However, these performances are unlikely to equate those of devices implementing generation 3 to generation 5 roaming algorithms, unless you can follow strictly the recommendations above. Building your cell design as recommended above will help to improve the roaming performances. It is not designed for real-time applications and frequent roaming. This type of algorithm, commonly categorized as "generation 2" (partly sticky), is typically designed for non-real-time applications, with a roaming frequency expected to be low. Stay there until the signal gets bad, and then move to another BSSID if it is really better". the IOS 8 roaming algorithm logic may be summarized as follows: "associate to 5 GHz if you can. *People is no longer also the number of wireless devices! Many people carry 2 or more wireless-capable devices. I had originally wanted to use one Ubiquiti UAP-AC at full power to achieve full coverage, but since my studies, that's changed. They have 20-60 people* in the shop at any given time, but they have one large wall that's clad with corrugated steel, so I've decided on 3 small AC WAPs (Ubiquiti AP-AC-LITE) for full coverage with output power set to match that of the wireless clients they'll be serving. My real-world takeaway: I'm about to deploy wireless for a coffee shop that focuses on its trendy environment and seeks customers by providing comfy chairs and private nooks, etc. It's also MUCH more beneficial for 2.4GHz deployments, where only 3 non-overlapping channels exist (assuming you're not already in a 2.4GHz-saturated area, in which case your low-power radios would cause problems with Signal-to-Noise Ratio, although you could put 5GHz radios at lower power). PS: More APs at lower power allows better coverage with less overlap, and mobile clients more readily roam to the next AP as a user moves around. Here's the question part: When I see people saying how much they love long range APs, I now wonder if it's because they think they're improving service, or cutting costs by deploying fewer APs, or if they have a real genuine benefit from the LR-APs?Īnd in other research, it appears that most APs can only handle active throughput for ~30 clients each, although they can support 2-4x more connected. So unless I'm MOSTLY serving wireless clients with strong transmit power capability (ie: laptops, wireless desktops, some full-size tablets, etc), I'm doing a disservice to my wireless network users. Just because I have a LONG warehouse that's covered by 3 WAPs at full transmit power and I can stay connected to the SSID no matter where I go in the building with high signal bars doesn't mean I can download a large file in a reasonable timeframe anywhere in the building!įURTHER, my client which is not being received well by the APs is bogging down the AP's capability to serve other wireless clients by hogging more airtime to poorly transmit traffic! I know I've fallen into that trap: If I've got quality signal from the AP, then I must be good, right? WRONG. Most of us deploy coverage-based WLANs, but that's rarely a quality deployment. I've recently finished a fairly technical manual about high-density wireless deployments, and one of the things I learned was that wireless networks are much more efficient, and clients will achieve better performance (in all cases, but especially high-density) when they're best able to transcieve, which means that a low-power handheld device must be able to effectively reach the AP. So this post is both informational and a question:
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |