Wireless links and networks: determining your antenna needs
In today’s industrial environments, wireless networks are becoming increasingly commonplace. Most process and automation engineers are not necessarily trained in radio physics, so here is a basic antenna primer.
The use of wireless technology in industrial environments is on the increase, whether that be common standards-based technologies such as 802.11 Wi-Fi, or proprietary systems for long-distance communication. But whatever the technology used, the signal must successfully propagate between a transmitter and receiver node successfully, and that means successfully converting the RF electrical signal from the transmitter into an electromagnetic wave in free space, and then converting that wave back again at the receiver. The device that couples the electrical signal to the air is, of course, the antenna.
The true design of antennas and how they couple electromagnetic energy would appear to many to be a bit of a black art: a piece of metal sticking up in the air somehow sends and receives information out of the ‘ether’. The true physics of antenna design involves the complex geometry calculus of electromagnetic field theory, and is a practical application of the famous Maxwell equations.
The antenna is possibly the most important and influential element of your wireless system design, because the right antenna can make your radio system sing, while the wrong one can be a disaster. Sometimes, in applications where radio reception is unduly influenced by interference, or where the signal strength is only just making it (and maybe fades in bad weather, for example) then choosing a higher gain antenna may solve the problem simply and easily.
In industrial environments there may be many elements that can interfere with the clear reception of a signal at the receiver, and a good application of the right type of antenna (or combination of antennas) can customise the transmission path to your needs.
This article is about the basic knowledge of the key performance characteristics of antennas commonly used. Of course, antennas come in many shapes and sizes, ranging from the small antenna hidden in your smartphone, to the multi-element arrays that your television uses, to the huge dish antennas you see at satellite base stations. This article will focus on the basic designs and how they are practically implemented.
General features of antennas
Every antenna has its own specific design characteristics that determine the range and radiation pattern of the radio signal. Also, if you analyse its radiation pattern for transmitting, its sensitivity pattern for receiving is the same. That is, the direction in which it transmits the strongest is also the direction in which it is most sensitive for receiving.
But what do we mean by radiation pattern?
If we transmit a constant power from an antenna, and then measure the strength of the radiation at a fixed distance in a sphere around the antenna, we get a variation in received signal strength depending on what angle we are to the antenna. This can be charted as a 3D geometrical shape showing where the signal is stronger (more focused) and where it is weak, or even non-existent. No antenna transmits equally in all directions spherically.
Two antennas will ‘couple’ best when they are geometrically aligned on the best transmitting/receiving axis. In between them is the free space over which the energy will propagate, and there is a ‘free space loss’ associated with the propagation that means only a portion of the transmitted energy will arrive at the receiver.
Gain
The shaping of the radio signal is what is referred to as antenna gain. The higher the gain of an antenna, the more focused the signal.
Antenna gain is measured in decibels. A decibel is a logarithmic ratio between a specific value and a base value of the same unit of measure. With respect to radio power, dBm is a ratio of power relative to 1 mW, where 1 mW equals 0 dBm.
Table 1 shows the logarithmic relationship between dBm and power (10log(P)): a small change in dBm results in a large change in power.
For every reduction of 10 dBm the power is reduced by a factor of 10 and power levels below 1 mW are negative decibels. The system’s power is reduced by half with a change of only -3 dBm.
Specifications for most antennas refer to the gain in either dBi or dBd. The first one, dBi, is decibels relative to an isotropic radiator — a theoretical model antenna with a perfectly spherical radiation pattern (a point). Such an antenna, if possible to create, would radiate all the energy in all directions equally, so would not add to the transmitted power in any direction. This unit is commonly used, and so all antennas have a positive gain in dBi. Sometimes antenna gain is specified in dBd: decibels relative to a simple dipole antenna. At the operating frequency of the antenna (the frequency it is tuned to) 0 dBd is equal to 2.15 dBi. In other words, a simple dipole antenna has a gain of 2.15 dBi.
