Today we announced the new ePMP 3000L which is an 802.11ac Wave 2 Access Point that is very well-suited for microPOP's and/or small density, small subscriber count sectors. It is a 2x2 AP supporting synchronization and compatibilty with Force 300 SM's and backwards compatible to 802.11n SM's like the Force 180.
If you're only deploying 802.11n subscribers then you won't see a benefit with an ePMP3000L AP. The ePMP3000 line employs the 802.11ac wave 2 chipset so compared to its 802.11n predecessors of ePMP1k/2k it's capable of a higher modulation mode (256QAM), larger channel size (80MHz) and the ePMP3000 is capable of 4x4 MU-MIMO while ePMP3000L is a 2x2 SU-MIMO device. Since the 802.11n radios cannot modulate up to 256QAM or operate in an 80MHz channel size they won't be able to take advantage of the new features of the ePMP3000L.
On a related note, if you were to deploy an ePMP3000 AP with a 4x4 MU-MIMO antenna it would be able to beamform to the 802.11n SM's and grant a 3dB gain on the UL due to array factor gain.
The ePMP 3000L is already released and shipping from the factory. Cambium resellers may only get their initial stock later this month.
The backward compatibility software is released on the AP and in final beta on the SM. We expect to have full final release around the end of April.
- Bruce
For what it's worth, we have ePMP3000 AP's connecting happily to ePMP1000 (1st generation connectorized) and to Force180 / Force190 / Force200 (all running 4.3.2-RC15 in 'compatibility mode') with no apparent issues. There may still be some performance/stability things being worked on, but for my 2c the RC15 is working pretty well for us so far.
For those of us in the US, what are its FCC approvals like, or when will they be coming?
- When will it have DFS?
- What is the maximum allowable EIRP on U-NII-1 and U-NII-3?
- How close to the band edge can it operate -- can a 20 MHz channel go down to 5165, or up to 5840?
- Once it has DFS, will it operate on those DFS-overlapping frequencies like 5720 and 5725?
The 3000 line seems to me to be behind the 2000 line in these regards, so not as well suited for shots that require either high EIRP or noisy places that need to avoid the popular Wi-Fi frequencies. Has that been solved?
One more thing to add about the ePMP3000L that is in the spec sheet that may not stand out is that it is compatible with the RF Elements Twistport Adaptor and supports all the RF Elements horns. This is a nice scenario we did not have before.
In mixed mode of 2000 and 3000 CPE, will MU-MIMO occur between the 3000 CPE?
Assuming you're using an e3k w/OEM antenna as the AP, then yes, MU-MIMO can be triggered by optimal placement of AC clients. Legacy clients, like the 2000, won't be able to participate in MU-MIMO. Please note that the 3000L in AP mode cannot be used to provide MU-MIMO to clients as it lacks the extra MIMO chain that the e3k AP has.
If you're only deploying 802.11n subscribers then you won't see a benefit with an ePMP3000L AP. The ePMP3000 line employs the 802.11ac wave 2 chipset so compared to its 802.11n predecessors of ePMP1k/2k it's capable of a higher modulation mode (256QAM), larger channel size (80MHz) and the ePMP3000 is capable of 4x4 MU-MIMO while ePMP3000L is a 2x2 SU-MIMO device. Since the 802.11n radios cannot modulate up to 256QAM or operate in an 80MHz channel size they won't be able to take advantage of the new features of the ePMP3000L.
On a related note, if you were to deploy an ePMP3000 AP with a 4x4 MU-MIMO antenna it would be able to beamform to the 802.11n SM's and grant a 3dB gain on the UL due to array factor gain.
In regards to the lack of gain for a 802.11n network
Knowing that current 1000 AP's have a max frame usage which can limit the total possible throughput. Would the 3000L see the same max throughput? Or would it allow a higher total throughput in the exact same setup?. So if i have 20 customers on the panel with all near 14MCS and link tests of something along the 70/10 lines, but realistic we only move a total of 60ish mbps at 100% frame usage. Will the 3000L allow more total? I know the 3000 do not show a frame usage, so i cannot compare to the 3000's we have deployed
@skullzaflare @cambium_Brian I’m interested in knowing that as well. Would be nice if we can give some sectors that have only N clients a bump in capacity before moving it to a full 3000.
Since 3k has higher PPS and “half the latency” vs N series ePMP, one could assume that a 3000L AP with only N clients will add some capacity vs the N AP, but how much…
