Holding industrial wireless vendors to account
While wireless technology has moved well beyond simple point-to-point connectivity, the fundamental tenets of the technology remain the same. However, one shouldn’t be put off, and we don’t all need to be certified RF engineers to start making informed choices. This article looks to highlight the fundamental tenets of RF technology and empower more informed decision-making in relation to which wireless technology to deploy relative to application need.
There can be little doubt that wireless technology is omnipresent in our professional lives at present. From the touting of standards such as ISA100, WirelessHART and IEEE 802.11, to wireless sensor-based networks (WSN) and mesh technology, wireless systems are becoming increasingly accepted and integrated into greenfield and legacy plants and applications globally.
However, as with much in life, a trade-off exists when deploying wireless. Informed decision-making means looking at the criticality and latency of the PV in the process, the volume of information needing to be transferred and the required communication distance and terrain - and all of this should be relative to the frequency waveform properties, modulation scheme and Australian Communication and Media Authority (ACMA) guidelines.
The wireless spectrum and terminology
Informed choices on wireless technology begin with the understanding that wireless equipment manufacturers have only a subset of the variables that they can control and even these are subject to regulatory compliance. The balance of factors are application or site specific and are decision variables for the site engineer.
In design, manufacturers can control the amount of RF power emitted (to regulatory standards), the amount of modulation (also, by default, a function of the regulatory body) and the receiver sensitivity (the lowest RF signal that the receiver can reliably detect). Other decision variables are application specific (Do I need 4-20 mA or live video feed?) or site-specific (such as, How far do I want to communicate and over what terrain?). These variables are pivotal and go to the heart of good wireless technology decision-making.
A summary of how wireless works.
Wireless communication involves modulating binary data onto a carrier waveform and propagating it via the Fresnel zone (elliptical path of RF) between transmitting and receiving antennas. The data is then removed (or demodulated) from the carrier wave for interpretation by the receiving device. It is the obstruction of the Fresnel zone relative to the frequency waveform properties, regulatory compliance, receiver sensitivity and modulation that will impact on reliable communications over a given distance.
Talking and comparing RF
Radio frequency (RF) signals are often characterised by two common measurements - frequency and power. Frequency is measured in Hertz (Hz) and RF signal strength is often specified in milliwatts (mW) or decibels (dB). When working with radio-based systems, it’s useful to understand both and convert between them, as they will often be used interchangeably between vendors. Moreover, conversion will be required to understand and comply with wireless regulatory approvals (for example, when deciding on overall antenna gain).
The relationship between milliwatts and decibels is defined by the following equation:
dBm = 10 log10(RF power in mW)
The above equation shows radio signal strength expressed in decibels with reference to 1 mW of RF power. Therefore, 1 mW of RF power = 0 dBm. Given that it’s a logarithmic scale, a doubling of RF power adds another 3 dB.
Understanding waveform properties
The question is often asked, how far will my radio signal reliably go? This is a function of a number of factors beginning with waveform properties.
From your technical training you should remember that wavelength is inversely proportional to frequency. Waveform diffraction (the ability of the waveform to bend around objects), reflection (the ability to bounce off objects) and general object penetration is better at longer wavelengths, and therefore lower frequencies, than it is at higher frequencies. Moreover, higher frequencies, with their smaller wavelength, are more prone to scattering upon meeting obstructions (known as multipath fading). This, and the modulation technique (discussed later), is the reason higher frequency devices often have two or more antennas compared with a low frequency device.
In summary, the general rule is that obstacles, and their location relative to the Fresnel zone, decrease the overall reliability and operating distance. Any blockage impeding the Fresnel zone from opening will decrease reliable communications distance, depending on the properties of the transmitted waveform.
RF power and distance
Irrespective of frequency, increasing a transmitter’s RF power and antenna gain will increase the communication range. This is due to the higher power mitigating the signal attenuation that occurs as the signal passes through, and reflects or bends around, obstructions. In effect, RF signals attenuate proportionally (through a constant medium) to the square of the distance. What this means, practically, is that to double your reliable distance, you need to increase your RF power level by four times (ie, add 6 dB). One way to do this is by using a higher gain (more directional) antenna. The benefit of a directional antenna is that it achieves greater power (and distance) in one direction while reducing spill (and sensitivity) to the sides and back, giving greater control over interference.
However, government regulations on the amount of emitted RF power are enforced to ensure coexistence and management of the radio spectrum. These allowable limits are known as effective isotropic radiated power (EIRP) or effective radiated power (ERP) and are referenced to an isotropic or dipole antenna respectively. As a result, an antenna with higher gain (such as a dish or array) will increase the EIRP in the direction the antenna is facing. To work out the effective power, you add the gain of the antenna in dBi to the transmitter power in dBm to get an effective dB power emitted, after allowing for insertion losses such as cable and surge arrestors.
It is timely that we introduce what are termed ‘licensed’ and ‘licence-free’ bands. In general, licensed systems are those that the ACMA grants for a given geographic area, given channel size and level of RF power. Licence-free or ‘industrial, scientific and medical’ (ISM) use does not require a licence to be granted, but users need to query vendors on the ability of their equipment to mitigate interference on the same band.
Effect of receiver sensitivity
Communication distance is also a function of radio receiver sensitivity levels, which are often specified at a particular bit error (BER) or frame error rate (FER), such as -108 dB @ 1x10-6BER. For a given frequency, radio products with an ability to receive at lower levels will outperform those with poorer levels of receiver sensitivity. Put simply, they can detect and demodulate more successfully over longer distances.
