LAA-LTE: Looking beyond coexistence
The first time I heard about LTE in the 5 GHz license-exempt band, my reaction was: “Why do we need it? Wi-Fi works well, it’s cheap, and it’s in every device.”
The growth in cellular data has been stunning, and we are only at the beginning of a transition to an environment with massive wireless connectivity through IoT and personal mobile devices. But the technology that taught us about wireless broadband and first made us believe smartphones are worth buying was 3G only in part – it was mostly Wi-Fi. The air interface that carries most wireless traffic today is not cellular – it is Wi-Fi. According to Cisco VNI, Wi-Fi carried 55% of Internet traffic in 2013, cellular 4%. By 2018, Cisco expects Wi-Fi to carry 61% and cellular 13% of Internet traffic.
So why do we need a technology that competes with Wi-Fi and is pretty much guaranteed to lose the battle against such a formidable opponent? Keep in mind that LTE unlicensed (LTE-U, also called LAA-LTE, for licensed-assisted access LTE because it works in conjunction with licensed bands) shares the 5 GHz band with Wi-Fi (as well as other technologies), and hence the two technologies compete for access to the same spectrum assets. As a result, Wi-Fi traffic will reduce the bandwidth available to LTE-U; LTE‑U traffic will reduce the bandwidth available to Wi-Fi.
At a workshop hosted by Qualcomm last week, I did find an answer to my initial question, but I also realized this is not the right question to ask in the first place. The answer I have for the role of LTE-U has three parts:
- From a mobile user perspective, there is little difference between LTE-U and Wi-Fi, as long as the device supports LTE-U and the infrastructure is equally available (i.e., the same density of access points, regardless of whether they are Wi-Fi or LTE-U). At home, users keep using Wi-Fi as always, so there is no change there.
- From an enterprise perspective, LTE-U does not have much of an impact on Wi-Fi networks. LTE-U requires access to licensed spectrum, so deploying it is not an option for enterprises. Depending on their preferences, enterprises may allow mobile operators to install LTE, LTE-U or Wi-Fi within their footprint if they want to, but they do not have to.
- For a mobile operator, however, LTE-U has major implications. Like everybody else, mobile operators have access to the 5 GHz band using Wi-Fi or other technologies (e.g., for backhaul). LTE-U gives them a new way to use that band, and it’s a technology that integrates better with their current networks. There are some differences in the air interface and in the way traffic in managed, but LTE-U is only a slight variation on LTE. Carrier Wi-Fi gives operators the tools to integrate Wi-Fi with cellular networks, but fundamentally Wi-Fi is a device-centric technology, and mobile operators prefer a network-centric one – and that’s what LTE-U is.
Integration of Wi‑Fi and cellular is possible, but it requires more work than integration of LTE-U and LTE. Some operators will prefer Wi-Fi integration, either because they have already made an investment in Wi-Fi or because they prefer using a mature and probably cheaper solution. AT&T in the US is a good candidate for this approach, because it is heavily invested in Wi-Fi. Other operators – e.g., Verizon in the US – may opt for LTE-U to add capacity in dense areas and continue to rely on Wi-Fi for residential and enterprise offload. A third group of operators may use a combination of Wi-Fi and LTE-U to boost capacity even further. How operators will decide remains to be seen, and this will determine the success of LTE-U.
Regardless of what mobile operators decide to do, Wi-Fi position in the market – and in the homes and hearts of users – will remain the same. LTE-U will have to coexist with Wi-Fi, and play nice even in markets like the US and China where this is not required by regulation – if for no other reason than to avoid a confrontational environment over a technology, Wi-Fi, that everybody loves. Standardization efforts at 3GPP confirm the commitment of operators and vendors to developing the tools for a friendly coexistence of Wi-Fi and LTE-U.
But here is why the “Why do we need LTE-U?” and “How can LTE-U compete with Wi-Fi?” questions are not the most relevant ones to ask. Operators do not need LTE-U to use the 5 GHz band (they can do so with Wi-Fi), but LTE-U a valuable tool for mobile operators. LTE-U is not going to challenge Wi-Fi, but it will carve out a separate, coexisting role that is complementary to Wi-Fi in the 5 GHz band. Operators already have access to the 5 GHz band, but LTE-U helps them access to it in a more efficient way – better spectral efficiency, easier network integration, and better traffic load management. A combined use of LTE-U and Wi-Fi has the potential to expand the combined capacity of licensed and unlicensed bands because of the increased efficiency in the use of network resources. In turn, this translates into a collective benefit for Wi-Fi and cellular users.
Beyond coexistence, however, the traction that LTE-U is gaining sends an important message – one that transcends the current tensions in the wireless industry between the Wi-Fi and LTE-U camps. LTE-U shows that, after residential users and the enterprise, mobile operators have finally started to embrace license-exempt bands, not as an alternative to licensed spectrum, but as an expedient way to increase network capacity in dense areas that is complementary – and fully integrated with – licensed spectrum. From the initial suspicion toward Wi-Fi, mobile operators have steadily moved toward leveraging unlicensed spectrum with Wi-Fi offload, but mostly in a passive way, using the infrastructure that was already in place or building a stand-alone hotspot infrastructure. With carrier Wi-Fi and LTE-U, operators can use unlicensed bands as an integral part of their network, and they have shown a commitment to doing so.
From a wider perspective, the use of unlicensed spectrum by operators may have the effect of reducing the value (and possibly cost) of licensed spectrum. More importantly, however, it should encourage telecom regulators to expand license-exempt spectrum allocations. The ability of Wi-Fi to pack more than half of all IP traffic primarily into 100 MHz of 2.4 GHz spectrum, and the eagerness of users, enterprises and service providers to use it, confirm that unlicensed spectrum is a powerful way to increase the utilization of spectrum assets.
