Poojan (Wagh) Blog

Requests for comment

Archive for the ‘technology’ tag

Mesh WiFi Recommendations

without comments

With many people having work-form-home constraints, I have received a few questions about at-home WiFi equipment. I have gone through dozens (dozens, I tells ya) of different WiFi systems in my house: Buffalo, Linksys, D-Link, Western Digital, TP-Link, D-Link (again). Some of this was my own doing: when things went from 802.11g (WiFi 3) to 802.11n (WiFi 4) to 802.11ac (WiFi 5), I upgraded my setup. (I will at some point switch to WiFi 6, but probably not for another 5 years or so.)

Here’s what I have learned along the way:

  • The best setup is to have multiple access points, each having a wired connection back to the main router. This is what I call wired backhaul.
  • You do want 5 GHz WiFi (especially 802.11ac AKA WiFi 5). It does not penetrate nearly as good as 2.4 GHz WiFi. But (because 2.4 GHz travels farther and because there are fewer channels), 2.4 GHz is more congested: your neighbor’s WiFi is interfering in your house on 2.4 GHz.
  • Powerline Ethernet connections are not rock solid. In some cases, they are better than WiFi, but it really depends on your house construction (as does WiFi).
  • Extenders don’t work. There is most likely an incompatibility with your router and some other manufacturer’s extender. It’s better to go with a centrally-managed system made from one brand (AKA a mesh system).

If you’re in the market, I recommend a mesh WiFi system with dedicated backhaul. What is dedicated backhaul? It is a separate channel and separate radio that can do the meshing. What that means is that each WiFi access point (each satellite that you put in your house) has two radios: Radio B connects back to the main router. Radio A is what services your laptop/desktop/Roku/cell-phone/etc. (I call laptop/desktop/Roku/smartphone the user traffic.)

This dedicated backhaul means that the satellite (Access Point) can service user data and relay it back to the main router simultaneously. However, there’s a problem with distancing: you still need to put the satellite in the range of the main router. Here’s are a couple of crudely drawn pictures showing the problem and solution:

Problem Case: the satellite (Access Point) is placed too far away form the main router; it cannot receive/send a signal to the main router.
Mesh Case: the satellite (Access Point) is placed within the wireless range of the main router. It can connect and create its own signal farther than the original main router.

This is a little worse than what I do myself, which is wired backhaul:

Wired Backhaul Case: the satellite (AP) is connected with a wired (Ethernet) connection to the main router. So, it does not need to be within wireless range of the main router.

What’s the difference? In the mesh configuration, you have to place each access point within the range of another access point. In the 2nd picture, you can place access points outside the range of each other. And, in fact, it’s preferred that they don’t overlap. (Non-overlap causes less interference and makes it clearer for the user device which network to connect to.)

Now, what if you set up things like so?

Daisy Chained Case: Traffic from rightmost access point must pass sequentially through each access point to the left to make it to the main router

Each radio can hop as many times as necessary to get to the main router. So, it’ll probably function. But it won’t work well. The reason is that AP1 inherits all the traffic from AP2 and AP3. This multi-hop adds delay to user traffic. Delay causes slowdown: both perceived slowness and actual reduction in data rates. This setup can also over-burden the satellites, because they are sending traffic in both directions on their backhaul radios. Finally, the backhaul channel can get congested from the proximity of multiple satellites.

I have simplified things a bit in the above diagrams: there are really 3 networks: the main router’s user network (blue), the satellite user’s network (green), and the wireless backhaul network (shared between the main router and sarellites). But the wireless backhaul on each device will have roughly the same extent as the blue/green network, so I didn’t clutter the pictures separating that detail.

From a user’s perspective, there is only one network in all of the above. Both the main router and satellite will appear to have one WiFi network. The user’s laptop (for example) will switch between the blue and green networks as necessary. The technical jargon for this is that both the blue and green networks should have the same WiFi name (SSID) and password (security credentials), so the user’s devices will see them as the same. They will be on separate frequencies, though, to prevent self-interference of the radio signals.

Now, for my recommendations. Most consumer solutions are primarily made to work as a mesh, not with wired backhaul. While you can do wired backhaul, it’s not something that really comes out-of-the box. So, I’m not recommending it unless you know what you’re doing. (In which case, you probably wouldn’t be asking me for advice.)

To plan this out, you need to know how far your WiFi propagates. If possible, you’ll want your main router somewhere centrally located. And the access points (satellites) concentrically located. For example, if you have a 2-story house with basement, put the main router on the 1st (ground) floor. Put a satellite in the basement and one satellite on the 2nd floor.

But where on each floor is a trickier question. If you place satellites directly above and below the main router, they will be as close as possible. But maybe that’s not where you have an outlet. And in some cases, the routers and satellites will be optimized to broadcast horizontally but not vertically (not directly above and below).

Really getting to the bottom takes a little bit of measurement. If you have an Android device, I recommend using WiFi-Man by Ubiquiti to map out your current situation. Start near your router and walk away. Go upstairs, downstairs, etc. You will see the signal change. You can get a crude map for where the signal propagates and where it doesn’t. Based on this information, you can decide where to place your satellites. And hopefully, the new router will have around the same radiation pattern as your old one. (There is an Apple version of this app, but I do not believe it has the same signal strength function. A google search for similar apps showed an iOS app that had too few reviews for me to recommend it here.)

I should say at this point: I don’t own a mesh system. So, I had to resort to some web researching. You should take the following advice with a grain of salt. A lot of the below is taken from a WireCutter article on Mesh WiFi. This article also corroborates my experience above. In fact, you should at this point read through the article for more detailed testing on all the mesh options out there.

