Quantenna is showing off two very interesting wireless products at CES, an 8×8 MU-MIMO radio and a software suite. As you probably realize both are aimed at different markets but both are interesting in their own way.
First up lets talk about the MU-MIMO radio, a true 8×8 chip that can support up to eight simultaneous 802.11ac streams. If you aren’t familiar with MU-MIMO and other 802.11ac Wave 2 technologies, read this. As we mentioned last time we wrote about the Quantenna 8×8 system, they are calling it a “10GbE” and we could not figure out how they got to that number. The answer is that the 8x 867Mbps gets you to the high 6.xGbps range then you add a few other wavelengths and technologies. Before you call this BS, Quantenna isn’t counting like some less ethical companies, this one is real.
Take that base 6.xGbps number and add in the 802.11n bandwidths to the total. Why is this not a scam? Because the Quantenna 8×8 solution can do that simultaneously with the 8 802.11ac channels. On top of that you can add in some older specs for a bit more bandwidth, all at the same time. In short the 10Gbps number can actually be attained, it is actually the throughput of the radio in question.
On top of this there are also some proprietary modes that Quantenna has, presumably for base station to base station repeater work, that could up those bandwidth numbers too. That said Quantenna doesn’t need to count these to get to the 10Gbps throughput number. Until physics are bent a little more by creative RF engineers though, we won’t get near that number for a single stream. Quantenna will be demoing their 8×8 silicon at CES, if we can get to them we will bring you more details and possibly a few pictures.
Along side that is a new software suite and tool set called Maui. This was an internal Quantenna tool for remote diagnosing and troubleshooting of field trial silicon deployments that grew into something useful. Sensing an opportunity to fill a market need with something they had in-house, Quantenna turned Maui into a tool for ISPs and other owners of large wireless networks. In essence the same tools that the engineers used to diagnose, fix, and predict problems on the early deployments can now be used for end-user support.
What does Maui do? It has two parts, an agent that runs on the Wi-Fi base station that collects data and a cloud based analysis suite. The agent is pretty simple and lightweight, it sits on your router and listens for errors, reboots, channel changes, interference, and other problems that users see as downtime or reduced bandwidth. More importantly it does not just log failures but interference that can result in lower throughput but not outright failure. These are sent back to the main tool suite for ‘big data’ type analysis.
Before we go too far into what it does, the most important question in SemiAccurate’s mind was privacy and the ability to turn it off. We are not big fans of corporate spyware and forced malware, this is the main reason we can’t use Windows and Microsoft products, so Maui set off alarm bells. We asked about turning it off and Quantenna rightly said that it is not their choice, it is up to the Wi-Fi provider to decide what abilities you have and do not have around Maui. This did not fill us with hope, we trust Comcast et al slightly more than Microsoft [0 * 1 = 0].
Luckily we then asked what data Maui collects and all of our fears were eased. Maui collects only very low-level radio data, not user level anything. It operates on OSI layer 1 and occasionally layer 2 which means at most it can see your MAC address. Since this is something you can’t hide and broadcast almost 100% of the time, other than that Maui can’t invade your privacy. If you really care and have a grown-up OS, IE not Windows, you can actually change your MAC pretty easily. Luckily for the Comcasts and Verisons of the world, they have their own sophisticated tools and DPI boxes to invade your privacy in a much more antisocial way, it is a lucrative business. Unfortunately for them and fortunately for users, Maui looks to be more than safe enough for SemiAccurate not to worry about and you shouldn’t either.
Ranting aside once the low-level data is aggregated there is a lot of useful insight that can be gleaned from it. Starting off with location, any ISP or corporate network monitor knows the location of each base station so you can map problems and see what other physically close radios are seeing. Is the problem only at a single location or is everyone in the area having the same issue? Is the problem occurring in a wide area, say around an airport or military installation? Knowing this can go a long way towards finding and avoiding a problem but relying on users to manually report it is a borderline lost cause. Having even unnoticeable degradations logged, timestamped, and aggregated automatically is a big benefit too.
That brings us to the problem of proper reporting, users are pretty bad at this. Instead of a call that says, “Your service sux0rs, fix it now so I can XBox more”, you get channel info, signal strengths, and the things that users can’t tell you easily if at all. And you usually get it before they call if they report it at all. You get everything you need and more by keeping users out of the loop. Luckily the Maui agent is smart enough to report only the problems it needs to, not everything. Since this data is low-level and specific, the bandwidth it uses is between minimal and ignorable especially if there aren’t actual problems to diagnose.
Armed with this data, Maui can offer predictive analysis to users so tech support doesn’t have to wait for you to get annoyed and call, they can call you. Actually they are more likely to simply push a fix to your Wi-Fi base station in the background, how they do this can be acceptable or not.
If you do call in, Maui has the potential to allow the flowchart monkey on the other end to say, “Your radio is set to channel 6 which three of your neighbors and the local wing of F-16s also use. You have been seeing this issue for three months since your router rebooted and changed channels to avoid a strong signal that has since stopped occurring. You might want to go to channel 10 where there is clear air to avoid this problem, shall I do it for you?”
That is what geo-data, historical data, multi-site aggregation, and predictive analysis can do for both users and ISPs. Without any single piece of the puzzle the entire whole would be impossible, together they are just really hard to pull off. One of the main benefits that Quantenna claims is not just logging and reporting but predictive analysis and actionable recommendations based on the analysis. Instead of a red blinking light at 1313 Mockingbird Lane, there is a red blinking light at 1313 Mockingbird Lane with a problem description and a suggested fix. In some cases the fix can be automated so the user never knows there is a problem, especially if the problem happens at regular intervals it can just be avoided.
From the sound of it, Quantenna has pulled it off with Maui and done it the right way for both users and the customers, in this case mainly ISPs. Unfortunately Maui will not be a tool that end users can directly use but SemiAccurate can see it being integrated into a large ISP’s self-help portal or other sadistic user frustration mires. Better yet Maui is not going to be restricted to Quantenna chips and radios, it will work with any Wi-Fi radio that an ISP wants to deploy it on so there are few if any barriers to deployment. We can’t say much more until we get a chance to play with it which will hopefully happen in the next few days. Stay tuned.S|A
Latest posts by Charlie Demerjian (see all)
- How is Intel solving their 14nm capacity problems? - Jun 13, 2019
- How big is AMD’s new Navi GPU? - Jun 7, 2019
- Intel kills off a (minor) product line - Jun 7, 2019
- A look at Intel’s Ice Lake and Sunny Cove - Jun 5, 2019
- Leaked roadmap shows Intel’s 10nm woes - Apr 25, 2019