However, receiver sensitivity by itself does not ensure reliable communications. There is also the effect of fade margin, which represents the difference between the dBm level of the received signal relative to the dBm level of RF background noise of the same, or similar, frequencies.
A system with a poor fade margin, even with superior receiver sensitivity, will not perform reliably when compared with a system of lesser receiver sensitivity but strong fade margin. This is particularly relevant where environmental factors can be transient (inclement weather) and add to the problem. Intermittent communications is a telltale sign in this situation.
So, good receiver sensitivity along with a good fade margin will have a high impact on reliable communications distance.
Wide versus narrow channels, band size and modulation.
Up to this point you could be forgiven for thinking that lower frequencies are the panacea. Well, that depends on the application and the required data throughput - there are always trade-offs in wireless physics!
While lower frequencies offer greater range, it is also the case that these frequencies are made up of either a single narrow channel or multiple smaller channels within a band. However, higher frequency systems have wider bands and the channels are wider.
More numerous and wider channels allow for greater modulation and potential data throughput. Why can’t we modulate more at lower frequencies? For each increase in modulation (and data rate) the spread of the frequency lobe (the size and number of sidebands) increases, potentially spilling into adjacent channels, creating potential interference.
In general, it is beneficial to know which modulation technique is being deployed as it will help achieve a more informed choice. There are three types commonly used.
Digital frequency shift keying (FSK) modulates data on a given carrier waveform and is typically the domain of lower frequencies with narrow channels. It is, therefore, limited in its data throughput capabilities. The limitation of FSK can be its susceptibility to interference on that frequency which can (not always) be a reason to use a licensed frequency.
Frequency hopping spread spectrum (FHSS) is a scheme using narrow channels within a band, scanning and hopping through available channels when communicating or experiencing interference. Again, channels are smaller in size and are typically in the range of 19.2 Kbps at 900 MHz and toward 250 Kbps for FHSS at 2.4 GHz (depending on channel size).
Direct sequence spread spectrum is a wideband modulation technique spreading the data across much of the band using differing variants of differential phase shift keying (DPSK). The concurrent spreading and even multidimensional sending of data streams (eg, multiple in/multiple out spatial multiplexing of 802.11n) allows for data throughputs to 108 Mbps and beyond.
Overall, 802.11 devices are not able to communicate as far but offer greater data throughput than lower frequency devices.
Repeatability and wireless mesh
Understandably, if repeating or wireless mesh technologies (by default repeating) are viable options, then greater communications distances may be achieved. Repeating of wireless communications has been available for some time but practical considerations include budgeting (including the cost of holding redundant spares for repeater-only units), site access considerations and associated infrastructure (such as antenna masts, power supply, etc).
The most common mesh architectures are those using a coordinator, a gateway or independent coordination and feature high or low RF power. Both coordinator and gateway networks typically feature low RF power and rely on redundant coordinators or gateways to manage network communications. When there are obstructions or long distances involved, low RF power will require the insertion of additional network nodes. Coupled with the addition of redundant coordinators or gateways, this adds to the cost of network infrastructure. Independent coordination networks, however, can be much different. Independent networks offer no single point of failure when a coordinator or gateway fails, and they typically provide higher RF power, promoting more reliable communication over distance.
When thinking about deploying wireless, the application and site specific questions you should be asking are:
What are the data throughput and connectivity requirements of my application?
Do you really need 108 Mbps or is it really just a want? Perhaps you might want a mixed system of I/O, VoIP and IP cameras - then the answer is not at lower frequencies but at higher frequencies. What are the data communication requirements of my equipment? Is it I/O, RS232 or RJ45? Do I need to communicate at different speeds (Modbus RTU to TCP) or are differing protocols at play? Put simply, but for cost, if you could cable between the two devices, would they interoperate? If so, and the data rate is low, then maybe the functionality and simplicity of a wireless modem will suffice. If not, then a gateway style of product may be applicable.
Over what distance and what terrain/obstacles am I looking to propagate?
Can I get site access for repeaters and do I have the budget for associated infrastructure (such as antenna masts, possible solar panels and the like)? What other RF is onsite that might interfere, or are there large vessels and the like in the propagation path? Clearly lower frequencies offer greater opportunity to get around obstructions, but can your process live with the lower data throughput? Are there sufficient data points/nodes that I can accommodate using low RF power mesh networks? If not, do I have the budget to accommodate this?
What are the latency requirements of the data communications for my process?
Can I engineer the process equipment to allow for the slower baud rates of a lower frequency in relation to the distance required? Can I live with the possibility that I lose communications for a time (such as with cellular network) or could it be that I can data agglomerate at given intervals (in which case cellular might be viable)?
Now take the vendor to task
Now you have a list of questions that are site or application specific. Armed with the above information, take the vendor to task! Quiz them on their product and its specifications relative to your application need, and make sure that their offering will suit your requirements reliably and effectively. Don’t accept a ‘one size fits all’ approach, but see if the vendor can tailor their offering to your real application needs.
By Brett Biondi, Elpro Technologies
In Part 1 of this article, we reviewed the history of wireless sensor networks (WSNs) and defined...
Today wireless instrumentation is becoming more commonplace in process plants and is a more...
Many operational and maintenance problems around a plant can be solved by deploying WirelessHART...