At this time it appears as though even with the MARS flat plate antennas, the PMP450 is NOT a direct replacement for the PMP320 outwards of 8 miles.
While I don't have any experience with the PMP320 system, I have used other 802.16d and e based WiMAX systems, and I've had excellent experiences with the PMP450 compared to these older systems. We do 8 mile shots quite regularly using the integrated SM's w/reflector dishes. We even have some 20+ mile shots using 2' dishes.
MARS makes good antennas, what is the gain on the antenna? Are you making sure that you have the flat panel setup in a dual-slant e.g. +-45deg orientation? Are you using this antenna on the AP side, or the SM side or both? Lots of questions here, but to sum it up, the PMP450 should provide better overall performance then the PMP320 under identical circumstances.
Our experiance is the exact oppisite in Nlos situations. The lack of more advanced arq and the processing that the 320 has for Nlos situations has shown that the 450 has less then a 25% opertuanity to be replaced. This is at less then 2 miles from the tower locaiton. This is also useing 13.3.x firmware.
Has the performance improved with the newest firmware is yes but it is not a replacement at this time. One issue that would be relitivly easy to resolve is the addition of a mid gain panel antenna to more effectivly utilize the ability of OFDM to reliaze its Nlos capavilities. A 14dbi antenna found on both the 320 and most other 802.16e CPE seems to be around the sweet spot of gain vs relective range.
The poster really didn't make it clear if this was a non or near LOS shot, or LOS. He just said he had trouble "outwards of 8 miles". He also didn't answer any of my questions regarding his deployment... that being said... if we only care about who has the best link margins under identical deployment circumstances...we can create an apples to apples comparison...
PMP320 vs PMP450 - LET'S GET READY TO RUMBLLLLEEEE!!!!!!!!
Let's assume a 10MHz channel width, which both products can do, and will also allow for the maximum 40dBm EIRP. Both system are 2x2 systems capable of mimo-a/b operation, that's a tie. As you mentioned, the PMP320 has HARQ, which is technically a bit better then PMP450's ARQ.. give the 320 a point. The stock PMP320 AP antenna and the PMP450 AP antenna are nearly identical performance-wise, so that's a tie. The PMP320 AP can push out 1.5dBm more power and still stay within 40dBm EIRP over the PMP450 (450 is limited to 22dBm @ 10MHz), so another point goes to PMP320. The PMP320 SM's integrated antenna has quite a bit more gain at 14.5dBi vs. the 450's 8dBi patch, a bunch more points for PMP320. Let's go to the judges for a tally... and assuming we're keeping it above the belt, and we're staying within legal 40dBm EIRP, the answer is... drum roll....THE WINNER, THIS ROUND, BY TECHNICAL KNOCK OUT.... PMP320!!!!
But of course... this is all just math and as I've shown under these conditions the PMP320 is going to have quite a bit more stamina...uhhh.... I mean gain and be able to negotiate links under more difficult circumstances.... but what if we want to see a fight that couples stamina and speed... stay tuned... because that's coming up NEXT!!!!!!
I believe the tables are turned however when you make a few strategic deployment changes. When you add a reflector to the PMP450's SM, you get about 6dBi gain over the 320 SM, and I think a reflector is a pretty typical deployment accessory across the 450 line. In addition, if you run a 20MHz channel width, you can bump the power up to 25dBm (both AP's and SM's). We also use very high quality 18dBi 65deg sectors with our PMP450 AP's. For the coup-de-grace, the 450 gives the 320 a black eye with a much better b/pHz ratio.. e.g. 60mbps vs 40mbps assuming 10MHz channel width. OH yeah, and the PMP320 SM's can't even do 40mbps, they can only push like 18mbps.
This is our typical deployment stratagy, and in regards to the original 'outwards of 8 miles' post, we have no problems doing this shot and getting 6X modulation and 80mbps aggregate link eval's on a 20MHz channel width. I think it's safe to say...
PMP450 IS THE NEW HEAVYWEIGHT CHAMPION OF THE WORLD!!!!
So cheesy boxing analogies aside... In regards to your post concerning non/nearLOS shots, I'll admit it, I don't have a lot of experience doing those kinds of shots because we like to avoid headaches.... that being said, based on my outlined deployment scenario, technically the PMP450 should outperform the PMP320. There's no mystery or magic (or actual boxing) going on here... just math.
3 Likes
@Eric Ozrelic wrote:The poster really didn't make it clear if this was a non or near LOS shot, or LOS. He just said he had trouble "outwards of 8 miles". He also didn't answer any of my questions regarding his deployment... that being said... if we only care about who has the best link margins under identical deployment circumstances...we can create an apples to apples comparison...
PMP320 vs PMP450 - LET'S GET READY TO RUMBLLLLEEEE!!!!!!!!
