Why I Don’t Like Wireless Networking

Originally Written 06 Mar 2003

Wireless networking appears to be emerging from a geeky toy to a viable network solution. However, it has many hurdles and several leaps and bounds to be made before I’ll personally consider it a viable solution. Two major hurdles are speed and security. Another major hurdle is survivability. Radio wave communications themselves have several drawbacks that might not affect most networks but can seriously affect large ones. The question you have to ask yourself is what makes better sense for you, copper or wireless.

Speed (in comparison with other methods of networking) has always been a drawback to wireless. The original IEEE 802.11 standard designated that a measly two megabits per second was attainable. The latest, still to be finalized, IEEE standard is 802.11g. This new standard can produce bandwidths of up to 54 Mbps. The flexibility of wireless intrigues many people but the slow speeds turn people away quickly. This is being improved upon but needs much work and only time will tell how the speed of wireless will increase. There are many environmental variables that affect the speed of wireless transmissions inside an office building or warehouse. Examples of these obstacles are walls, interference from lighting and electrical circuits, any solid metal such as conduit, support beams, storage racks (in warehouses) can cancel or reflect signals. Distance from the access point itself will also affect connectivity to your network. Typically the further away you are from the access point the lower the bandwidth of an established connection. These are all things that have to be dealt with before wireless networking really catches on.

The main reason I will not endorse wireless networking is the number of flaws in security that have plagued its existence. The security nightmare occurs when one realizes that the access point has to be protected from external or Internet intruders and that the wireless side has to be protected even more so due to the fact that access to ones network is only limited by distance and not by physical connectivity to it. This means that a wireless subnet has a back door in itself simply due to the fact that someone could be out in your parking lot with a laptop and an antenna. While the copper subnet of the same network has numerous back doors that can be protected by firewalls restricting where traffic can go and by lack of physical connectivity behind the firewall. An encryption method called, Wired Equivalent Privacy (WEP) is included in IEEE’s 802.11 standard but is so flawed that it has been deemed a mere hurdle that a hacker has to hop over before gaining access to a wireless network. There are other practices that help in establishing a more secure wireless network but all seem to be a series of hurdles that would only turn away the inexperienced or not as dedicated hacker. The most common additional practice is MAC address security. At the wireless access point (WAP) a table of valid MAC addresses is created so that only those MAC addresses can access that WAP. This works but MAC addresses themselves are easily spoofed (depending on your network card it can be changed in the device’s properties). MAC security and WEP security are two somewhat simple hurdles before entering a wireless network. The two practices combined have been called, “better than nothing.” There are a few other more advanced methods of securing wireless networks but those are the two most common. As a matter of fact the only true way to secure a wireless network is via a VPN. Another, sometimes unrealized mistake by administrators is setting up DHCP on wireless networks. This is great because any one can turn on their laptop and get all the needed network addresses to begin using the network with little to no configuration by the administrators. The bad thing about this is that DHCP on solid-state networks it requires a physical connection to that network where as on a wireless network a user only requires the right hardware and air for a connection. Security (or lack thereof) in wireless networking itself is reason enough for me to not use it.

If you’re looking for a way to secure a wireless network I would recommend reading the Project WARTA: Wireless Authentication, Routing, Traffic control and Accounting paper.

Larger wireless networks will also start running into to numerous uncontrollable issues. Bad weather can seriously degrade radio waves thus making a wireless corporate WAN or inter-building wireless network susceptible to outages. Other things such as power lines and other radio or electromagnetic wave generating objects will force a wireless network to succumb to the wills of science. Another, less heard of, problem is jamming. Jamming a wireless network in theory could be quite easy. All it would take would be a radio wave consisting of a similar frequency and higher power to absolutely scramble signals leaving a WAP. A worse case scenario situation is that a signal leaving the WAP could jam a signal coming towards the WAP if had opposite polarities or phase. These are all problems that have to be accounted for when creating a wireless network of significant geographic size.

Comparison of copper versus wireless has many positives and negatives on both sides but at the date of this article copper is winning. We all agree that wireless networks are flexible. They’re also insecure and due to a lack of available security measures highly vulnerable. Wireless hardware is also quite a bit more expensive as well. Copper is cheap; transmitters, receivers, and repeaters for wireless networks are not. Copper yields higher bandwidths that can’t even be obtained by the current standards of wireless networking. When you put speed into a cost comparison between hardware needed to establish a network based on wireless or copper topologies, wireless is clearly the more expensive network to create. Maintenance of a wireless network can carry a hefty price tag as well. Hardware for wireless networks is quite expensive in comparison to copper based networks. I will continue to use copper until speed, security, and cost have been balanced between the two. Once that happens wireless will have the advantage because of its flexibility.

It seems that at this point wireless networking’s success has been attributed to a bandwagon effect of sorts. Even though there are flaws in them, wireless networks will eventually become faster, more secure, and more stable and will become a viable networking solution in the future. Nevertheless, I will not be building any wireless networks in the near future.