4G LTE Signal Interference

  • 1
  • Question
  • Updated 4 years ago
I am using ZoneDirector 3000 to manage ZoneFlex 7343. Recently, i am facing a problem about 4G LTE signal interference. The wireless signal still can maintain around -69 to -75 dbm. However, the throughput has been dropped below 1Mb. Do anyone has the similar problem? Do you have any workaround solution for it? It is because we can't manage or control the carrier to deploy their equipment.
Photo of vip maddog

vip maddog

  • 10 Posts
  • 0 Reply Likes

Posted 4 years ago

  • 1
Photo of Primož Marinšek

Primož Marinšek, AlphaDog

  • 413 Posts
  • 48 Reply Likes
Are you sure about this? Have you checked this with a spectrum analyzer?
Photo of vip maddog

vip maddog

  • 10 Posts
  • 0 Reply Likes
I have two sites had similar problem.
Site A: There is 4G LTE antenna beside the ZoneFlex 7343. We did the tests before and after power off 4G LTE equipment. It proves the throughput is affecting by the 4G equipment.
Site B: There is 4G LTE outdoor antenna in opposit building and we place the ZoneFlex 7343 in corridor. We did the tests before and after close the windows curtain. It proves the throughput is affecting by the 4G equipment.
I am not an expert about the wireless signal. So, I used the simple tests to prove it.
Photo of Primož Marinšek

Primož Marinšek, AlphaDog

  • 413 Posts
  • 48 Reply Likes
Well, that doesn't prove anything.

But anyway. How far is the antenna from the AP at site A? It could overpower the AP. Although that doesn't explain site B.

What freq is the LTE modem on? Buy a spectral analyzer and check. LTE shouldn't interfere with WiFi.
Photo of vip maddog

vip maddog

  • 10 Posts
  • 0 Reply Likes
For Site A, the 4G LTE antenna is very close to the ZoneFlex 7343 and we arranged the tele-company to relocate their antenna. Now, it has around 1 meter distance between antenna and 7343. The testing result is acceptable.

For Site B, I can't apply the same solution for it.

LTE frequency will be 1800Mhz, 2300Mhz or 2600Mhz.
Photo of Keith - Pack Leader

Keith - Pack Leader

  • 860 Posts
  • 51 Reply Likes
Some of our newer models such as the ZF7352 (which is similar to your ZF7343), ZF7372 and ZF7982 have special interference filters which help reduce this specific problem.
Photo of Primož Marinšek

Primož Marinšek, AlphaDog

  • 413 Posts
  • 48 Reply Likes
That's the kind of info that would be welcome sooner rather than latter :).
Photo of Michael Brado

Michael Brado, Official Rep

  • 2114 Posts
  • 297 Reply Likes
We would like 2 meters separation from 4G LTE antennas and APs if possible.

For both sites, collect AP support info with APs hard-coded to channel 1, 6, 11,
and analyze Athstats RF Histogram, showing 2 minute sample of RF interference
levels, to help determine if one of these three channels is any better/more useable.

"Good" levels would be reported with less than 5000 PHY errrors per second seen.

Histogram of PHY errors per second (pcttime in each range)
0 1-500 ..1K ..2K ..5K ..10K ..20K ..50K .100K more
0 92 4 2 1 0 0 0 0 0
Photo of Michael Brado

Michael Brado, Official Rep

  • 2114 Posts
  • 297 Reply Likes
Histogram of PHY errors per second (pcttime in each range)
0 1-500 ..1K ..2K ..5K ..10K ..20K ..50K .100K more
0.......92......4......2......1.......0........0........0........0.......0
Photo of vip maddog

vip maddog

  • 10 Posts
  • 0 Reply Likes
Thank you for both of your reply, Michael and Keith.
Photo of vip maddog

vip maddog

  • 10 Posts
  • 0 Reply Likes
and Primož.
Photo of vip maddog

vip maddog

  • 10 Posts
  • 0 Reply Likes
------ Interrupt Stats ------
34017706 RX interrupts(rx_i)
1116133 TX interrupts(tx_i)
410225480 beacon alert interrupts(swba)
59 dynamic wmm engaged(dwmm_engaged)
59 dynamic wmm disengaged(dwmm_disengaged)
2 recv overrun interrupts(rxorn)
135642 recv eol interrupts(rxeol)
90367 global txmit timeout interrupts(txto)
448919 carrier sense timeout interrupts(cst)

------------ TX/RX Stats ------------
45491 tx failed 'cuz too many retries(tx_xretries)
836671 long on-chip tx retries(tx_longretry)
146954 tx frames with no ack marked(tx_noack)
0% rts fail rate
41931 tx frames with short preamble(tx_shortpre)
7118498 rx failed 'cuz of bad CRC(rx_crcerr)
56 rx discarded 'cuz frame too large(rx_toobig)

------------ PHY Error Stats ------------
1517757176 PHY errors since clearing all stats (rx_phyerr)
1841066746 phy ofdm group
3971657447 phy cck group
279 phy custom group
0 PHY errors since clearing delta stats (1 sec)
Histogram of PHY errors per second (pcttime in each range)
0 1-500 ..1K ..2K ..5K ..10K ..20K ..50K .100K more
0 0 0 0 0 100 0 0 0 0