Omnidirectional antennas
The dipole antenna mentioned above is an example of an omnidirectional antenna. As the name implies, an omnidirectional antenna transmits and receives radio signals equally in all directions. But omnidirectional does not mean the same as isotropic: the omnidirectional antenna radiates equally in all directions on its axial plane, but the signal weakens at angles off-centre to that plain. In other words, the radiation pattern looks like a circle when viewed from the end of the antenna, but is a toroid, or doughnut-like shape, when viewed from other angles (see Figure 1). A good example of an omnidirectional antenna is a radio station transmission tower or the antennas used on mobile phones or walkie-talkies.
Notice from Figure 1 that the decreased energy sent vertically increases the energy transmitted horizontally. The radiation pattern stretches to extend the range, focusing the signal along a horizontal plane. When we say a dipole antenna has a gain of 2.15 dBi we are referring to the central horizontal axis of the radiation pattern (where it is strongest). Notice there is also only a weak signal very close to the antenna, so that the signal may be lost if the angle off centre is even very small.
In a wireless network, omnidirectional antennas are best suited to indoor environments and for devices at the centre of a star topology network. For long-range point-to-point communications, omnidirectional antennas would not be the best option, since they would waste energy radiating in unwanted directions.
Directional antennas
Only dipole antennas can really be considered omnidirectional. Other designs tend to focus the energy more in a given direction or axis. There are many designs of directional antenna.
There are two general techniques that are used to do this. One technique is to use large metal surfaces such as parabolic reflectors (think satellite dish) or horns, which change the direction of the radio waves by reflection or refraction, to focus the radio waves from a single low gain antenna into a beam. This type is called an aperture antenna.
A second technique is to use multiple dipoles in an array, fed from the same transmitter or receiver. If the currents are fed to the antennas with proper phase difference, due to the phenomenon of interference the waves from the individual antennas combine (superimpose) at various angles around the array. In directions in which the waves from the individual antennas arrive in phase, the waves add together (constructive interference) to enhance the power radiated. In directions in which the individual waves arrive out of phase, with the peak of one wave coinciding with the valley of another, the waves cancel (destructive interference) reducing the power radiated in that direction.
A Yagi-Uda antenna (usually known simply as a Yagi), named after its Japanese inventors Shintao Uda and Hidetsugu Yuda, is a commonly used version of a multi-element array. It is made up of a series of different sized dipoles spaced at specific distances from each other, to enhance the signal strength in one direction and weaken it in another, as they interfere with or enhance one another. Only one element is driven while the other passive elements act as directors or reflectors, depending on their position and length. Depending on the design, a Yagi can typically have a gain of up to 20 dBi.
Yagi antennas are suitable for long-range communications. In process control and SCADA networks, Yagis are often used in outdoor applications like tank level monitoring. There must be a line of sight between the antennas.
The importance of line of sight
When using directional antennas, any obstructions, including buildings, trees or terrain, that interrupt the visual path between antennas will also interfere with the radio signal transmission, resulting in multipath fade or increased signal attenuation.
Multipath fade is the result of radio signals reaching the receiver via two or more paths. In industrial settings, a received signal may include the line-of-sight signal in addition to signals reflected off buildings, equipment or outdoor terrain.
There is a space around the radio ‘beam’ known as the Fresnel Zone, in which obstructions, even if not in the direct line of sight, can cause reflections and attenuation of the signal. The Fresnel Zone is an ellipsoid shape, and is thickest at the centre-point of the radio path. Antennas need to be raised high enough so that no obstructions come within the Fresnel Zone.
The size of the Fresnel Zone at any point on the path can be calculated, and is based on wavelength and the distance between the measurement point and the two antennas.
where:
r is radius of the Fresnel zone in (m)
d1 is the distance from the transmitter (m)
d2 is the distance from the receiver (m)
D is the total link distance (m)
f is the link frequency (MHz)
This is not the height above the ground, but the radius of the zone around the line-of-sight path.