@Let's assume a 10MHz channel width, which both products can do, and will also allow for the maximum 40dBm EIRP. Both system are 2x2 systems capable of mimo-a/b operation, that's a tie. As you mentioned, the PMP320 has HARQ, which is technically a bit better then PMP450's ARQ.. give the 320 a point. The stock PMP320 AP antenna and the PMP450 AP antenna are nearly identical performance-wise, so that's a tie. The PMP320 AP can push out 1.5dBm more power and still stay within 40dBm EIRP over the PMP450 (450 is limited to 22dBm @ 10MHz), so another point goes to PMP320. The PMP320 SM's integrated antenna has quite a bit more gain at 14.5dBi vs. the 450's 8dBi patch, a bunch more points for PMP320. Let's go to the judges for a tally... and assuming we're keeping it above the belt, and we're staying within legal 40dBm EIRP, the answer is... drum roll....THE WINNER, THIS ROUND, BY TECHNICAL KNOCK OUT.... PMP320!!!!
But of course... this is all just math and as I've shown under these conditions the PMP320 is going to have quite a bit more stamina...uhhh.... I mean gain and be able to negotiate links under more difficult circumstances.... but what if we want to see a fight that couples stamina and speed... stay tuned... because that's coming up NEXT!!!!!!
I believe the tables are turned however when you make a few strategic deployment changes. When you add a reflector to the PMP450's SM, you get about 6dBi gain over the 320 SM, and I think a reflector is a pretty typical deployment accessory across the 450 line. In addition, if you run a 20MHz channel width, you can bump the power up to 25dBm (both AP's and SM's). We also use very high quality 18dBi 65deg sectors with our PMP450 AP's. For the coup-de-grace, the 450 gives the 320 a black eye with a much better b/pHz ratio.. e.g. 60mbps vs 40mbps assuming 10MHz channel width. OH yeah, and the PMP320 SM's can't even do 40mbps, they can only push like 18mbps.
This is our typical deployment stratagy, and in regards to the original 'outwards of 8 miles' post, we have no problems doing this shot and getting 6X modulation and 80mbps aggregate link eval's on a 20MHz channel width. I think it's safe to say...
PMP450 IS THE NEW HEAVYWEIGHT CHAMPION OF THE WORLD!!!!
So cheesy boxing analogies aside... In regards to your post concerning non/nearLOS shots, I'll admit it, I don't have a lot of experience doing those kinds of shots because we like to avoid headaches.... that being said, based on my outlined deployment scenario, technically the PMP450 should outperform the PMP320. There's no mystery or magic (or actual boxing) going on here... just math.
The math is not that easy. 802.16e has some mechanisms (e.g. PUSC) to work under bad conditions. Think of it reducing the channel size further to concentrate the TX-Power on a smaller Channels. This is done for very weak clients like USB-Sticks.
Of course under good conditions a 20MHz Channel with 256QAM is way better.
Well Eric, even if ste isn't amused, I got a kick out of that...
Personally, if the 3.65 GHz PMP-450 is anything like the rest of the product line, I would gladly punt every single PMP-320 we own and reuse our license grants with gear that actually works reliably. Our towers are close enough that we don't need links as long as you guys are talking about on any platform other than 900 MHz. Mostly a necessity of the terrain - no praries here!
(Actually, once they're paid for, all of the 320 is coming down and 2.4 GHz 450 is going up. All the math and channel theory in the world doesn't make up for the bottom-quality hardware and software that explodes at the slightest hint of interference and is completely unmanageable.)
1 Like
@salad wrote:Well Eric, even if ste isn't amused, I got a kick out of that...
Personally, if the 3.65 GHz PMP-450 is anything like the rest of the product line, I would gladly punt every single PMP-320 we own and reuse our license grants with gear that actually works reliably. Our towers are close enough that we don't need links as long as you guys are talking about on any platform other than 900 MHz. Mostly a necessity of the terrain - no praries here!
(Actually, once they're paid for, all of the 320 is coming down and 2.4 GHz 450 is going up. All the math and channel theory in the world doesn't make up for the bottom-quality hardware and software that explodes at the slightest hint of interference and is completely unmanageable.)
You are right. Quality and Manageability is a problem of the PMP 320 line. From what we see the problem is at the cpe side at most. So using other cpes helped in a many cases. We don't switch to 450 due it's low power. We work licensed in 3.5 which change parameters a lot. So we move on with even higher power gear. You may guess the coverage improvements. There is where wimax fulfill one of its promises. We can change vendor when one vendor drop his development.
1 Like
@ste wrote:You are right. Quality and Manageability is a problem of the PMP 320 line. From what we see the problem is at the cpe side at most. So using other cpes helped in a many cases. We don't switch to 450 due it's low power. We work licensed in 3.5 which change parameters a lot. So we move on with even higher power gear. You may guess the coverage improvements. There is where wimax fulfill one of its promises. We can change vendor when one vendor drop his development.
That's certainly true. That's not the first time I've heard something similar. If I recall correctly, back on the forum before this one, some folks had better luck with indoor SMs than the Gemtek integrated units.
Hilariously enough, yesterday morning I got an email about the PMP320 -> PMP450 tradein program. $1000 bucks back for an AP... We're checking with our sales rep to see if we can get 2.4 GHz gear instead of staying in 3.65 GHz. There's just too much noise in that band here, between in-band competition and users that can't/won't stay in their 3.5 license.
I can say the "terrible about interference" part with some authority as I've had to redo frequencies on our WiMAX deployment two or three times in the last year... so much for "licensed"! We have less trouble out of our 2.4 GHz 802.11 equipment.