Page 1 of 7 2014-CIC-0050

Fundamentals of Wireless Communications

Candice King
Technical Applications Engineer, GE
IEEE Member
2308 Riverway Drive
Dallas, Texas 75227, USA
candice.king@ge.com

Abstract - Radio Frequency (RF) communications are an important smart grid enabler for functions such as volt/VAR control,
recloser control, feeder restorations and isolation, fault detections and sensor alerting in other industrial markets such as cement,
mining, and transportation. This paper will give a basic tutorial on the types of radio frequency communications and the benefits
and liabilities of each. Specific topics to be explored will be licensed versus unlicensed frequencies, distance between remote devices
and base stations, and communication architectures.
Radio technology is often referred in numerical ranges or frequencies. The decision on which frequency to employ in a network
depends on a few key variables. Prior to deciding which frequency for a network, the application for the radio use will assist with
dictation of which frequency range to utilize. Applications such as recloser control and volt/Var control may require a radio device
that can provide a high bandwidth/fast speed solution. Other SCADA applications such as sensor monitoring may only require small
bandwidth and for data delivery to be at a much slower speed. Besides SCADA applications, wireless communications is now
required in other industries such as cement and mining. Another variable when deciding on a radio network is the distance from the
main hosts to end remote devices such as RTUs or PLCs in the SCADA environment and sensors and devices for monitoring kiln
states in the cement industry. Lower end frequencies (100MHz-900MHz) provide further coverage and greater distance from base
stations/Access Points to remote end devices, whereas higher frequencies (2.4GHz-5.8GHz) provide shorter distance coverage, but
higher bandwidth and relay data back to SCADA hosts much faster.

I. INTRODUCTION

As with increasing technology advancements, all market segments and industries feel the need to automate networks. This
automation often comes in the form of some wireless solution. The wireless solution can come in the form of radio technology
or mobile and cellular technology. Any of these solutions can work great in a network, but understanding the application
utilizing the wireless devices and how they will fit into the network is vital when deciding on which technology to pursue.
Understanding of the network should include having a breadth of knowledge of the application for which wireless coverage is
desired, budgetary proposals for the project, layout of the network that includes all nodes such as Human Machine Interfaces
(HMIs) or SCADA hosts that handle and maintain the network, as well as all characteristics of the end devices such as
interfaces and speed requirements that will be physically connected to the wireless solution are all considerations that need to
be addressed. Physical interfaces may not be the traditional serial or Ethernet as in some industries. The utility market is
heavily known for using devices that have an Ethernet or serial connection. Other markets such as mining and cement may not
use devices that are capable of serial or Ethernet, but instead employ sensors that require either an analog or digital
Input/Output interface. The cement industry often utilizes kilns that reach very high temperatures, which would make it
humanly impossible to constantly monitor, whereas a remote wireless device could be attached and provide the consistent
monitoring required of such critical equipment.
For example, Fig.1 shows approximate distances (in miles) a particular frequency (in MHz) can cover. From Fig.1 noted
below, the use of licensed frequencies will ideally provide more coverage of a network versus using an unlicensed or registered
frequency. The distance can be determined based on the two endpoints that require connectivity. There may be two relays that
need to communicate with each other and are located miles apart, thus utilizing a wireless radio device with a frequency that
can provide this coverage. Of course, if the relays are more than 20 miles apart, obviously requesting a wireless device that
operates in the 3650MHz range would not be the best recommendation.

978-1-4799-3294-8/14/$31.00 2014 IEEE

 2014-CIC-0050 Page 2 of 7

5800

4950

3650

Unlicensed
1800
Licensed
Registered
900

450

200

0 10 20 30

*900MHz can be used as Licensed or Unlicensed

Fig.1. Distance Coverage for Frequency Ranges

The distances are approximate in that absolute values can only be identified with a field survey and performing site visits.
These field visits will help determine true distance coverage needed by taking into account the weather, surrounding obstacles,
if any, and any other variables such as buildings, trees, and any other wireless devices that might already be a part of the
desired network coverage.
As distances are vital considerations prior to a wireless solution choice, so too are the physical interfaces of the end devices.
These interfaces will determine which specific type of radio device will work in the network. Such interfaces could be either
an RJ45 or Ethernet connection or serial devices such as RS232 or RS485. Even with consideration of the serial devices, the
physical interfaces could either have 9 Pins or 25 Pins for connectivity. There are some wireless devices that may support both
Ethernet and/or serial and then there are those devices that support only one or the other. Fig.2 shows an example of a wireless
device that supports both types of physical interfaces that will allow for the connectivity to the end device regardless of the
physical interface [1].
Wide range power
RJ45 port for Ethernet input (10-30 VDC)
connectivity