A site survey should always be conducted before committing to antenna placement.
Calculating link margin
Radio receivers have a minimum sensitivity (measured in dBm) below which they will not receive a usable signal. In order to select an antenna pair with the necessary gain to ensure a received signal, it is necessary to calculate the link budget. Essentially, the link budget is an indicator of link performance. Once we know the transmit power, the transmitting and receiving antenna gains and the free space loss over the link path, then we know how strong the signal is at the receiver.
Using decibels means that the formula for link budget is a simple addition of gains and losses. Gain (G) is positive in dB and losses (L) are negative. A full calculation also takes into account small losses for connectors and cables, as follows:
where:
Prx is the received power (dBm)
Ptx is the transmitter power output (dBm)
Gtx is the transmit antenna gain (dBi)
Ltx is the transmitter losses, coax, connectors etc (dB)
Lfs is the free space path loss (dB)
Grx is the receiving antenna gain (dBi)
Lrx is the receiver losses, coax, connectors etc (dB)
An additional element for fading and other path losses (if estimated) can also be added.
Another term you may come across is EIRP, or effective isotropic radiated power. This is the first two elements of the equation above (transmitter power plus antenna gain) and represents the effective power transmitted in the centreline of the main radiation lobe.
The difference between the received signal strength and the receiver sensitivity is known as the link margin. The link margin needs to be greater than 0 dB to be a working link (the received signal is greater than or equal to the receiver sensitivity).
Since it is an idealised calculation, not taking into account natural variables such as weather or other forms of interference, the link margin is effectively how much additional attenuation can be tolerated by the receiver: the higher the margin the better.
Free space loss
Free space loss is caused by the geometric spreading of the wavefront. The energy is spread over an area that increases with distance from the transmitting antenna, so the power density per unit area diminishes. Figure 4 shows free space loss versus distance for two common Wi-Fi frequencies.
The idealised free space loss is calculated using:
where:
Lfs is the free space path loss (dB)
d is the distance in (km)
f is the carrier frequency (MHz)
For example, a 2.45 GHz signal transmitted over 5 km will exhibit an ideal free space loss of -114 dB.
Assume we have a receiver with a sensitivity of -82 dBm. If the signal is transmitted from a transmitter with a power of 20 dBm and an antenna with 10 dBi gain, the receiver antenna has 14 dBi gain and the cables cause a loss of 2 dB at each end, then we have a received power of:
The link margin is therefore +8 dB, so the link should work.
Conclusion
As you can see, determining the appropriate components for your radio system is not as complex as it might appear, without detailed knowledge of radiophysics.
Data sheets for all radio transmitters, receivers, antennas, cabling and connectors should list the transmit power, receive sensitivity, antenna gain and cabling losses. Connector pairs may have a 0.5 dB loss while a lightning arrestor could include a loss of 0.5 to 1.5 dB. Cabling losses vary by manufacturer and are typically listed per unit length of cable.
The only parameter to be determined for the specific application is free space loss. Using the equation for free space loss, you should be able to determine if there is sufficient link margin for each link in a point-to-point radio system.
Disclaimer
I should add a small disclaimer. There are other parameters that affect antenna performance, which are beyond the scope of this article. For example, if an antenna is mounted on a metal structure or surface, the structure acting as a ‘ground plane’ can alter the radiation pattern of the antenna, and focus it more or less in different directions, depending on its shape, so placement can be important if your link margin is small.
Also, every antenna has a characteristic impedance, and for best results (minimum losses), the cabling and connector system should match the antenna’s impedance as closely as possible. If you are buying your equipment from a good supplier, this should be easy to sort out.
There are also many types of antenna design: this article only gives general examples.
Mineral processing: a eulogy for analog
Leading mines have already accomplished an automated, digitally connected mine and are reaping...
What is TSN and do we really need it?
Whether or not TSN becomes an industry-wide standard remains to be seen.
AI and condition monitoring
The rise of artificial intelligence is seen as particularly useful in the field of condition...
