We have long been using WiFi just as a technology to connect my laptop and mobile phone to the internet, but the technology does not stop there. Nowadays, WiFi is also a key part of the Internet of Things (IoT), linking not just our phones and computers, but also many devices in our homes and workplaces. This includes everything from light bulbs that you can control with your smartphone to smart thermostats that learn your schedule. WiFi makes it easy for these devices to communicate with each other and with the internet, making our lives more convenient. By bringing all these different gadgets together, WiFi is opening up new ways for us to interact with the technology around us, making our environments smarter and more responsive to our needs.

In this note, I am going to talk about how WiFi is used for IoT and overcome various technical challenges for the application.

• WiFi IoT Devices
• Where are we with WiFi IoT ?
• Regular WiFi vs IoT specialized WiFi ?
• Technical Challenges to WiFi IoT
• Reference

WiFi IoT Devices

Today, we live in a world where WiFi and Internet of Things (IoT) devices are everywhere. It's not something from a science fiction story. It's real. These devices make our homes, work, and even the outdoors better and smarter. They connect and talk to each other, making things easier for us. Below is a table with many different kinds of these smart devices. It shows us that the future of smart technology is already here. We use these amazing devices every day. They help us in so many ways, from keeping our homes safe to making sure we're healthy.

For the WiFi design, the device type matters less than two properties: where the power comes from, and how much data moves. Let's sort the table by those two. A smart TV, a camera or a smart speaker is plugged in and moves a lot of data, because it streams video or audio. A thermostat, a smart lock or a tire pressure sensor runs on a battery and moves very little. For example, a sensor that reports 100 bytes once a minute needs only 100 x 8 / 60 = 13.3 bit/s on average.

The first group is well served by ordinary WiFi. The second group sets the hard problems of this page. Its devices need years of battery life from a radio that was designed to stay awake, and they often sit far from the access point. The sections below come back to these two groups when they compare regular and specialized WiFi.

• Power source and traffic sort the devices : mains powered and high rate on one side, battery powered and low rate on the other.
• Low-rate devices need little air time : their problem is the energy spent to stay connected, not the data rate.

Where are we with WiFi IoT ?

WiFi-connected devices are everywhere, and they're a big part of our world now. These numbers show us how fast they are growing. We're going to look at just how many smart devices are out there today and what this means for our future.

The chart below illustrates the global IoT market forecast, showing the number of connected IoT devices in billions up to the year 2027. It differentiates between actual data up to Q4/2022 and projected growth beyond that. A Compound Annual Growth Rate (CAGR) is given for two periods: 2021-2022 and 2022-2027, across various connectivity types like 5G IoT, LPWA, and WLAN. The CAGR shows the expected annual growth rate in percentage for each connectivity type. The overall trend indicates a significant increase in IoT connections, with a notable rise in cellular 5G IoT connections.

NOTE : Even though every access technology is growing for IoT, WLAN/WPAN is still dominant technology and the situation would not change.

Source : IoT Business News

• Totals : 14.4 billion connections in 2022, 16.7 billion forecast for 2023 and 29.7 billion for 2027. The labels +23%, +18% and +16% mark the growth rate at three points on the curve.
• WLAN : a CAGR of 21% for 2021-22 and 16% for 2022-27, the same rate as the market as a whole.
• WPAN : 12% and 16%. The WPAN layer at the bottom and the WLAN layer above it are the two largest layers in every year.
• Cellular 5G IoT : 200% and 87%, the fastest growth, but from a very small base, so its layer stays thin even in 2027.
• LPWA : 38% and 27%, the fastest growth after 5G IoT.

Let's check the headline numbers against each other. From 14.4 billion in 2022 to 29.7 billion in 2027, the growth factor is 2.06, and the fifth root of 2.06 is 1.156. So the market grows by about 16 percent per year, which matches both the +16% label and the WLAN CAGR. This is the practical meaning of the NOTE above. WLAN does not grow the fastest, but it grows with the market from one of the two largest bases, so its lead does not shrink. The chart counts installed connections, not shipments, so it shows how many devices a WiFi network has to carry at the same time.

• WLAN keeps its share : its CAGR equals the market CAGR of about 16 percent.
• A high CAGR on a small base is still small : 87 percent per year for 5G IoT does not change the ranking by 2027.

Regular WiFi vs IoT specialized WiFi ?

Do we need a specific variants of WiFi for IoT ? or is it same as regular WiFi ? The answer depends on three properties of the device: its power source, its distance from the access point, and the number of other devices that share the channel with it.

IoT devices can use regular WiFi, but there are also specific WiFi standards designed for IoT applications. Regular WiFi can be used for devices that need higher data rates and have more power, like smart TVs and cameras. For IoT devices that need less power and have lower data requirements, like sensors and smart bulbs, WiFi variants like WiFi HaLow (802.11ah) are more suitable because they offer longer range and can connect to more devices. This helps save energy and extends the life of the device's battery.

There are a few reasons why specialized variants of WiFi have emerged for specific IoT use cases:

• Key Differences:
• Power Consumption: IoT devices often need to run on batteries for long periods. Regular WiFi can be too power-hungry. Specialized IoT WiFi variants, like Wi-Fi HaLow (802.11ah), focus on low-power operation to extend device battery life.
• Range: Many IoT applications need extended range to cover sensors in a factory or farm. Wi-Fi HaLow is optimized for longer distances than standard WiFi.
• Number of Devices: Regular WiFi might struggle with a large number of connected IoT devices. Newer generations of WiFi (like WiFi 6 and WiFi 6E) are better equipped to handle dense device environments, improving IoT stability.
• When can I Use Regular WiFi for IoT ?
• For many home IoT applications, regular WiFi works perfectly fine. Here's when regular WiFi might be sufficient:
• Home Environment: Smart lights, thermostats, cameras in your home likely function well on regular WiFi.
• Limited Devices: If you have just a few IoT devices, your standard WiFi should handle it perfectly.
• Power Availability: If your IoT device has consistent access to power (not battery-operated), a standard WiFi connection might work just fine.
• When Specialized WiFi Might Be Better ?:
• Battery-Powered Devices: If power conservation is crucial, Wi-Fi HaLow or other low-power options might be necessary.
• Large Areas: Industrial settings, agriculture, or outdoor IoT deployments needing better range would benefit from Wi-Fi HaLow.
• Dense Device Networks: Warehouses, hospitals, etc., with many IoT devices may require the greater scalability and features of WiFi 6/6E.

Let's look at what makes Wi-Fi HaLow different in numbers. 802.11ah reuses the OFDM design of 802.11ac, but it runs the clock ten times slower. So a 20 MHz channel becomes a 2 MHz channel, and the subcarrier spacing becomes 31.25 kHz. The useful symbol grows from 3.2 to 32 microseconds, and with the guard interval a symbol lasts 40 microseconds, or 36 with the short guard interval. The channels are 1, 2, 4, 8 or 16 MHz wide, in license-exempt bands below 1 GHz.

The table below applies the rate formula of the Evolution page to these parameters. The first row is MCS10, which repeats every bit twice for extra robustness.

The formula reproduces the two figures that are usually quoted for HaLow. The minimum is 150 kbps, and four streams at 16 MHz give 4 x 86.7 = 346.7 Mbps. An IoT sensor needs only the low end of that range, and the low rate is also the source of the long range.

The range gain comes from two sources, and both can be computed. First, free-space path loss grows with 20 x log₁₀ of the frequency. Going from 2437 MHz to 915 MHz therefore saves 20 x log₁₀(2437 / 915) = 8.5 dB. Second, the thermal noise in a 1 MHz channel is -114 dBm, against -101 dBm in 20 MHz. That is 13.0 dB less noise. Together, the link has about 21.5 dB more margin at the same transmit power. With a path loss exponent of 3, that margin multiplies the range by 10^21.5/30, or about 5.2. In free space, with an exponent of 2, the factor is about 11.9.

• HaLow is 802.11ac clocked ten times slower : narrow channels, long symbols and the same rate formula.
• The range comes from frequency and bandwidth : about 8.5 dB from the lower band and 13 dB from the narrow channel.
• The rate is the price : a 1 MHz channel carries between 0.15 and 4.44 Mbps per stream.

