Wireless Sensor

Mixing Multiple Wireless Technologies

December 1, 2007 By: Joel K. Young, Digi International Sensors

With the variety of wireless technologies available, outfitting an industrial or commercial environment for wireless communications may involve mixing and matching technologies to find the optimal solution. Here are the issues to consider to make your mix a successful one.


As consumers, we're familiar with pervasive wireless gadgets and comfortable with the idea of easy, low-cost wireless communications. However, we should remember that commercial needs are different; we can't always rely on a human to adjust for the "number of bars" or to initiate a "new call." We also need to account for the thousands of units in a commercial or industrial setting (in contrast to the relatively few wireless devices in the consumer space) and acknowledge that our needs for reliability and mission criticality have an impact and a cost. Can wireless communications in the commercial and industrial world be cost-effective? The answer is a resounding yes. In fact, many organizations are beginning to mix multiple wireless technologies to create the optimal commercial wireless solution. In this article we'll discuss each of the common industrial wireless technologies and what to consider when evaluating them for your application.

Wireless Technologies
The different wireless technologies available today include cellular, WiMAX, WiFi, proprietary RF, ZigBee, and Bluetooth (Figure 1). With the exception of Bluetooth, all of these technologies are commonly used in commercial applications.

 

figure
Figure 1. Wireless technologies

W-WAN: Cellular & WiMAX. There are three forms of wireless wide area network (W-WAN) technology: time division multiple access (TDMA), used in GSM and digital cellular networks; code division multiple access (CDMA), used in CDMA cellular networks; and orthogonal frequency division multiple access (OFDMA), used for WiMAX and other 4G cellular data systems. Depending on the needs of the application, there are several key issues to address when choosing a W-WAN system:

  • Data throughput, latency, and quality of service. First assess what kind of data throughput, latency, and quality of service is needed. Regardless of the technology, the higher the performance required, the higher the cost.
  • IP addresses & routing. IP addresses are often private and dynamic, so destination routing usually doesn't work. Therefore, use device-initiated connections or simulate an extended network using something like a virtual private network (VPN).
  • Data plans. You need to pay to transmit data over any of the cellular networks and there are very few unlimited data plans for a nonhuman tethered device. However, very low data-usage telemetry plans are available if you know how much you are going to use. Know your data needs and select a plan that allows pooling of data.
  • Carrier cooperation. Whoever the carrier might be, make sure that the device is supported and certified by its network. Just because a SIM card works doesn't mean that a device using that card can connect into a network.
  • There is no ubiquitous coverage. Even the largest carriers can't provide coverage everywhere, and the networks have the greatest capacity where humans are present. When deploying to a remote site, you may need to employ extraordinary measures to get coverage.

WiFi. WiFi most often refers to the 802.11 family of standards, ranging from 802.11 and 802.11a all the way through 802.11s. WiFi is used mostly as a wireless local area network (WLAN), but it is also used for some wide area network (WAN)-type access. Note that the WAN environments are really local area networks (LANs) masquerading as WANs but without any of the channel-access quality of service attributes derived from W-WAN technology. This is because WiFi is a shared bandwidth system where access is handled using a collision avoidance system (CSMA-CA) akin to a road intersection without traffic lights or stop signs—it works well when there isn't much traffic.

figure
Figure 2. Star topology

The typical architecture for a WiFi system is a star topology (Figure 2) where users associate with Access Points. Mobility is allowed between two Access Points but often becomes problematic if the Access Points are on two different subnets. Probably the most important thing to consider when deploying WiFi is that manufacturers tend to cater to consumers. The rapid turnover of chipset vendors looking for lower cost solutions may not be appropriate for long-term commercial or industrial deployments. In addition, WiFi systems are not well suited for low-power consumption applications. Matching the security policy of your environment is also a necessity.

Consider the following issues when choosing WiFi for commercial and industrial applications:

  • Plan to fit the WiFi network into the existing infrastructure. One of the key benefits of WiFi is that so many networks are already deployed (one of the most common reasons to choose WiFi). Remember that the network was probably deployed with humans in mind, so coverage might not exist where your device needs it. It's a good idea to perform a site survey to understand the system's range.
  • Interoperability starts and ends with security. If the network is already deployed, it already has a security policy. WiFi-related standards have myriad different encryption and authentication methods; any new device on the network must conform to that policy. Choose an embedded WiFi supplier who has implemented the full range of security options.
  • Choose embedded partners wisely. While this relates to security, it also means choosing a radio design that will be available for the long term. Be wary of consumer radios that will trigger embedded driver upgrades. The cheap radio may not actually be low cost in the end.

ZigBee & Related PANs. Personal area networks (PANs ) were defined originally as a network for a very small area around a person. Over time, this definition has expanded to any wireless network that is not WiFi (WLAN). Key attributes commonly associated with PANs are low power and low cost.

It is important to first separate two concepts: (a) a mesh network and (b) a point-to-multipoint network. A point-to-multipoint network is a star topology (like WiFi) where a central point maintains a relationship with each end device. The central point may or may not provide routing to other end points. A mesh network dynamically forms routes between any two points via any other routing-capable point that it sees. Point-to-multipoint networks are simple to manage, but require every end point to have a view of the central point or hub. 802.15.4 is a common standard used in point-to-multipoint PAN implementations.

