|
TRENDS From the October 2004 issue of Communications News |
Las Vegas meshes its network As one of the fastest-growing metropolitan areas in the country, Las Vegas is experiencing what most fast-growing cities do–increased traffic congestion. In order to manage that traffic better, the city is testing a relatively new concept in wireless technology–mesh networking. Implemented in conjunction with the Freeway and Arterial System of Transportation organization (FAST) of Nevada, the high-speed wireless data network is expected to facilitate traffic-management applications in the downtown area, according to Jorge Cervantes, assistant traffic engineer for Las Vegas.
The city chose mesh-networking technology from Mesh Networks, installed by Cheetah Wireless Technologies. City agencies, including public safety and code enforcement, will test the network, with the potential of expanding the coverage over the entire 58 square miles of metro Las Vegas. The solution allows the creation of a wireless backhaul mesh among wired access points and wireless routers. This reduces system backhaul costs while increasing network coverage and reliability. In addition, meshing enables wireless peer-to-peer networks to form between and among client devices, and does not require any network infrastructure to be present. In this case, clients can hop through each other to reach other clients in the network. MeshNetwork’s Enabled Architecture (MEA) offers symmetric uplink and downlink data rates of 512 Kbps to 1.5 Mbps, and provides always-on connectivity, even at highway speeds above 100 mph. The city hopes to use the system for high-speed transmission of large data files to laptops and PDAs, such as blueprints, barricade plans, access to the city’s geographic information system files, aerial photographs, service requests and work orders, and other high-bandwidth applications. “Latency per hop is a key limiting factor in a large-scale mesh system,” says Rick Rotondo, vice president of technical marketing for MeshNetworks. “To minimize per-hop latency, we use a Layer 2 routing protocol and engine. Physical layer variables, such as per-path interference and packet completion rates, can be monitored and acted on much faster when routing decisions are made at Layer 2. Key network settings, such as optimal path and highest reliable transmission rate, can be set on a packet-by-packet basis by the Layer 2 protocol. The result is a per-hop latency of three to four milliseconds.”
More than one approach, however, can minimize the degradation of bandwidth for conventional ad hoc mesh as the number of hops increases. “The primary problem with conventional ad hoc mesh is that all the mesh nodes talk to each other on the same channel,” says Francis daCosta, CTO of MeshDynamics of San Jose. “The interference and contention that result leads to the degradation. Also, with all nodes on the same channel, you can only use one radio per node to talk to the mesh, further increasing the degradation.” The solution that MeshDynamics employs is to use multiple channels and multiple radios in the backhaul path through the mesh, so that the communications on a particular hop are happening on a different channel from neighboring hops.
“Even a latency-sensitive application like VoIP can successfully be deployed,” says daCosta, “Our current two-radio 11b solution supports it with a measured latency per hop of only three milliseconds” VoIP across the wireless mesh network is a viable application for both approaches. “With a system using our technology,” says Rotondo, “sending a VoIP call over 10 hops would not add a material amount of latency.” When choosing which applications to run on a wireless mesh network, Rotondo advises that enterprises thoroughly test the particular solution/vendor chosen. “Depending on the performance of a particular mesh network,” he says, “you can then look at which applications are appropriate. Evaluation must be done with the actual applications and demand profile you expect to run on the network. Just as important, you must physically test the system with the density of users and access points that will be found in various parts of your network. Low-density tests may reveal coverage issues, while high densities may uncover interference and power-control issues.” “In the enterprise,” says daCosta, “we expect mesh networking to be used in facilities where wiring is difficult, and in campuses where the wide-area benefits of mesh come into play.” For more information from: Cheetah Wireless Technologies A difference of opinion The majority of end-users do not regard spam as a problem in their workplace (50.8%), according to a study conducted by Insight Express for Symantec. The majority of IT managers do (79.1%), however. The majority of IT managers (70.9%) also regard spam as a problem they will likely be dealing with in 2007.
Most IT managers (58.2%) say they have seen a significant increase in spam in their organization over the past 12 months, but only 35.5% of end-users say the same. Fewer IT managers (56.4%) than end-users (68.2%) describe their company’s current spam situation as under control, and more IT managers (33.6%) than end-users (23.4%) describe it as barely under control. Both end-users and IT managers are skeptical regarding the impact of state or federal legislation on improving the spam problem. Only 24.1% of end-users say it will have a big impact; 20% of IT managers indicate it will have a neutral impact, while 53.6% say legislation might have a minimal impact. The majority of end-users (71.9%) and IT managers (81.8%) agree that worms and viruses will likely get worse in the future. According to end-users, technology will have the biggest impact on stopping or slowing spam (35.5%), with user behavior considered the second most powerful deterrent, (25.4%) and legislation the third (24.1%).
|