Technical Challenges to WiFi IoT

Integrating WiFi into IoT devices brings its own set of hurdles. In this section, I am going to explores the intricate technical challenges—from power management and signal reach to ensuring secure and seamless communication among a multitude of diverse devices—underpinning the sophisticated world of WiFi IoT.

• Power Consumption:
• Many IoT devices are battery-powered, requiring long battery life. Regular WiFi can be power-hungry, draining batteries quickly.
• Techniques like power-saving modes and optimized sleep/wake cycles are needed to conserve energy.
• Range:
• Standard WiFi's range might be insufficient for applications like smart agriculture or industrial monitoring, where sensors are spread out over large areas.
• Solutions like WiFi HaLow (802.11ah) cater to long-range requirements but may have trade-offs in data rates.
• Scalability:
• Home WiFi networks struggle when overwhelmed with a large number of IoT devices competing for bandwidth.
• Newer WiFi standards (WiFi 6/6E) introduce technologies to manage more devices efficiently, but older routers create bottlenecks.
• Security:
• IoT devices are potential entry points for cyberattacks. Weak security protocols on devices or in network configuration leave systems vulnerable.
• Robust encryption, secure authentication, and regular updates are essential for IoT security over WiFi.
• Interference:
• The 2.4 GHz band (frequently used by WiFi) is crowded with Bluetooth devices, microwaves, etc., causing interference and affecting IoT device reliability.
• WiFi 6 and 6E use less congested frequency bands (5 GHz, 6 GHz) to mitigate interference, promoting greater stability.
• Connectivity in Harsh Environments:
• Industrial or outdoor IoT deployments expose devices to extreme temperatures, vibrations, and dust. WiFi equipment needs to be designed to withstand these challenges.
• Device Management and Updates:
• A large-scale IoT system requires managing many devices remotely, deploying software updates, and ensuring configurations are updated over WiFi. This needs seamless orchestration.
• Additional Considerations
• Cost: Low-power WiFi variants can be costlier to implement for small-scale IoT projects.
• Limited Expertise: Specialized WiFi IoT protocols and features might require specialized knowledge for proper implementation.

The power consumption challenge is the one that decides whether a battery-powered device can use WiFi at all, so let's put numbers on it. The values below are assumptions for illustration, not measurements of a real chip. A battery holds 2500 mAh. The radio draws 10 microamperes while it sleeps, and 50 mA while it is awake to receive.

Suppose first that the device stays associated in the classic way. It wakes for every beacon, which comes every 100 TU, or 102.4 ms, and it stays awake for 2 ms each time. The average current is 50 x 2 / 102.4 = 0.98 mA, and the battery lasts about 2530 hours, or 106 days. Now suppose instead that the device sleeps for 5 minutes, then wakes, sends one report in 50 ms at 100 mA, and sleeps again. The average current drops to 0.010 + 100 x 0.05 / 300 = 0.027 mA, and the battery lasts about 10.7 years. The only difference is how often the radio wakes up.

This is why Target Wake Time, or TWT, matters for IoT. TWT came from 802.11ah and was adopted by 802.11ax. With TWT, the station and the access point agree on when the station will wake. The station can then skip the beacons in between, and the access point buffers its traffic until then. The same idea appears in the DTIM period and in the listen interval of older WiFi, but TWT lets each station negotiate its own schedule.

• The wake-up rate sets the battery life : in the example, the same battery lasts 106 days or 10.7 years.
• TWT gives each device its own schedule : the device sleeps through the beacons it does not need.
• Check the assumptions : real sleep and receive currents vary widely between chips, so the example shows the method, not a product figure.

Reference

• State of IoT 2023: Number of connected IoT devices growing 16% to 16.0 billion globally – Wi-Fi, Bluetooth, and Cellular driving the market - IoT Business News (May 2023)
• 60+ IoT statistics and facts - Comparitech (Jan 2024)
• Internet of Things statistics for 2024 – Taking Things Apart - DataProt(Feb 2024)