In contrast, a mesh network doesn't care about a line of sight to the central point, enabling ad hoc deployment. Mesh networks are easy to deploy but may be difficult to troubleshoot if there is a problem. ZigBee is a popular mesh networking PAN standard, the latest version being ZigBee 2007. Note that there are also many proprietary implementations of point-to-multipoint and mesh networks which may better suit your application requirements.

When choosing a wireless PAN technology, there are four critical considerations:

  • Power. Will nodes and routers be battery powered or is there access to continuous mains power? Keep in mind that extreme low-power solutions usually have restricted range and duty cycles.
  • Range. How far do you need to go between RF points? This is a direct tradeoff with power consumption. When traveling long distances on a small budget, 900 MHz might be best. If power is available but the location of end points is uncertain, then a mesh network might be best.
  • Environment. Is it noisy or quiet from an RF perspective? Does the environment change based on time of day or other characteristics? Are there multiple equipment vendors? For example, if the 900 MHz spectrum is crowded, then a 2.4 GHz solution may be more suitable. If a multivendor environment is desired, then ZigBee 2007 is best.
  • Data Flow. How should the data flow? Do they always flow to a centralized point? Are there any latency restrictions? What are the throughput requirements? How often do data flow? For example, if data always flow to a reasonably close, centralized point, then a point-to-multipoint network may be your best choice.

Also consider the following when deploying a PAN for commercial and industrial applications:

  • Assess the environment. Use a PAN when no other infrastructure is present or when there are low power consumption requirements. Then consider the spectrum. If it is noisy, a frequency-hopping or frequency-agile solution is probably necessary. Is power available?
  • Test before deploying. Even though you might be deploying an ad hoc network on the fly, it is still important to identify the RF weak spots and single points of failure. This includes assessing the environment at different times of day.
  • Assess practical interoperability over the air. What level of interoperability among different devices is required? Interoperability means more than just conforming to the standard. A standard only goes so far and almost never guarantees application compatibility. Rather, they define the level of interoperability that can be expected. Application context and provisioning are very important for a practical sensor system.

Putting the Pieces Together
A practical wireless system may involve wireless components at the WAN, LAN, or PAN level (Figure 3). Select the critical components in a well-designed system by understanding the flow of data and type of service required.

 

figure
Figure 3. A mixed wireless system

Critical Decisions. There are four important considerations when defining the ideal multiple wireless environment solution:

  • Define how the application should work. Assess the end-to-end functionality, level of service, location of intelligence, and security levels. The question of functionality usually relates to whether the application is for logging, control, alarming, or a combination of all three. Level of service identifies whether the data is mission critical or best attempt (not mission critical); this helps drive where to place the intelligence. As a general rule, intelligence can be placed at the device, at an intermediate point, or at the enterprise. It is generally unwise to put intelligence everywhere, and mission critical communication almost always requires intelligence at the end device. Then overlay security on top, evaluating what happens if the system is compromised from both an access and eavesdropping perspective.
  • What infrastructure is already in place? Evaluate the environment. Determine if there are opportunities to use existing cabled communications and local power, and check for the availability of wireless infrastructure, such as WiFi and cellular signal strength.
  • Fill in the gaps. After determining application needs and evaluating the available infrastructure, it is time to complete the puzzle. This involves doing an environmental assessment, a site survey, and considering the cost tradeoffs for different deployment options.
  • Look beyond the pilot. Deploying a mixed wireless system can be complex in terms of both the environment and the deployment itself. As such, it is important to determine how the system will be deployed over many sites. Is it scalable? Is it maintainable? It's often easy to get an initial system in place only to find that it is difficult to duplicate, extend, or maintain it.

Guiding Principles. For best results, follow these guiding principles for system design and deployment:

  • Focus energy on high payback areas first. Don't try to solve world hunger from the start or you will have an unwieldy system to troubleshoot. Do, however, consider potential expansion.
  • Focus intelligence at a common level. Sometimes there is a natural tendency to build intelligence into the system at every level in the mistaken belief that it makes a more robust system. Unfortunately, custom logic and filtering at too many places make troubleshooting difficult, especially when using multiple wireless technologies. Place decision making where it is most efficient.
  • Minimize the number of vendors and/or different technologies. Multi-vendor environments are often quoted as the true benefit of standards. Remember that standards provide multiple vendor sources but don't require mixing and matching. Too much mixing obviates suppliers from responsibility because they can always point to the other vendor.
  • Match the network to the criticality of communications. Don't try to over-engineer the system. If the system communications aren't mission critical, don't try to make it that way, it will only add cost and complexity.
  • If you have a cable, use it! Wireless technologies are wonderful, but a short cable always works better.

 


Add Comment










Sensors 2015 Call for Papers



Twitter Feed

Find It Fix It Forum

Sensors invites you to join the Findit-Fixit Forum, where you can get answers to your sensing questions—concerning technologies, products, methods, applications, and services--and also offer help to your fellow engineers. The Forum covers all kinds of topics, from the basics to the extraordinary.

Join the discussion!


© Copyright 2014 Questex Media Group LLC. All Rights Reserved. Sensorsmag. Privacy Policy | Terms of Use

If you are having technical difficulties or considerations, please contact the webmaster.