So if you are in the market for a mesh WiFi solution, which one should you get? It depends on whom you trust.

WireCutter rated Eero the best. But Eero is owned by Amazon, so consider whether you trust Amazon. Synology also has a good name with tech enthusiasts; and it has a good upgrade path. WireCutter considers it their 2nd best WiFi router (outside of mesh), and you can add up to 6 mesh devices later; that said, WireCutter says the setup is more complicated than Eero.

If it were me, I would probably go with Synology. They are well-known in other network device markets. They also have a good track record of updating their routers (for security, etc). While the setup may be more complicated, I like that you can add more devices in the future.

I have seen many other tech-enthusiasts go with Google/Nest WiFi. (I bought my WiFi equipment from someone who was selling it because he bought the Google WiFi solution.) But Google has a history of dropping products, and I worry that WiFi could be next. Finally, WireCutter said it was behind Eero in performance.

Written by PoojanWagh

June 7th, 2020 at 9:56 pm

Posted in Desktop Computing

Tagged with ,

Technology Trends

without comments

One can see a distinct trend of phone screens getting larger. For example:

Left: HTC EVO 4G; Middle: Galaxy Nexus, Right: Nexus 4

Left: HTC EVO 4G; Middle: Galaxy Nexus, Right: Nexus 4

On the left is an HTC EVO 4G (circa 2009), a Galaxy Nexus (2011), and a Nexus 4 (2012). Based on these observations, I’ve extrapolated what the next-generation wristwatch will look like:

Left: Timex; Right: future

Left: Timex; Right: future

The one on the left is a Timex (circa 2008). The one on the right is a prototype.

Written by PoojanWagh

April 15th, 2013 at 1:34 am

Posted in Humor

Tagged with ,

I’m Leaving Motorola

with 2 comments

I’ve accepted a position elsewhere. My last day of employment at Motorola will be Nov. 21st. Before I get into where I’m going, I’d like to reflect on / point out a few things with embedded humor:

  • I’m not dissatisfied with my position at Motorola. Indeed, I am in an ideal environment for someone who both likes to invent and innovate, and likes to see things ship. The role in which I found/positioned myself allowed me to investigate some really new technology, without the ridiculous schedules that plague market-reactive engineering environments. At the same time, the stuff we made actually gets shipped, tested, and developed further. It impacts customers.
  • I have learned how to present and explain things to external customers that aren’t up to speed with the daily tasks of the project. I’m still not great at it, but I am comfortable doing it–and I’m better at it due to mentoring by some amazing communicators.
  • I’ve learned a lot about making decisions and plans (strategy). It is always the case (and always going to be) that we can’t do everything. Figuring out what not to work on can be as essential as figuring out what to work on. It’s important to have some criteria guiding these decisions. Goal-setting is important to decision-making. You don’t have to write down the goals, but have them.
  • I’ve learned that it’s always better to be pleasant and kind about technical disagreements. Okay, all disagreements. Even to the people that bug you. They usually think they’re helping. Sometimes, the other guy/gal is right. Okay, more than sometimes. It’s very easy to think that what you’re doing is optimal, because you’ve already laid out the plan and can’t see the other alternatives. An open mind is essential.
  • At the same time, don’t spend time arguing with others if you’ve convinced yourself. There’s always going to be things you could do better. Accept that what you do won’t be perfect–and other people will point it out, but it wasn’t their decision to make, was it? You’re in charge for a reason. Take ownership of what you do, and you will do it better. Get someone else to write the project report, though.
  • The most important thing is the business. We’re not here to have fun. We’re here to make money for the company. Almost all the time, you can have both. Question if what you’re doing is what is best for the customer/department/business. You may have to do some things that aren’t fun, but most of the time, success will follow–and that’s always fun.
  • I have had the honor of being a member of very passionate groups that are doing amazing things with technology. These amazing things don’t get realized all of a sudden on a glorious day. The problems are so complex that one ends up solving very small problems for a long period of time to get incrementally closer to the goal. It’s hard work, and believing in the goal is helpful. Having a management team (and customer) that understands the labor of research is essential. Celebrating small advancements (for example, as feedback to your peers) makes for good teams.

I’m not leaving because things are bad at Motorola. In fact, it is the focus on technological work that makes me want to expand into more areas of technology. If I wasn’t happy at Motorola, I would probably be getting my MBA and be considering business instead. To all executive recruiters: I am not ruling the MBA out.

Where am I going? To a private hedge fund in Chicago. Why? Because:

  • I want a change. I want to learn something new. I want to broaden my knowledge.
  • I like the environment. Working in the financial industry is good. Working with technology is good. However, it’s more important to me that the firm invests in their employees. They’re willing to hire a person–okay, it’s me–because they think he’s smart. I like their attitude toward people.
  • The people I know at the firm are some of the most ethical and just people I know.
  • This opportunity isn’t going to come around again.
  • Free lunch.

Will I be a financial analyst? No.

Will I have stock tips? No.

What will I be doing? I don’t know: stuff. Smart stuff. Mostly software initially.

Is this change necessary? Absolutely not. However, there’s no reason to wait until change is necessary to develop and grow. If you–well, not you specifically–embrace change on your terms, rather than the terms of your environment, you get to pick how you develop, and you get to pick your strengths. There’s a school of thought that states that evolution picks convenient solutions, not optimal solutions.

I sincerely plan to stay in touch with the people I’ve met at Motorola and continue to keep up with my friends from Freescale. This blog is a good way to contact me (see voicemail page to the left). Or, consider subscribing (RSS/Atom or Email).

Written by PoojanWagh

November 7th, 2008 at 4:26 pm