2 independent RS232 ports for

serial connectivity

Fig.2. Wireless Device (Serial and Ethernet Port)

The physical interfaces are an important aspect for consideration in the wireless network because not only do they provide
connectivity to endpoint devices and controllers, but these interfaces also help to determine the bandwidth needed for a
seamless network. Serial devices operate at much slower speeds than those devices connected to Ethernet ports. The serial
devices allow connectivity to devices that do not require high latency or quick response times. Some examples of these

978-1-4799-3294-8/14/$31.00 2014 IEEE

Page 3 of 7 2014-CIC-0050

applications that do not require quick responses are those systems that may do a system poll and only require a response when
or if there is a problem. Other applications with low latency or response times are those that monitor a system for an ON or
OFF response. The bottom line is that low latency or slower response times are not a bad notion in a network, it is just a
reality. There are many applications that do not require an immediate response or the host system does not require a quick
reaction time. In these instances, a serial interface device is a perfect solution and will provide high reliability as well. Though
serial devices often operate in the 4800 to 9600 bits per second data realm, they are often the most reliable and provide the
longest coverage range with respect to distance. The ability to cover long distances better than Ethernet devices makes using
serial devices much more attractive. The reason for higher reliability is that serial devices do not allow for the capability to be
hacked into like an IP/Ethernet device. The serial communication protocol does not allow for this type of network
infringement because of the dedicated physical resource of the protocol, unlike the IP/Ethernet protocol which uses non-
dedicated physical connections and Over-The-Air communications that can sometimes be compromised. Fig.3 shows how the
RS232 serial protocol communicates between devices or hosts.

Fig. 3. RS232 Serial Link

The components of a serial link are the UART, the serial channel, and the interface logic. An interface chip known as the
universal asynchronous receiver/transmitter or UART is used to implement serial data transmission. The UART sits
between the host computer and the serial channel. The serial channel is the collection of wires over which the bits are
transmitted. The output from the UART is a standard Transistor-Transistor Logic/Complementary Metal-Oxide-Semiconductor
(TTL/CMOS) logic level of 0 or 5 volts. In order to improve bandwidth, remove noise, and increase range, this TTL logical

level is converted to an RS-232 logic level of or volts before being sent out on the serial channel [2]. With the
consideration of distance between host or end devices or controllers and the idea of which physical connection to be used in the
wireless network, a better understanding of frequencies is the next step in finalizing a true network solution.

II. LICENSE VS UNLICENSED

There are really only a few reason that separate the choice between having a licensed wireless network versus and
unlicensed network. One of the first things to note is the greater distance licensed frequencies have the ability to cover.
Frequency is the most fundamental characteristic of a radio signal that describes the rate and indicates the number of times per
secon that a signal cycle comes from its maximum value to its minimum value and then back to its maximum value [3].
Microprocessor based protective relays can greatly utilize the RF of a wireless network since decisions are based on a
calculated magnitude and angle of the input waveform [4].
In the topic of using a licensed or unlicensed frequency, things for consideration are costs of the licenses (for licensed) and
the ability to have a virtually privately owned network or whether to use an unlicensed frequency that reduces costs since
unlicensed frequencies have no fee. The fees for licensed frequencies are made by the Federal Communications Commission
(FCC) for the United States and such similar entities for other non-American countries. The fees for a pair of licenses can vary
from a few thousand dollars to tens of thousands of dollars. Typically the FCC will provide a pair or 2 frequencies for a
licensed system. These systems are often referred to as MAS because there is usually a Master/Host location that provides RF
coverage to multiple Remote/Controller devices. Each of these locations and devices are considered an address in the network.
The pair of frequencies is needed so that there is a frequency to Transmit data and the second frequency to receive data. This is

