6 GHz Spectrum and Wi-Fi 7 Driving the Demand for Greater RF Filter Complexity

The rapid expansion of the available unlicensed spectrum in recent years has enabled great leaps in Wi-Fi throughputs and latency rates, but has also heralded significant challenges for Radio Frequency (RF) filters to contend with. The demand for greater RF filter complexity is being boosted further by the increasing regional variations in spectrum accessibility and power level regulations, the coexistence with 5G, and the new technical features of Wi-Fi 7, notably Multi-Link Operation (MLO). New filter designs and a greater number of filter derivatives are helping to address the challenges, but further advancements are required in order to fully harness the full power of the newly available unlicensed spectrum in all markets and the promised technical improvements of Wi-Fi 7.

RF filters are vital components in all Wi-Fi products, and are available from a range of vendors, including Akoustis, Broadcom, KXcomtech, Murata, NXP, Qorvo, Qualcomm, and Skyworks. In simple terms, the role of RF filters is to enable Wi-Fi devices to access the correct spectrum by accepting or rejecting selected RFs. Given the rapid expansion of the available unlicensed spectrum in recent years (with the unlocking of the 6 Gigahertz (GHz) band in the United States essentially doubling the unlicensed spectrum in the country overnight), it should come as no surprise that the market demand for greater RF filter complexity has also increased. While the unlocking of the 6 GHz spectrum band promises to revolutionize performance by delivering greatly enhanced throughputs and latency speeds, it also poses a particular challenge for RF filters, as the distance between it and the legacy 5 GHz band is only 110 Megahertz (MHz) wide. Without an adequately designed RF filter, a portion of the UNII-5 band (5925 – 6425 MHz) will be clipped during the transition, reducing the usable band. Thus, RF filter design becomes essential in ensuring Wi-Fi 6E- and Wi-Fi 7-enabled equipment is able to effectively deliver on the promises of the 6 GHz spectrum.

Yet spectrum enlargement alone is just half the story; matters are further complicated by the fact that each nation has its own regulatory body governing areas, such as Wi-Fi frequency availability and power levels. While the world’s two largest global regulatory authorities (the U.S. Federal Communications Commission (FCC) and the European Union’s (EU) European Telecommunications Standards Institute (ETSI) do offer some degree of international guidance, each nation typically tailors its regulations to its own unique country’s needs. For example, they might reserve certain segments of the spectrum for incumbent users, or restrict outdoor power levels to a greater degree than the FCC. The fractured nature of spectrum management is particularly acute in Asia, where the regulatory bodies of large nations in the region have chosen highly divergent paths. This is perhaps most evident when considering the case of the four largest economies in Asia: South Korea has adopted the entire 6 GHz spectrum band (5925 – 7125 MHz); Japan has already allocated the lower portion of the band (5925 – 6425 MHz) and is considering the upper portion; India is still debating how much of the 6 GHz spectrum it will designate for unlicensed use; and China has chosen to reserve the entire spectrum band for 5G cellular use. To meet the specific needs of each market requires a greater number of filter derivates.

There are numerous reasons why it is important for Wi-Fi devices to leverage all of the available spectrum in a given market. The most obvious reason is that if a device cannot access the entire available spectrum, or is unable to conduct the fast transitions necessary to avoid band clipping, then the full potential of the device cannot be realized. Another critical, but often overlooked reason is that regulators often adhere to a “use it or lose it” approach to spectrum allocation, meaning that if it is not used by Wi-Fi devices in the market, the regulator will choose to reallocate that portion of the spectrum to another party. Thus, the full utilization of the 6 GHz band is important if the Wi-Fi industry wishes to retain the rights to access to the band.

There are a variety of RF filter types that are best suited for different applications, and for devices that must operate across the 5 GHz and 6 GHz bands, with Bulk Acoustic Wave (BAW) filters the optimal filter type. Original Equipment Manufacturers (OEMs) and Original Device Manufacturers (ODMs) designing multi-band Wi-Fi 6E and Wi-Fi 7 Access Points (APs) should, therefore, be incorporating BAW filters into their designs. There are many reasons why BAW filters are the best suited: they offer a high Q factor (meaning less loss and rounding of filter corners), low insertion loss (reducing the loss of signal power from the insertion of the component), better power handling capability (essential for the higher-power requirements of the 6 GHz spectrum), excellent isolation (necessary for separation from and coexistence alongside numerous other signals), superior temperature stability, and a reduced size (essential for the fast growing mesh Wi-Fi market). Some of the leading RF filter vendors have developed proprietary technologies to augment the function of BAW filters. Examples include Akoustis, with its XBAW technology designed to address the most stringent frequency selectivity requirements, and Qorvo, with its bandBoost BAW filters that can maximize channel isolation in multi-band mesh networks. These two companies are examples of vendors that produce modular RF components, which stand in contrast to those who create end-to-end designs. Qualcomm is an example of the latter, with its recently announced Wi-Fi 7 Front-End Modules (FEMs) co-designed to incorporate the company’s ultraBAW filter technology. OEMs should assess both modular and end-to-end Radio Frequency Front End (RFFE) components to determine which is best suited to their needs, assessing areas such as performance, flexibility, customization, and value.

The one drawback of BAW filters is that they take an extended time to develop. Dielectric ceramic filters are a quicker alternative that can also manage high frequencies, but they cannot match BAW filters on size and temperature stability. In some circumstances, this can be an acceptable compromise; for example, if an OEM wishes to quickly get a product to market and is not concerned by the size or need for large fans for ventilation. This was the case with some early Wi-Fi 6E routers and Wi-Fi mesh networks, which prioritized speed to market over aesthetics, size, and noise levels. ODMs with similar speed to market goals may wish to investigate dielectric ceramic filters, but for the majority of residential routers and mesh nodes, for which aesthetics and size are important, BAW filters remain the best suited filter type.

The arrival of Wi-Fi 6E spurred investment into the capital-intensive processes to create RF filters operating in the 6 GHz frequency and at higher power levels. The total time to develop the RF filters operating in this new spectrum took several years, and the price premium on Wi-Fi 6E filters is currently approximately 50% above that of Wi-Fi 6. Fortunately, now that the manufacturing processes for RF filters in the 6 GHz spectrum are already well established, future incremental increases in spectrum coverage or bandwidth frequency in the filter design will take significantly less development time. The development cycle of new filter derivatives can be expected to take roughly 6 to 9 months. Yet, because of the large variation between spectrum allocation globally in both the 5 GHz and 6 GHz bands, there is an increased need for further narrowband variants, as certain markets require filters that can maximize the available channels for use.

Additional advancements to RF filter designs will be required as the industry transitions toward Wi-Fi 7. One particular area that RF filters will need to innovate for is the Multi-Link Operation (MLO) feature of Wi-Fi 7, which allows for the aggregation of different radio links of multiple frequency bands. From a design point of view, additional diplexer components will be necessary to enable MLO. Early RF filter designs will struggle with challenges, such as heat dissipation, which will inhibit, to an extent, the performance of the MLO. Only after vendors have become more adept at designing for MLO will these issues be ironed out and the MLO will improve. As MLO is one of the headline features of Wi-Fi 7, Wi-Fi equipment ODMs should take note of the above.

Aside from the development timeline for new filter derivatives, equipment vendors need not be overly concerned with the availability of RF filters. In contrast to the situation for chipsets, RF filters are not currently experiencing any major supply chain challenges. This is due to the fact that they are not relying on any components that are in short supply, and also because many major RF manufacturers, such as Akoustis and Qorvo, are in control of their own foundry, so they are not facing the competition for capacity that many chipset vendors are currently tackling.