------------ Airtime Stats ------------
Airtime utilized (percent) since clearing delta stats (1 sec)
Busy: 4 RX: 46 TX: 1 Total: 51
Histograms of airtime (pcttime in each range) since clearing all stats
0 0-10 ..20 ..30 ..40 ..50 ..60 ..70 ..80 ..90 ..100
busy: 0 100 0 0 0 0 0 0 0 0 0
rx: 0 0 0 1 11 73 16 0 0 0 0
tx: 12 88 0 0 0 0 0 0 0 0 0
tot: 0 0 0 0 3 38 57 1 0 0 0

38947 periodic calibrations(per_cal)
Ack RSSI [combined]: 13
Ack RSSI Ch0: ctl 6, ext -1
Ack RSSI Ch1: ctl 12, ext -1
Ack RSSI Ch2: ctl -1, ext -1
Rx RSSI Ch0: ctl 21, ext -1
Rx RSSI Ch1: ctl 27, ext -1
Rx RSSI Ch2: ctl -1, ext -1
NF Ch0 -102[-108], ext -102[-104]
NF Ch1 -102[-104], ext -102[-104]
NF Ch2 -88[0], ext -88[0]
Photo of vip maddog

vip maddog

  • 10 Posts
  • 0 Reply Likes
result from site B
Photo of vip maddog

vip maddog

  • 10 Posts
  • 0 Reply Likes
I'm sorry that it should be information from Site A that the AP with 4G antenn beside.

The information of Site B is listed below:
Photo of vip maddog

vip maddog

  • 10 Posts
  • 0 Reply Likes
------ Interrupt Stats ------
313701266 RX interrupts(rx_i)
223733126 TX interrupts(tx_i)
892403106 beacon alert interrupts(swba)
30712 dynamic wmm engaged(dwmm_engaged)
30712 dynamic wmm disengaged(dwmm_disengaged)
305044 recv eol interrupts(rxeol)
1555 global txmit timeout interrupts(txto)
396720 carrier sense timeout interrupts(cst)

------------ TX/RX Stats ------------
2029502 tx failed 'cuz too many retries(tx_xretries)
140935302 long on-chip tx retries(tx_longretry)
4865279 tx frames with no ack marked(tx_noack)
116891 tx frames with rts enabled(tx_rts)
0% rts fail rate
112258941 tx frames with short preamble(tx_shortpre)
31954225 rx failed 'cuz of bad CRC(rx_crcerr)
12 rx failed 'cuz of rx unknown (rx_unknown)
1 rx failed 'cuz MIC failure(rx_badmic)
1 rx failed 'cuz HW provided wrong keyIx(rx_badkeyix)
1 rx failed 'cuz frame too short(rx_tooshort)
156 rx discarded 'cuz frame too large(rx_toobig)

------------ PHY Error Stats ------------
2340635190 PHY errors since clearing all stats (rx_phyerr)
1491713038 phy ofdm group
848295472 phy cck group
626680 phy custom group
0 PHY errors since clearing delta stats (1 sec)
Histogram of PHY errors per second (pcttime in each range)
0 1-500 ..1K ..2K ..5K ..10K ..20K ..50K .100K more
0 45 16 10 5 4 10 9 0 0

------------ Airtime Stats ------------
Airtime utilized (percent) since clearing delta stats (1 sec)
Busy: 13 RX: 19 TX: 1 Total: 33
Histograms of airtime (pcttime in each range) since clearing all stats
0 0-10 ..20 ..30 ..40 ..50 ..60 ..70 ..80 ..90 ..100
busy: 46 40 10 3 0 0 0 0 0 0 0
rx: 0 5 53 27 8 4 2 1 0 0 0
tx: 4 91 2 1 0 0 0 0 0 0 0
tot: 0 2 37 27 14 12 4 2 1 0 0

60050 periodic calibrations(per_cal)
Ack RSSI [combined]: 9
Ack RSSI Ch0: ctl 3, ext -1
Ack RSSI Ch1: ctl 8, ext -1
Ack RSSI Ch2: ctl -1, ext -1
Rx RSSI Ch0: ctl 24, ext -1
Rx RSSI Ch1: ctl 34, ext -1
Rx RSSI Ch2: ctl -1, ext -1
NF Ch0 -102[-103], ext -102[-104]
NF Ch1 -102[-106], ext -102[-104]
NF Ch2 -88[0], ext -88[0]
Photo of Primož Marinšek

Primož Marinšek, AlphaDog

  • 413 Posts
  • 48 Reply Likes
Histogram of PHY errors per second (pcttime in each range)
0 1-500 ..1K ..2K ..5K ..10K ..20K ..50K .100K more
0 45 16 10 5 4 10 9 0 0

Too much errors. The last time I've seen this much errors I had to cheange from 7341 to 7363 and use the 5GHz.
Photo of vip maddog

vip maddog

  • 10 Posts
  • 0 Reply Likes
We just deployed the solution around 1 to 1.5 year. It is quite difficult to replace all the APs from Single band to Dual band. Moreover, we can't control the frequency of guest device. As a result, we are still facing the problem in 2.4GHz.
Photo of Primož Marinšek

Primož Marinšek, AlphaDog

  • 413 Posts
  • 48 Reply Likes
I'd say it's more expensive for the customer to have a solution that can't function than to spend some additional cash for a working solution.

Spectrum is spectrum. What seems to be happening is that the LTE is triggering APs ED all the time or overpowering it or clients at site B.

I think that there was a company that produced wall paper that they claim could stop a Wifi signal from propagating through. Maybe you could try that.
Photo of Primož Marinšek

Primož Marinšek, AlphaDog

  • 413 Posts
  • 48 Reply Likes
Call the USAF and ask if they have any of that F117 paint leftover lying around.