Claims 802.11ax standard for ultradense WiFi should be ready in 2018, with commercial chips the same year
The prototype equipment reached 10.53Gbps in the conventional 5GHz unlicensed spectrum band. Much effort around superfast WiFi has focused on the 60GHz band, which supports high speeds and capacity, though typically over short distances. But Huawei is driving similar data rates in the lower frequencies, with their superior propagation, and is chairing the latest in the IEEE's alphabet soup of WiFi taskgroups, 802.11ax, which is looking at very fast networks especially for ultra-dense hotzones.
In this way, the company aims to put itself in the vanguard of next generation wireless technology development, and it hopes the standard will be ratified in 2018, with commercial equipment in the same year, though likely based - as is WiFi's habit these days - on pre-standard implementations.
The demonstration in Shenzhen is based on Huawei's own technology, since the 11ax specifications are not real yet, but the firm will submit some of its innovations to the taskgroup and so aim to seize the steering wheel in next-wave WiFi. It has been engaged in R&D in this area since 2010, focusing on "innovation surrounding new WiFi architectures that would break the logjam of classical WiFi wideband radio and baseband processing to increase user data rates".
It is using some of the core technologies which are also expected to be prominent in cellular '5G', including MIMO-OFDA, dynamic spectrum allocation, interference coordination and hybrid access techniques, and claims a tenfold improvement in spectral efficiency by combining these tools. MIMO-OFDA is of particular interest to hardware makers, combining large MIMO arrays with OFDM schemes, which divide sub-carriers into separate data streams. Huawei did not specify how many channels were used to get the peak speed, or how wide they were.
The IEEE board approved the 802.11x effort in March after a year of preliminary work.
Start-up will launch first product next month, working with Quantenna to bring 802.11ac to rural service providers
The funding round was led by previous investor New Enterprise Associates, with participation from Oak Investment Partners., taking the start-up's total financing to date to over $38m. The injection also sees Mimosa coming out of stealth mode and promising a product launch next month.
Unlike established carrier WiFi vendors - which have tended to play up the opportunity in urban areas where cellular networks are under strain from rising data usage - Mimosa is focusing first on the traditional market for broadband wireless platforms, the underserved rural regions. It is banking on the fact that demand for broadband, especially in emerging economies, will far outstrip the pace of roll-out of wireline connections by telcos or under national broadband plans.
The Californian company was founded in 2012 by Brian Hinman and Jaime Fink. The former previously co-founded videoconferencing vendors PictureTel and Polycom and home networking specialist 2Wire. Fink was CTO at 2Wire and worked at set-top box maker Pace when it acquired the smaller firm.
Like most carrier WiFi innovators, Mimosa's performance derives from its particular take on MIMO antenna arrays (from which it gets its name). It has been partnering with Quantenna Communications, the pioneer in gigabit WiFi chipsets with advanced MIMO, and the latter's 802.11ac wave 2 silicon will power Mimosa's debut products. Quantenna has consistently stolen a march on Broadcom and Qualcomm Atheros with the latest fast WiFi developments, and recently promised the world's first 10Gbps WiFi chipset, using eight-stream (8x8) MIMO. This will appear in Mimosa offerings next year.
The two start-ups are also cooperating on joint development of enhanced physical layer and link layer technologies.
While Mimosa currently concentrates on 5GHz, the usual band for the latest WiFi iteration, 802.11ac, it believes new unlicensed spectrum will be needed to cope with data demand. Some WiFi vendors are interested in 802.11ad, which supports very high speeds in the 60GHz band, but has limited range. For Mimosa's rural model, range will be important, so it favors the 10GHz spectrum for future WiFi expansion, and has petitioned the FCC to open this up under a lightly licensed scheme (this is being opposed by amateur radio enthusiasts, in particular).
Mimosa claims it has developed high speed equipment in low cost form factors, which will offer 10 times the speed and five times the capacity of current wireless broadband options open to rural ISPs.
"The next great frontier in wireless technology is the ability to transmit independent streams to multiple clients simultaneously. With constantly increasing internet demands, wireless hardware must leverage these innovations to leap forward in speed and client capacity," Hinman said in a statement earlier this year. "Quantenna's development efforts focus on using wider channels, more streams and multiuser MIMO to address the next phase of network demand, making their chipsets a perfect fit for the Mimosa product suite."
By Peter White, Principal Analyst Wi-Fi and digital multimedia
Last week saw a significant breakthrough for US Wi-Fi spectrum in terms of flexibility and capacity. The FCC signed a ruling which increases the radiated power that can be applied in parts of the 5GHz unlicensed spectrum; removes the indoor-only requirements in the 100MHz from 5.150-5.250GHz; and extends the 5.8GHz band by an extra 25 MHz, from a prior 5.825GHz upper limit to 5.85GHz.
Some of these changes are to bring the US in line with other regimes across the world, while others come as a result of two key factors – cable industry lobbying pressure and a period of experience where no real complaints against unlicensed usage have been upheld. Indeed, every case of interference registered in the past has now been attributed to equipment that was either faulty or which did not adhere to the standard, proving that unlicensed spectrum works and is safely managed.
And with that, the FCC commissioners promised to make at least another 195MHz of further spectrum available in future in the 5GHz band, along with the creation of more unlicensed spectrum use in the 3.5GHz and 600MHz bands, the latter to be freed up by broadcasters under FCC incentives.
The initial order allows Wi-Fi equipment to target performance above 1Gbps in operation and as further chunks of spectrum are made available, it could one day double that.
There are a few important things to take away from this move. Firstly that the rest of the world will mostly follow. Wi-Fi spectrum is not unified, but there is a commitment across the globe for regulators to try their best to make it easy to export devices from one part of the world to another, with minimal adjustment. Where this has not yet been done in other parts of the world, it will follow.
Secondly the speed at which this came about shows that those lobbying in favor of Wi-Fi are strong, and the economic benefits of having lots of unlicensed spectrum for Wi-Fi and other technologies are measurable and good for any country issuing more.
Thirdly this sends a message to the cellular community that ‘the water is fine, come on in’. Efforts, led by Qualcomm, to demonstrate 4G in this same unlicensed spectrum is one way to go, squeezing Wi-Fi in its own space. However, if the cellular community were to do this, it would crush down the effectiveness of this new spectrum, by creating interference and by making investment in Wi-Fi less profitable.
This order from the FCC makes it clear that MNOs would be welcome to build out their own Wi-Fi, but then like all Wi-Fi, it needs to be for the benefit of everyone, not a single MNO customer base. The powerful Wi-Fi lobbying community would explode if 4G limited its options to a single MNO in any of the major western territories.