978-1-4799-3294-8/14/$31.00 2014 IEEE

 2014-CIC-0050 Page 4 of 7

referred to as a network operating in a duplex mode where the devices can talk on one frequency and listen on the other in
order to prevent data loss or data collision. The duplex mode is not required, but often desired. There are occasions when
there is only the need to use one frequency to both transmit and receive on. In this case, this wireless system is operating in a
simplex mode.
An example of a typical licensed frequency pair is 928.01234 and 952.01234 for a 900MHz license request with the entire
900MHz licensed band ranging from 820MHz to 870MHz, 880MHz to 915MHz, 926MHz to 936MHz and finally 928MHz to
960MHz. The FCC determines frequency assignment to a customer based on the channel size request of 12.5 KHz, 25 KHz or
50 KHz. Similarly, the 450MHz licensed band ranges 400MHz to 512MHz. Ensuring you request the best license for your
network solution again requires that you understand the distances that separate your end/controller devices from the host
location. Knowing that a 450MHz license would give better network coverage is quite helpful, but also remembering that your
latency or response times on those devices will be slower than those operating on a 900MHz frequency is even more beneficial.
The biggest advantage to having a licensed wireless network is that the network is private in that no one else can operate on the
assigned frequencies provided by the FCC. If an operator is found to be utilizing an assigned frequency, the FCC will fine the
user for trespassing in your network and this is fully enforced and taken very seriously by the FCC. Other users operating in
licensed frequencies opt for this more expensive wireless solution because the application that requires monitoring or
connectivity to wireless devices are important and cannot afford the risk of having unavailability or downtime throughout the
year. This availability requirement, which is often 99.9% or better for a device to be operable and available annually, is
another motivating factor for a licensed wireless network. In a licensed network, you own the network and control when
outages can occur. Some customers will often test run for outages and have the capability to do so since they will only be
affecting their network. Considerations for having an unlicensed wireless network are slightly different versus those for a
licensed. There is still the conscious thought of costs, but there is no longer the cost of a license to be put into budgetary
projections. Unlicensed frequencies do not incur a cost by the FCC, but there are rules to utilizing this frequency band. For
utilization of frequencies of the ISM band (902MHz to 928MHz), the wireless devices must have a built-in mechanism that
performs frequency hopping across the frequency spectrum or otherwise known as FHSS (Frequency Hopping Spread
Spectrum). This particular type of technology is intuitive in the name in that the RF signal must actually have an algorithm
that will create a hopping sequence from one frequency within the 902MHz-928MHz range so that no one particular
frequency is always utilized by one customer. Many wireless devices will have such algorithm, but it is up to the consumer to
ensure any wireless device introduced into the network follows the rules of the frequency hopping technology. This
technology may seem unattractive in that the signal is constantly moving, but what makes it attractive is that the coverage is
relatively decent as if it were a licensed 900MHz band device, without incurring the licensed fee. There are downfalls in the
unlicensed system as well because the ISM band operates in a license-free arena, meaning the wireless network is one that can
be shared by many users within the network coverage. However, because everyone is constantly hopping between
frequencies, interference by other users should be mitigated by this fact.
Some wireless networks take advantage of registered band frequencies that are considered unlicensed. These frequencies
are not necessarily assigned by the FCC, but the FCC does have a particular frequency band that can only be used by these
registered frequencies. Currently, this registered band includes frequencies ranging from 3650MHz to 3750MHz. The
advantage of using a registered band is that it is similar to a licensed band in that you are allowed to have devices on one
particular frequency without having to hop among different frequencies, but you do not have to pay a fee to the FCC for any
particular frequencies. This band is also known as the WiMax band that can make end devices respond faster and have a much
higher bandwidth capacity. This high bandwidth is often needed for applications that might include having video surveillance
or any application that is supplying a large amount of data transfer between the host and the remote or controller. What makes
WiMax different from FHSS or MAS systems is that in the RF world, reflections usually will kill or drop an RF signal, but
WiMax uses Orthogonal Frequency Division Multiplexing (OFDM). The technology of OFDM uses reflections to an
advantage since the multiplexing is frequency based versus Time Division Multiplexing or time based in other wireless
solutions. Again, because of the frequency operating in an upper band, coverage distance by a WiMax device is not as far as
with licensed or even other unlicensed frequencies. One stipulation by the FCC when using a frequency within the registered
band is that each location where a wireless device is connected, must have the actual latitude and longitude of the site
registered or sent to the FCC for recording. If another user is found to be operating on a frequency for sites that are
registered with your network, the FCC does not fine these users, but will instead only inform the user that the frequency is
already in use and that a frequency plan must be determined amongst the parties involved. The registration to the FCC just
shows who asked for permission first to use a certain frequency within the registered band. This first-come first-served basis
can mean leverage when developing the frequency plan with the other party involved. The FCC does not partake in any part of
this frequency planning and requires that the two parties work the interference problems out amongst each other.

978-1-4799-3294-8/14/$31.00 2014 IEEE

Page 5 of 7 2014-CIC-0050

III. APPLICATIONS

A. Distribution Automation - Advanced Metering

There are many applications that a wireless network can enhance or even make possible. One such application
under the SCADA umbrella would be that of advanced metering. With the addition of smart meters or meters that
can be remotely maintained and monitored, there is the need for the ability to retrieve this data remotely. Fig.4
shows an example of an Advanced Metering Infrastructure (AMI) that is operating via a wireless network [5]. The
wireless devices or controlling devices are connected to meters that are located at consumers homes. These
connections can either be made with a serial connection or an Ethernet connection. The remote devices from the
home then communicate back to an Access Point or host and this Access Point will report or transmit the data back
to the network Data Center for processing, monitoring and diagnostics as required. The advanced meter reading
application has become quite popular and most cities in the United States are taking advantage of this application
with the use of a wireless network solution. With the addition of wireless devices for the AMI application, utility
companies not only have the ability to read meters remotely, but billing is now the most accurate it has been since
consumers are now billed based on actual usage and not on usage that was rounded to the nearest kilowatt as done in
the past. Utility companies also have the ability to turn a new customers service on in half the time it has taken in
the past prior to using a wireless network. Service can be turned on remotely in many instances and no longer
requires a visit by the utility companys crew.

