Our team is planning to make a HAB (High Altitude Ballon) project for our university. We are planning to use XBee pro 868 RF modules (we are in Europe so we can not use 900 MHz band). We have to communicate to the base station from the payload from 20-30 km distance. My question is, what kind of antennas will we need to use? The Xbee data-sheet says that with high gain antennas the module can communicate up to 40 km with clear LOS (line of sight). If we put high gain antennas like YAGI types on the base station, do we still need to attach high gain antennas or are dipole antennas enough for the payload ?

Note : I know that at some distance the connection will be lost and we have to move our base station according to last GPS data to find payload location landing area. But we are trying to do best as we can to extend communication duration before we lose the connection.

Module specifications:

enter image description here

  • \$\begingroup\$ What data rate do you have to communicate or is it voice or voice and data? What power output is the RF module? Is it two-way comms? What is the stated sensitivity for your receiver? Can you link a data sheet? \$\endgroup\$
    – Andy aka
    Commented Nov 29, 2014 at 17:25
  • 3
    \$\begingroup\$ Since you're a student I will mention this but given that English may not be your first language, I won't comment on those edits, but in general, you should treat this as you would your school assignments. Your question as posted was very sloppy with poor attention to presentation. Attending university isn't just about the course work, it's also about communicating and making yourself understood. \$\endgroup\$ Commented Nov 29, 2014 at 18:08
  • \$\begingroup\$ The specifications you list here say that 40km is attainable with dipole antennas, which are hardly directional at all. With more directional antennas, even greater distances should be attainable. \$\endgroup\$
    – Phil Frost
    Commented Nov 29, 2014 at 20:23
  • \$\begingroup\$ Instead of posting a snapshot of what the spec says why don't you state the data rate YOU need to use. If you don't understand my previous comment you may be over your head on this. \$\endgroup\$
    – Andy aka
    Commented Nov 30, 2014 at 1:01
  • \$\begingroup\$ Consider a low cost cell phone. Good range, clear communications, video, built in GPS, light weight, easily replaced, external battery options, some support Data modes, built in antenna. \$\endgroup\$ Commented Nov 30, 2014 at 2:41

3 Answers 3


What you need is a link budget. The specifications you supply say that the maximum transmit power is 25 dBm, and the receiver sensitivity is -112 dBm. This means that you can afford, at best:

$$ 25\:\mathrm{dBm} - (-112\:\mathrm{dBm}) = 137\:\mathrm{dB} \text{ of loss } $$

You will, of course, want to leave a healthy margin for robustness, but it gives you some bounds.

Loss calculation is greatly simplified for a balloon, since we can reasonably assume a clear line of sight. Atmospheric conditions (fog, rain, etc) can increase the loss, and you may have to compete against noise from other radios on the same frequency, but that's what the margin is for.

The most obvious loss is that due to the distance between the antennas. This is called the free space path loss, and we can calculate this loss, in decibels, as:

$$ 20 \log_{10}(d) + 20 \log_{10}(f) + 32.45 $$


  • \$d\$ is the distance, in kilometers
  • \$f\$ is the frequency, in megahertz

So for your specified distance of 30 km, and a frequency of 868 MHz:

$$ 20 \log_{10}(30) + 20 \log_{10}(868) + 32.45 = 120.76 \:\mathrm{dB}$$

This loss (121 dB) is less than the maximum loss based on the transmit power and sensitivity above (137 dB), so in theory, the link should work, even with an isotropic antenna.

In fact, you have a margin of \$137 - 121 = 16 \:\mathrm{dB}\$. Any gain that your antennas have is going to increase this margin. It doesn't matter if you add antenna gain at the receiver, transmitter, or both. Because of reciprocity, any gain in the system helps the same way. Additional margin may also allow the transmitters to operate at a lower power, which will increase your battery life, if that's a concern.

There's another source of loss that may not be obvious: polarization loss. Since the balloon is spinning about, you don't know what its orientation will be. Satellite communications have the same problem, and the canonical solution is circular polarization.

