The RF Crowd

Among other things, it's important to consider the RF band when building a wireless network

• By Ray Shilling
• May 31, 2007

TO take advantage of the powerful and ubiquitous TCP/IP communication platform, manufacturers of security products such as network video cameras, gate access controllers, biometric scanners, perimeter fencing systems and mobile covert monitoring solutions are gradually migrating products to a digital platform. In many cases, a wireline network connection is not available at all locations in the facility. Wireless alternatives are increasingly being considered to deploy these Ethernet devices.

With the dramatic increase of wireless the past five years, the unlicensed radio spectrum is becoming increasingly overcrowded in many urban areas. A wireless system installed today that functions well can fail dramatically next week, next month or next year simply as a result of multiple products installed and operating on the same RF band. 

Designing the System
In the United States, the three most common unlicensed ISM bands in use for commercial wireless Ethernet products are 900 MHz, 2.4 GHz and 5.8 GHz.

Prior to certifying a wireless product, the Federal Communications Commission carefully examines output power and modulation techniques of the radio transceiver. However, the FCC does not regulate the density of installed wireless products in each ISM band in the public domain. As a result, ISM bands are becoming increasingly crowded.

Before beginning the design of a wireless system, users should carefully research the current state-of-the-practice to learn more about the features and/or characteristics of today’s wireless digital Ethernet products to help to ensure a successful wireless installation. The most critical factors in mitigating RF overcrowding are to be sure to specify FCC-certified radio transceivers with a discrete fixed-frequency, non-overlapping channel architecture; avoid using frequency-hopping radio technologies that consume the entire ISM band in the area for a single application, leaving no room for others; and, where possible, deploy systems with adaptive frequency agility that can heal autonomously in the event of future new sources of RF interference introduced.

It also is important to use directional or sector antennae to focus RF energy at the target to avoid transmitting or receiving undesirable RF pollution.

It is important to understand the implementation of Ethernet or TCP/IP wireless transmission protocols. With rare exception, it is preferable to transmit content from an edge device in digital format, as opposed to analog. There are two principal reasons for the preference.

First, it is more difficult to provide adequate security for analog transmissions. Armed with a basic RF scanner, a person with less than honorable intentions can easily move to within range of an analog wireless transmission and intercept unencrypted content and extract the information. Conversely, a digital data stream can be easily encrypted and secured from unwanted intrusion. Second, a digital stream of data is better equipped to handle sporadic RF interference since TCP/IP protocols include packet retransmission routines and are designed to be fault-tolerant. When a digital signal experiences RF interference, the radio transceivers can usually mitigate the interruption by retransmitting any lost packets of data a short time later. As a result, while the digital network can be slowed down by incidental interference, it is unlikely to be completely shut down.

Extreme Makeover: Wireless Edition
Armed with this preliminary know-how, you are now ready to continue the process of designing a robust, fault-tolerant wireless solution.

There are dozens of network design methodologies and system architectures that can be followed when building a wireless network. The three wireless Ethernet network topologies most commonly used today are mesh networks, point-to-multipoint (star) networks and point-to-point bridges.

Entire textbooks have been written on the advantages and disadvantages of each network platform and how to design solutions using each topology. To avoid or mitigate RF interference in system design, consider pros and cons of the network methodologies.

By the very design, mesh networks must use omni-directional antennae and therefore are least likely to be able to resist nearby, in-band RF interference. Point-to-multipoint systems can be deployed with directional antennae at the edge of the network can be somewhat resist RF noise, but are vulnerable. The winner of the RF interference resistance challenge is the point-to-point bridge system with directional antennae that only transmit and receive in the desired direction and are less likely to pick up undesirable noise.

Focusing on Products, Expectations
When selecting a wireless hardware manufacturer, be sure the products supplied are FCC certified. Spending tens of thousands of dollars installing a system only to find out later that it is not certified can be a costly mistake for all parties involved. Ensure all products and radio-antenna combinations meet FCC power output guidelines. Installing illegally overpowered wireless devices should be avoided at all cost. Be certain that your hardware provider can assist before the purchase in providing system design and layout support. And following the installation, ask if the manufacturer provides unlimited, free telephone technical support to resolve RF interference issues or discuss ways to mitigate issues.

It also is important the manufacturer offer both U.S. and international support with personnel that have local domain expertise in the specific regulatory requirements of the country, state or province. Also examine what the return policy is in event of system underperformance or malfunction due to RF interference in the local environment.

Meeting the customer’s expectations is one of the most significant challenges in designing and deploying an integrated wireless network in the increasingly crowded ISM band space. Of course, this can be hard to achieve, even in the most ideal circumstances and time-tested technologies. However, it becomes a more important factor to consider when designing and deploying a cutting-edge wireless network in an unknown and rapidly changing RF environment.

Be sure to involve the customer in every decision about the network design. Resist the urge to surprise the customer with skill and creativity in designing and installing the network. If at all possible, the customer should be willing to sign off on the risks associated with deploying wireless technologies. They need to know up-front there are inherent pitfalls during the initial system installation, as well as a variety of yet unknown problems that may arise in the future. Many topics in this series have implications for system under-performance or outright failure if not implemented correctly. For instance, a point-to-multipoint or mesh system may be subject to outright failure if another high-power mesh network or frequency-hopping RFID scanning system is installed on a neighbor’s property.

Educate and involve the end user in the process from the beginning of the design phase through installation. Unlike wired installations—where system repairs and fixes are often a simple case of more money or labor resources—in the world of wireless technologies, there may simply be no way out, and the customer should share this risk with the system integrator.

Ever-Changing Technology
As the project owner, when determining the project budget, and creating specifications and bid documents, it is important to note several key points.

It is recommended project owners allocate funds to perform a preliminary site survey and system design. While the performance of most wired analog and digital, IP-based security devices installed today are consistent from site to site, wireless networking is highly site-specific and dependent upon local conditions. Since the installer/integrator may be subject to a dynamic RF environment at the site and an evolving technology feature set, it might be advisable to include a general integration services component to the project so an additional revenue stream is provided to the integrator to help offset costs associated with solving any challenges.

Wireless technologies also are changing rapidly. If a project design and bid process is anticipated to take more than 12 to 18 months, it is highly likely the features and performance characteristics of the final delivered products will change. So, be sure to build in some flexibility in design and bid documents to substitute products as needed.

Finally, as with any rapidly changing technology environment, feature bloat and/or creep is a common occurrence in a longer project design lifecycle. As the project owner, it is important not to continue to expand expectations of system performance in projects mid-stream. Newer, faster products are likely to be released during project implementation. However, it may not be advisable to alter existing plans and derail the progress made of the installer/integrator by asking them to change a system design to fit the new product. Instead, it might be advisable to agree to lock down the system design as of a particular date and live with the system originally contracted to build. An installer will thank you for it.