Fig. 4. Advanced Metering Application (SCADA)

B. Protection and Control - Transmission Fault Detection

Many systems involve relays that have the ability to detect faults on a transmission line. When faults are
detected, service that is interrupted is required to be restored very quickly and accurately. This is especially true in
an urban area such as large cities that are heavily populated. In these instances, the utility crews that are dispatched
to handle these types of outages rely largely in part on a system that usually contains a line current differential. This
differential has to provide the most accurate information possible; otherwise the magnitude of inaccurate data is
huge and has proven to be quite costly. You may be wondering how a wireless device can be used to enhance this
functionality since most line current differential systems detect line on a fault and this fault is handled by an internal
system. In Fig.5, the addition of wireless devices can be added to the relays via serial or Ethernet [6]. There are
often times when the distance between substations is substantial and can only be handled with the use of an RF
signal. Not only will the remote endpoints connect the lines for fault detection, but the wireless devices allow for
the network to be monitored remotely and they system can be controlled remotely. The use of this type of remote
system will allow for a Control Center to be able to accurately and quickly dispatch a crew the fault since the
wireless network provides the full architecture of the network and can often help bring the network back to service
without a utility crew dispatch if not needed.

978-1-4799-3294-8/14/$31.00 2014 IEEE

 2014-CIC-0050 Page 6 of 7

Fig. 5. Fault Detection

IV. CONCLUSION

When designing and planning a wireless network, ensure some extensive research is conducted prior to
purchasing. This research should make sense for your network. No two networks are the same and there are no
cookie-cutter solutions. Wireless solutions have great benefits and often may seem expensive, but the benefits more
than make up for the investment costs. Wireless networks are used as a cost-reducing measure for many companies
as well as a means to promote a safer environment with the use of remote access to places that would otherwise be
deemed hazardous work sites. Some such places include those who work in the utility industry, wastewater and
landfill environments or more industrial locations in the cement and mining environments as show in Fig.6 just to
name a few.

Fig. 6. Cement Processing

Some drawbacks to building a wireless network include the actual buildup of the infrastructure such as having
lattice towers or monopoles, inner-city tower structure availability such as telephone poles or the lack of these types
of structures used for mounting antennas to be used in the network. Often times, many companies do not have
access to these large and often expensive structures, but if you do, utilize them fully as they will be the best
investment made for the network. The application for use in the wireless network is the main priority for driving a
final solution recommendation. The use of licensed or unlicensed frequency as well as the types of interfaces
needed at the controlling devices is important knowledge to build the network, but fully understanding the
application serve best to answer the "whys and whats" are needed when choosing a wireless solution for your
network.

978-1-4799-3294-8/14/$31.00 2014 IEEE

Page 7 of 7 2014-CIC-0050

REFERENCES

Technical Reports:
[1] GE Publication 05-2806A01-J MDS iNET Series Transceivers, 2010.
[2] GE Publication Fundamentals of Industrial Wireless Communications, 2008.
[3] GE Publication (2012). MDS Mercury Series Brochure, [Online]. Available:
http://gedigitalenergy.com/products/brochures/MDS/Mercury5800_web.pdf.
[4] GE Publication (2012). Fast & accurate double-ended fault location in urban environments. [Online].
Available: http://www.gedigitalenergy.com/multilin/transm_spotlight.htm.
[5] GE Publication (2012). Industrial Communication Solutions for the Mining Industry Brochure, [Online].
Available: http://www.gedigitalenergy.com/IndSolutions/brochures/ICS_Mining.pdf.

Periodicals:
[6] N. Drakos (1993). Micro-controller Learning Modules, Computer Based Learning Unit, University of
Leeds [Online]. Available: http://controls.ame.nd.edu/microcontroller/main/main.html.

Papers Presented at Conferences (Unpublished):

[7] R. Hunt and T. Smith, "Current Transformer Saturation Effects on Coordinating Time Interval", 2012
Annual IEEE Pulp and Paper Industry Technical Conference (PPIC), April 2011, p.2.

978-1-4799-3294-8/14/$31.00 2014 IEEE