Since you don't need a lot of gain (and in fact, too much gain will make it hard to aim the antennas), a turnstile antenna may be a good option. It's circularly polarized and easy to construct. Adding a set of reflectors as in the first image from that Wikipedia article might not be a bad idea for the ground antenna, just for some extra margin:

turnstile antenna with reflectors

This antenna could also be described as two crossed Yagi antennas on the same boom and fed in quadrature, so to calculate the geometry of the antenna elements you can use existing Yagi designs. If you research Yagi antennas for satellite communication you should find ample information.


The issue that you will have in any high gain antenna is the loss of spatial diversity. Simply put the more gain, the better the tracking had better be. The balloon-borne side of things dictates that you probably can't have any tracking there simply because of weight, power and complexity. However, consider that you don't need to be transmitting upwards, so an antenna with only a forward (downward) component will give a nice trade off between antenna gain and spatial diversity, there really is no trade off there. If you have a good idea of the wind conditions (i.e. predicting down wind travel distance) that are typical and the height you intend to reach then you might be able to restrict your antenna directionality further and still not have to have tracking.

A nice directional antenna like a YAGI with tracking at the base station sounds like the right idea. Especially if it is mobile and you can move to keep closer.

  • \$\begingroup\$ What about using YAGI for base and using PATCH(180 degree) antenna for payload ? \$\endgroup\$ Commented Nov 29, 2014 at 17:49

If you are going to be transmitting 20-30 km from the antenna, I would suggest using am omnidirectional antenna on the balloon. You can't have a directional antenna pointing straight down, and if you try to mount it at an angle, you won't be able to control the direction since the balloon can rotate in any direction.

Although it would seem that the longer the antenna the better, actually a 1/4 wavelength (86 mm at 868 MHz) monopole "whip" antenna is the best choice for you. For other lengths such as 1/2 or 5/8 wavelength there will be a reactive component at the feed point which requires additional loading to match to 50 ohms.

The advantage of monopole antennas is its radiation pattern is uniform in all directions, unlike for example a dipole antenna (which are typically 1/2 wavelength) which has radiation patterns concentrated into two lobes.

Whip antennas are designed to be mounted above (or below, in your case) a ground plane, such as the roof of a card, and there is no radiation below (above) this ground plane. Pretend the car below is your balloon.

enter image description here

So you need to provide either a ground plane, or provide a pseudo ground plane using additional elements angling down from the antenna as shown below on the right. Thee are actually four, the two perpendicular to the others are not shown. They can be any stiff material; coat hangers would do fine. These other elements are the same length, 1/4 wavelength, as the antenna. The metal ground plane below the antenna on the left is the roof of a car.

At your base station I would use a directional antenna such as a Yagi since you can precisely control the direction.

enter image description here

I have successfully used this arrangement to receive data at 920 MHz from a transmitter on a balloon flying at 70,000 feet and 240 miles away from my base station (a truck), using a whip antenna at both ends.

  • \$\begingroup\$ Why can't a directional antenna point straight down? \$\endgroup\$
    – Phil Frost
    Commented Nov 29, 2014 at 19:12
  • \$\begingroup\$ Also, a 1/4 wave monopole has exactly the same radiation pattern as a 1/2 wave dipole, except for the lack of radiation below the ground plane as you mention. The dipole's radiation pattern is a doughnut, and the monopole is a sliced bagel. A horizontal dipole over ground can have its doughnut pattern distorted by the image antenna created by the ground, but this isn't going to happen in a balloon. \$\endgroup\$
    – Phil Frost
    Commented Nov 29, 2014 at 19:18
  • \$\begingroup\$ Also, coat hangers, usually being made of mild steel, aren't a great choice for antenna material. The high permeability of steel (iron in particular) increases skin effect, making the lossier than it needs to be. The lacquer coating on a typical coat hanger is also a pain to remove for a good electrical connection. Aluminum or copper are better choices. \$\endgroup\$
    – Phil Frost
    Commented Nov 29, 2014 at 20:20
  • \$\begingroup\$ "Why can't a directional antenna point straight down?" I hadn't thought of that, but I imagine it would work. I'm not sure it would have any advantages over the monopole, since you would lose half of the power in wasted the backward lobe. Re the coat hangers, that was suggested in the article accompanying that diagram, but you make some good points. \$\endgroup\$
    – tcrosley
    Commented Nov 29, 2014 at 20:28
  • \$\begingroup\$ Why the downvote? \$\endgroup\$
    – tcrosley
    Commented Nov 29, 2014 at 20:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.