 20% of the U.S. population by June 14, 2022, and to at least 70% of the U.S. population by June 14, 2023. Dish has committed a mere $10 billion in investment to accomplish this.
Dish has now announced that it will introduce its first cellular coverage area in Las Vegas “later this year.” To reinforce the message that it intends to meet that commitment, Dish has trumpeted every one of those leading-edge ideas with plans to deploy them in the AWS cloud with the promise to “achieve agile and cost-effective operations while seeking to redefine the practical applications of 5G.” This is shaping up as a David and Goliath battle if ever there was one.
I have been more than skeptical about Dish’s plans and its potential for success in creating a new nationwide cellular network from scratch, but now we’re going to find out if Dish’s promises hold water.
Are These Technologies Really a Big Deal?
The short answer is, “Yes” — and in a very big way. Transitioning cellular networks to new types of infrastructure mirrors the migration we saw in premises voice systems over the past few decades. But the cellular migration must be done on a much larger scale, with far more complexity, and with more critical interdependencies — i.e., all of these cool new ideas have to work together right out of the gate.
The first step in this transition is virtualization, or the process of transitioning a product from hardware to software that will run on standard hardware platforms. In the enterprise, we accomplished that with the move from PBXs to IP-PBXs. Our old AT&T System 85 PBXs disappeared, and the functionality of those big cabinets became a virtual instance running on a data center server.
The mobile carriers’ switching platform is the mobile switching center (MSC); however, Dish’s vision will take virtualization far beyond that. The carriers have tons of intelligence housed in their internal core networks that interconnect the MSCs and, in particular, in the RANs managing all of the crazy requirements of providing reliable mobile radio connections to millions of subscribers scattered over millions of square miles.
Dish is planning to virtualize and cloudify all of those functions, or, to quote their press release:
“DISH will leverage AWS’s proven infrastructure and breadth of services to deploy a cloud-native 5G network that incorporates O-RAN — the antennas and base stations that link phones and other wireless devices to the network — and the 5G Core — the logical architecture that directs traffic flow within the network. AWS will also power DISH’s fully automated Operation and Business Support Systems (OSS and BSS) that will enable the company to provision and operate its customers’ 5G workloads and monetize its network.”
Not only would Dish be virtualizing the RAN control functions, but also potentially sourcing the physical radio units (RUs) from different vendors, as is a principle of the ORAN concept. Dish recently announced a “large purchase” of RUs from Fujitsu along with an open virtualized RAN (vRAN) software solution from Altiostar and
Mavenir.
Dish sneaks in the biggest revolution further on in the announcement, where it describes APIs that would allow developers to “to engage with data on DISH network attributes such as user equipment latency, bit rate, quality of service, and equipment location.” That sounds a lot like the types of functional business connections to the cellular network I referenced in my
earlier piece. So, Dish is talking about not only changing how they deliver the service, but also radically broadening the range of services they will be able to provide.
ORANs, CRANs, and the rest potentially provide lower cost, more flexible mechanisms for delivering cellular service. These APIs would allow Dish to offer new types of differentiated services, or potentially to allow enterprise customers and app developers access to information about mobile stations that could be used to develop new capabilities or services. Security and privacy are still concerns, but this holds the potential to redefine what the mobile network can do… maybe becoming a mashup between a mobile carrier and a CPaaS provider a la Twilio.
The technology visionaries have dropped a hook in the water with these new concepts for cellular network infrastructure, and clearly Dish has taken the bait — and possibly the first several inches off the rod! The incumbent operators are testing the waters on these initiatives as well, but they already own and operate massive networks with hundreds of thousands of cell sites tuned to deliver a phenomenal radio service, so they can’t just turn in a new direction overnight. Dish does, however, face a completely different, though no less intractable, set of problems in actually delivering service to customers using these novel concepts.
Dish has swept the Double Jeopardy round for industry acronyms, and if they can really deliver things like opening up access to selected parts of the network signaling information through APIs (with appropriate privacy and security protections), the potential impact will go off the charts. Even the ability to know the caller is on a cell phone gives us the ability to communicate with them in SMS text. That in combination with location or motion information could lead to countless new services.
Will It Really Work? Now We Get to Find Out
This unique type of business challenge has come about due to a very peculiar environment. We have three massive incumbent utility providers running networks with book values in hundreds of billions of dollars and a fourth guy, Dish, that is now starting from scratch using a gaggle of new technology ideas and a $10 billion budget. Maybe Dish plans to keep the capital investment down by shifting more of the budget to OPEX.
Despite all this technology stuff, mobile operators are essentially utilities, whose business is to deliver a technically complex service on a large scale with extremely high reliability. This mandate more or less dictates a certain form of management. As a product of that environment (i.e., AT&T alumnus) I can vouch for the fact that this form of management works for delivering the goods, and the reliability requirement demands prudence and caution. However, if you’re looking for bold moves, leaps of faith, or technical risk-takers, you should probably be looking elsewhere.
What Dish is proposing is a new type of utility built on the full embrace of technical ideas in common with what we have seen in other parts of the communications industry. Most of these new technologies Dish is counting on have not been widely proven in this application, and certainly not at this scale. And Dish is talking about virtualizing literally every key element of their operations, from the RAN to the switching, QoS provisioning, and core networking right down to the billing. Did I mention that getting this cellular thing to work at all was a challenging endeavor in and of itself?
I have no doubt unforeseen surprises will pop up, and over time reality will tell us more precisely where each of these concepts works and where they fail. While I am perennially skeptical (that’s what clients pay me for), I do hope that Dish will be successful in bringing at least some of these technical concepts to market. However, if they can truly expand the definition of what mobile carriers can offer, that’s truly revolutionary.
The U.S. has been the driver with most of the ideas and applications that have fueled the mobile revolution. I would very much like us to maintain that lead. If Dish is successful, this could drive the cellular industry in a whole new direction.
