The widespread popularity of Wi-Fi and the continually increasing volumes of traffic on Wi-Fi networks mean that Wi-Fi spectrum needs will need to be planned carefully over the next 5 to 10 years. The Wi-Fi Spectrum Needs Study Final Report was released in February 2017 by its authors, Quotient Associates Ltd, to the Wi-Fi Alliance, with forecasts and suggestions for what will be required. Bandwidth, latency, congestion, and possible inter-Wi-Fi network interference are all topics under consideration, and each has its own type of impact on the Wi-Fi communications of the future.

An Extra 500 MHz to 1 GHz of Spectrum for 2025

Current data and estimates from the Cisco Virtual Networking Index and Maravedis indicate 40% annual growth of consumer Wi-Fi traffic, leading to a doubling of traffic over the period 2015 to 2020. Current Wi-Fi communication bands are located at 2.4 GHz, 5 GHz, and 60 GHz, of which the 5GHz range is the most targeted for reasons of availability (the 2.4 GHz range has been exhausted) and performance (the 60 GHz range offers higher data rates but lower propagation). Using extrapolation, the report suggests that for 2025, regions will need to allocate a minimum of between 500 MHz and 1 GHz of additional spectrum within some suitable range (for instance, 2 to 10 GHz) to meet projected increases in traffic.  One of the key recommendations of the report is to ensure that any expansion of the spectrum allows enough contiguity to construct the wide or very wide communications channels (160 MHz or more) needed to send tomorrow’s volumes of content back and forth with ease.

Business Needs are More Modest than Residential Needs

The amount of additional spectrum (communications frequency range) varies by geographical area and by type of usage. Residential usage will have the greatest need, as users employ ever more powerful computing devices to send and receive ever larger quantities of content. By comparison, office traffic volume is estimated to be a quarter of consumer traffic volume, and data usage outdoors is generally much lower than indoors, perhaps by a factor of 10. Usage in areas such as shopping malls or in individual venues offering guest Wi-Fi may offer more comfort to users than strict outdoors usage, but still tends to favor mobile devices for Wi-Fi communications with less voracious demand for content than larger screen laptops and PCs.

Taking Busy Hours into Account

Residential and business usage also differ in terms of busy hour usage. Business and retail outlets may spread usage out more evenly during a day, whereas residential usage shows a clear busy hour pattern concentrated into the hours of 7 to 11 in the evening. As Wi-Fi spectrum needs must be met in a way that allows not just quiet times but also busy hours to be managed without degradation of performance for users, overall residential spectrum needs rise yet again.

Changing Wi-Fi Standards

Past Wi-Fi networking standards have already been superseded, because they could no longer provide the performance that users demanded. The same will happen to today’s standards, as technology progresses. To make the most of existing Wi-Fi channels, basic smartphones will change from 1×1 (one input and one output antenna) to 2×2 antenna devices, while premium devices are likely to become 4×4 antenna devices. One of the latest versions of the Wi-Fi networking standards, IEEEE 802.11ax, is set to allow up to 8×8, but clearly both devices and Wi-Fi network access points (Aps) must support such a maximum for it to be of use. Meanwhile, today’s 11n standard is limited to a 40 MHz channel width, even if later versions are likely to extend to 80 and 160 MHz (see above).

The Impact of LTE

The Wi-Fi Spectrum Needs report also finds that mobile computing devices are likely to dominate in mall and external environments over PCs and raises the question of LTE (long term evolution) communications as an alternative communications channel. Today Wi-Fi carries 80% of the total traffic in these environments, leaving 20% for LTE. If LTE increases in popularity, the report suggests it will be by taking traffic away from Wi-Fi, rather than by increasing the overall volume of traffic.

Planning the Wi-Fi Network of Tomorrow

Finally, the report offers pointers for planning future Wi-Fi networks for the best performance depend on the type of usage. In some cases, maximum speed will predominate, for example for video streaming. In other cases, latency may be a critical factor, for instance for voice applications. Elsewhere, congestion levels may spoil the user experience. As a rule of thumb, a higher concentration of Wi-Fi access points will tend to alleviate these problems, although possibly at the expense of some interference between the Aps themselves. To offer a good user experience into the future, Wi-Fi network design goals suggested in the report include being able to carry a minimum of 95% of the traffic offered and to keep AP loading below 70%.

However, the report does not address however the developments of LTE in the 5GHz band not the various schemes of spectrum sharing such as CBRS  which will have a profound impact on how both licensed and unlicensed bands are used. We will be writing about spectrum sharing very soon.

Wi-Fi Spectrum Needs Study Final Report to Wi-Fi Alliance, February 2017, © Quotient Associates Ltd.