Wi-Fi is omnipresent in our lives yet few of us understand how it works, and more importantly, how to make it work best for us. If you are frustrated with the performance of your Wi-Fi at home or work or want to learn more about Wi-Fi, this should help you first to understand and then in later posts to diagnose and solve most performance issues.
One of the most common confusions is the difference between a byte and a bit. Most of us are accustomed to using and measuring digital files in Kilobytes, Megabytes and Gigabytes (a typical mp3 for instance is around 3 Megabytes, most USB flash drives today come with at least 8 Gigabytes of space). In the same way that 1 Gigabytes is the same as 1024 Megabytes, and 1 Megabyte is 1024 Kilobytes, 1 byte is equivalent to 8 bits.
While most of us are accustomed to using and measuring things in Megabytes, your Internet Service Provider (ISP) advertises their Internet connections in Megabits, an important distinction. For example if you want to download a 3MB file from a 3 Mbps connection, it won’t take 1 second but 8 seconds, or in other terms you would need a 24 Mbps Internet connection to download the same file in 1 second.
Check out these screenshots from Hong Kong Broadband Network (top) and Netvigator (bottom). They are advertising their speed in Megabits; the lowercase or missing ‘b’ is important!
Additionally it is important to remember that these are theoretical max data rates and not guaranteed rates. Why is that? Well not everyone in Hong Kong is going to be using the Internet at exactly the same time. By sharing the infrastructure amongst users you can reduce the amount of infrastructure you need to build dramatically. It is the same with many other infrastructure system such as the MTR, which is extremely busy at 6:30PM when everyone is going home but not so busy at 9PM when most people are already home.
In order for ISPs to more efficiently use and build Internet infrastructure, no one is guaranteed the stated bandwidth at all times, but everyone is guaranteed a connection at all times. By taking advantage of small time differences between when people are actually using their connection (ie. downloading a file or loading a webpage) they are able to dramatically increase the efficiency of the network. This lowers the cost dramatically for the ISP and for you, the end user. It also explains why you might have problems streaming HD movies at 7PM but have no problems at 11:30PM or 5AM, when many people are asleep.
Internet vs. Wi-Fi
Another important thing to keep in mind is that the Internet and Wi-Fi are not the same thing.
- You do not need Wi-Fi to access the Internet (Ethernet cables, LTE)
- Wi-Fi doesn’t need to be connected to the Internet (private network).
The Internet is the network of interconnected computer networks around the world all speaking the same language, TCP/IP. Wi-Fi is a wireless network. It serves as an easy way to wirelessly connect all your devices together so that they can communicate; it also allows you to connect all of those devices to the Internet. As such your Wi-Fi network makes up the first and last mile of everything you do online.
When dealing with many separate systems it is inevitable that different systems are going to differ in terms of their physical locations and connection speeds. These slower connections are known as bottlenecks; the fastest possible rate you can achieve. When this happens, your observable connection will only be as fast as your slowest connection. Take a look at the simplified example below.
- We have a mobile device (iPhone, iPad, MacBook, etc.) connected to Wi-Fi.
- The Wi-Fi has a 100 Mbps internet connection from the ISP
- There are three website that you can go to, Yahoo!, OpenRice and Google.
So what is the fastest possible connection speed I can achieve to Yahoo, OpenRice or Google? In this simple example our bottleneck is right in our first mile, so we can only be as fast as our Wi-Fi connection. The maximum possible speed we can achieve is 40 Mbps. What if we changed out our Wi-Fi router or modified the settings? Would things be different?
Test your understanding by taking a look at the modified example below. What are the fastest possible connection speeds I can achieve to Yahoo, OpenRice and Google? Do they differ? Where are the bottlenecks below?
A Wi-Fi by any other protocol
Did you notice a difference between the Wi-Fi connection speeds in the different examples? What could have caused this? Well, Wi-Fi has been around since 1997, a time when dial-up modems were the most common form of Internet connections in the home. As ISPs offered faster Internet connections, users needed advances in Wi-Fi technology to match the speed and deal with performance and security issues. The result is a wide variety of standards in use today with no obvious meanings and large differences in performance. Below is a quick breakdown of the different standards in use today.
|Wi-Fi Protocol||Year||Frequency||Max Data Rate|
|n||2009||2.4GHz / 5GHz||600Mbps|
|ax||2019||2.4GHz / 5GHz||NA|
*actual Max Data Rates may differ according to hardware.
Now that we understand more about what Wi-Fi is and isn’t, let’s try a quick test. Grab your laptop and put it with a clear line of sight to your Wi-Fi router and follow the steps below to see if you have a Wi-Fi bottleneck.
Do you have a Wi-Fi router bottleneck?
- Connect to Wi-Fi and click here to do a speed test.
Click “Begin Test” and record the download and upload speeds.
- Disconnect from Wi-Fi, remove the Ethernet cable from the back of your Wi-Fi router connected to the port that says Internet and repeat the speed test.
- Compare your results from 1 and 2. If your results from 1 are significantly lower than 2, you have a bottleneck that is impacting performance. This could be a result of old hardware, misconfigured hardware, interference or distance (we will go more into optimizing in a later post).
- It also a good idea to compare the results from test 2 with what your ISP promised you. The speed will fluctuate but it should be extremely close to what your ISP has promised you.
- Reconnect the Ethernet cable you disconnected in step 2 by plugging it back into the port labeled Internet.
That completes the basic Wi-Fi primer for homes and small businesses. For large network events, a more robust setup such as that offered by TAPevents would be needed to guarantee speed and reliability for all users. In my next post, I will explain how you can optimize your network to get better speeds and reliability.