Cheap wireless switch?

I want to create VLANs for three subnetworks that would link to a Linux router with the Linux router as VLAN1 (administrator) and the wireless switches as separate broadcast domains (virtual) with wired hosts configured as VLANs2-x running linux or windows, doesn't matter. if possible, i am thinking of having a separate VLAN1 if it wouldn't interfere with latency. Can i get any cheap one somewhere? really need help, the stuff i see on the net are either not what i need or big pipes and big bucks, if i could get a simple wireless switch i will be very happy. the switches need to have STP and VLAN 802.1Q trunking. thanks

Reply to
davidodimegwu
Loading thread data ...

snipped-for-privacy@hotmail.com hath wroth:

I think you might be mixing up a "wireless switch" and a "wireless access point that support 802.1q". The only real difference is that the "wireless switch" is usually centrally managed while the wireless router can be stand alone. You really only need a "wireless switch" if you want centralize management and monitoring or if you have a huge number of access points in a single building and management is a problem.

DD-WRT and OpenWRT running on cheap Linksys WRT54G routers will do

802.1q and VLAN's.
formatting link
formatting link
Note that some models running OpenWRT will NOT support VLAN's. Check the feature list before spending money:
formatting link
Reply to
Jeff Liebermann

Reply to
davidodimegwu

snipped-for-privacy@hotmail.com hath wroth:

Here's an article on the current state of the art for wireless security: |

formatting link
you're genuinely concerned about security, I suggest you adopt a holistic approach that looks at the security for the entire system which includes physical access, denial of service, and sniffing the wired part of the network.

Speed? What are your expectations for thruput? Note that wireless is a shared media system. Only one transmitter for a given system can be on the air at time in a given air space. It's like one big party line that includes other users on the channels. This is the real problem with large systems. With only one xmitter on the air at a time, the total available bandwidth is divided by the number of xmitters.

Speed is also a function of distance. The farther you go, the weaker the signal, the more the noise, and therefore the slower the data rate necessary to retain some semblance of a decent BER (bit error rate). The standard reference for receiver sensitivity is based on a BER of 1 hiccup every 10^5 bits (or 10^6 in some tests) or about 10% PER (packet error rate). See the table at: |

formatting link
6dB increase in fade margin or SOM is good for a doubling in range.

That's good. Just realize that most such wireless switches are not intended to operate as mesh networks with repeaters and such. Every access point requires a CAT5 backhaul to the central switch. If this system is intended to replace a previously functional wired system, you just put the wires back.

Not that I know about. They tend to be horrifically expensive because there are few customers that really need a wireless switch. The price scales nicely with the associated access points as these are produced in substantial quantities and are often just adaptations of existing commodity wireless access points. However, the central switch is pricey. The inability to mix and match wireless access point vendors on a wireless switch system locks the customer into a single proprietary vendor. This closed model tends to be pricey and often predatory.

Vendor list: Symbol Airespace (Cisco) Aruba (HP) Entrasys Xirrus Meru Networks Foundry Networks Trapeze (3Com) Chantry (Siemens)

formatting link
formatting link
formatting link
formatting link
formatting link
formatting link
(etc)

I would advise that you rethink your strategy. The problem with buying into proprietary systems is that they do not survive obsolescence very gracefully. Visualize buying a wireless switch system perhaps in 2002 when such systems were initially proposed. 4 years later, you would be stuck with an 802.11b only system, possibly with only WEP. The alternative would be wholesale replacement of the access points with 802.11g or MIMO capable radios. If you have confidence in your vendors ability to provide feature updates without also requiring hardware upgrades, then perhaps proprietary systems are tolerable. However, if you're planning on keeping up with the state of the art, you'll need to consider obsolescence, upgrades, and expansion costs.

Hint: Without numeric specifics, I can only supply general answers.

Your posting IP of 196.3.62.3 resolves to Nitel Networks in Nigeria. Are you sure you have the funds available to utilize a wireless switch?

Your proposed topology sounds more like an attempt to connect separate buildings to each other. This is NOT an appropriate use of a wireless switch. Wireless switches are best deployed where you need to connect a large number of users (laptops) to a central WLAN and where the high initial cost is offset by the relative ease of management and monitoring. Schools, large office buildings, government offices, "campus" like areas, and some municipal networks are good candidates. Point to multipoint connectivity between buildings is NOT.

Reply to
Jeff Liebermann

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.