For extreme applications, it's obvious why we need the best antenna possible. 
For Internet of Things (IoT) applications, antenna performance may seem unimportant. The whole idea of IoT is based on having many devices close together with inexpensive transceivers. Despite the close range and low cost, though, antenna performance still matters. Scroll down to Example for a real-world example of a manufacturer focused on improving antenna performance of an IoT product.
Why does antenna performance matter?
Noise
An IoT-enabled device may be used in noisy environments where more power is needed to overcome interfering signals on the receiver's antenna. Regardless of the receiver's sensitivity, if it's receiving -80dBm of noise power, either from unintentional emissions from equipment or from any of the non-Bluetooth devices that share the ISM bands, it will need roughly -60dBm of signal to overcome the interference. How much power is needed to overcome interference depends on the nature of the interference and the type of signaling the IoT system uses.
Fading
In an indoor environment, radio waves bounce off walls and objects. The reflections combine constructively or destructively, depending on their phase. If you spend a long time using a wireless device indoors, you can eventually find unlucky locations where the multipath interference is particularly destructive and the signal drops 30dB. A better antenna won't affect the depth of these “fades”, but it will raise the signal strength so that the fades don't cause the signal strength to fall below the receiver's sensitivity.
Efficiency
Every dB of increased antenna performance is potentially 1 dB less output power needed, decreasing the IoT device’s power budget. If improving the antenna decreases the number of packet retries, it can significantly improve battery life.
Calculating Range
The signal strength at the receiver is equal to the transmitter power plus the antenna gains minus the path loss. There is a straightforward formula for calculating path loss. If you run the formula for the 2.4GHz ISM band at a distance of 10 meters, you find a path cost of 60dB. A typical low-power transmitter's output is 1mW [0dBm]. A typical IoT receiver requires at least -85dBm of signal to work reliably. In this scenario if the antennas have zero gain and there are no obstructions between them, the signal strength at the receiver will be -60dBm, 25dB more than needed. This “extra” signal strength is called link margin. We need this link margin to handle fades, interference, detuning of the antenna due to proximity of the human body, and objects that block that path. If the antennas have 3dB loss on both sides, the receiver only gets -66dBm of signal. -66dBm > -85dBm, so the system will work, but the lower link margin means it will be less robust against fades and interference.
-6dB equates to one fourth the power. That means in the above scenario with -6dB of antenna loss, they could use one fourth the output power and get the same performance.
Example
Here is an example of an IoT wall switch from Amatis Controls that uses a chip antenna. We solder a small connector onto the board and measure the impedance of the antenna on the vector network analyzer (VNA). On a Smith Chart view, the center of the plot represents a perfect 50 ohm match. With the default match network from the datasheet, the antenna is not well matched in the 2.4 GHz band.
The Smith chart view lends itself, after some experience using it, to seeing which matching network values will move the impedance of the antenna closer to matched. For more details on the matching procedure, see this post on matching an on-board antenna.
After several iterations of matching, the antenna is matched more closely to 50 ohms.
It’s a good idea to do field strength measurement before and after matching to see how well the received signal strength from the matched antenna agrees with strength predicted by the path loss equation. Usually better matching means more radiated power, and radiated power will equal nearly the transmitter's output minus the reflected power observed on the VNA. This is not always true. In an extreme case, we could put a 50 ohm resistor (called a “dummy load” in the amateur radio world) that would dissipate the power as heat and hardly radiate at all. We could have a very bad antenna that we manage to match to 50 ohms. There is a saying that says “you can match to a rock, but it does not mean it's a good antenna.” For this reason it’s good to check field strength after matching.
The original design from the chip datasheet resulted in a mismatch that cost only a few dB. They probably had enough link margin that users wouldn't notice at first if they did not match the antenna. Amatis engineers told us the IoT world is very competitive. If someone compares their product to a competitor’s they want the Amatis product to have better range and less susceptibility to fades or interference.
Top Comments