- This topic has 31 replies, 10 voices, and was last updated 8 years, 8 months ago by Aabid Baba.
-
AuthorPosts
-
-
March 23, 2016 at 9:28 am #33606jyoti rainaSpectator
hi all..
can u people plz tell me how can i make my results better..i am designing a simple four channel WDM system at 10 Gbps..my fiber lenght is 150km and transmitter power i hv taken as 10dBm..i get good quality factor around 25 bt BER is very high..
it s abt 0.345 e-012..i hv to get it close to 1.
Plz can u people tel me hw i reduce it..
Thanking u ppl in advance. -
March 23, 2016 at 3:27 pm #33632Dr. Dhiman KakatiParticipant
Hi Jyoti Raina,
Your results are as expected, because when the Q-factor will go high the BER will definitely become low. and one more thing you need to be very much clear that BER or bit error rate is calculated from the no of bits that is received at the receiver (Which was transmitted from the transmitter section) by mistake or system. so for a system to be efficient it should have high Q-factor and BER as low as possible. and in you case you are getting better results. no need to worry about that. and I need to mention that you are not getting high bit error rate you can see that the power of 10 (i.e. “e”) is negative that means very low value. and BER of near to value 1 is not acceptable.If you need further improvement in your system you can attach here with your design file so that I can have a look into it.
Thanking you..
Regards,
Dhiman-
March 24, 2016 at 8:06 am #33692jyoti rainaSpectator
hi dhiman sir.. i am thankful for ur reply.
i have actually made a mistake in writing.. i have written 1 instead of writing 0.. what i want now is ber close to 0 so that my eye diagram is clearer than before.
this time my eye diagram is not that good also..so i want to make eye diagram good n also low bir error rate..quality factor is not that bad.. anyway I appreciate your efforts.
-
-
March 24, 2016 at 3:48 am #33665FAYIQA NAQSHBANDISpectator
HELLO JYOTI RAINA..
Well i think you are interpreting the results for bit error rate in a wrong way.. You know for nay system to be designed or designed already what we desire is lowest possible value for bit error rate (BER) and when the value for BER will be 1 ..it means your system performance is very poor and low value for quality factor.. U must try to get your BER as low as possible i.e close to zero not 1..
I hope u understand .
Thanks & regards
-
March 24, 2016 at 8:14 am #33695jyoti rainaSpectator
hello fayiqa naqshbandi mam.. thanx fr replying.
i already told that i have actually made a mistake in writing.. i have written 1 instead of writing 0.. what i want now is ber close to 0 so that my eye diagram is clearer than before.
this time my eye diagram is not that good also..so i want to make eye diagram good n also low bir error rate..
How cn i make my eye diagram better nw.??-
March 24, 2016 at 8:28 am #33708FAYIQA NAQSHBANDISpectator
HELLO JYOTI..
It is alright this happens sometimes while typing but anyway your system will work if the bit error is close to zero.. I twill be an ideal situation when your BER will be zero which is not possible..
To improve your eye diagram try increasing the global symbol rate or maybe power for that reason but it will introduce more itineraries..
u can also use optical amplifiers with distance upto 5m or any amplifier.. try these things and see for the results. I hope it will help.Thanks & regards
-
March 24, 2016 at 10:37 am #33719jyoti rainaSpectator
hi fayiqa mam..thank u fr the help.
i tried ur solution and my eye diagram is better than before..i tried increasing the power to 15 db and i kept edfa amplifier length 6m instead of 3m.. the ber is now 0.234 e-037 which as per aabid sir is good bcz he has mentioned that higher the exponential term good is the result.. thanx agn fr the help…can i improve it more?
-
-
-
March 24, 2016 at 4:59 am #33668Aabid BabaParticipant
Hello jyoti,
As far as your query is concerned , i completely agree with what fayiqa has stated here. What mostly we look for our system design is to work on least bit error rate and maximum quality factor pertaining to good overall performance of our system. So you should try to get it as small as possible which means higher negative raised to power term. In order words , e^-50 will give you better results than e^-12 or e^-10. I hope you find it helpful.
Regards-
March 24, 2016 at 8:17 am #33698jyoti rainaSpectator
hi aabid baba sir..thanx fr ur reply.
i have written 1 instead of writing 0.. what i want now is ber close to 0 so that my eye diagram is clearer than before.
this time my eye diagram is not that good also..so i want to make eye diagram good n also low bir error rate..
cn u plz tel me hw can i get good eye diagram vth BEr as low as possible.. i wud be thankful to u.-
April 3, 2016 at 9:26 am #34854Aabid BabaParticipant
Hi Jyoti,
I suggest you trying the things which other members have suggested you. Hope that Helps.
Regards
-
-
-
March 24, 2016 at 7:53 am #33685Naazira BadarParticipant
Hi jyoti raina..
Your bit error rate results are already better than the minimum acceptable value.
And like already mentioned by others , BER being equal to 1 indicates worst possible performance by the system.
I hope it is clear to you by now.Regards
Naazira. -
March 24, 2016 at 7:55 am #33688burhan num mina llahParticipant
Hi All
As rightly mentioned by forum members above jyoti you need to get your BER aa low aa possible and that would be by making the negative term raised to power e as high as possible and u want ur quality factor to be high. That is how you move towards having better result and a perfect eye in BER analyzer.
Regards
Burhan
-
March 24, 2016 at 8:22 am #33702jyoti rainaSpectator
hi naazira n burhan..
i hv made a mistake by writng 1 instead of 0.. i shud hv written 0 there..
anyway can u people tel me hw can i make my ber very low and improve the eye diagram.. i hv tried various things bt still no improvement in the eye diagram nor in the bit error rate.
plz suggest something to improve my results. -
March 24, 2016 at 10:43 am #33725burhan num mina llahParticipant
Hi jyoti raina
Yeah that seemed to be a mistake actually. And regarding the help about better result i would suggest you to upload your .osd file of the project you are working upon . That would give direction to our help for you in case of your project precisely.
Regards
Burhan
-
March 24, 2016 at 10:52 am #33728jyoti rainaSpectator
hi burhan mam.. i made my results good to some extent by increasing power level and the length of edfa to 6m from 3m..
still m uploading the test file.. it is for 16 channel and i hv used only 4 channels with 10 Gbps data rate.. plz check for four channels n plz rectify it fr good results..i ll be grateful 2 u..
thanx in advance -
March 24, 2016 at 1:03 pm #33735Sanjeev kumarParticipant
Bit error rate is less than 10^-9 so it is good response u are getting. as all are replying that u misinterpret this i think u would be able to move on further with your project. thank you
-
March 25, 2016 at 2:57 am #33793jyoti rainaSpectator
hi sanjeev sir.. thanx for replying.
i hv taken all the suggestions and what in infer from this is that more negative the exponential term better are the results.. i hv seen this now bcz i got my ber very negative and the eye diagram also improved by that..
so i think it was good suggestions by all..
thanx agn
-
-
March 24, 2016 at 1:19 pm #33738SAHIL SINGHParticipant
Hi Jyoti Raina,
I too agree with Sanjeev Kumar that if Bit error rate in your system is less than 10^-9…. The response is no doubt good.. Also increasing power level and the length of edfa will help you improve your results as you might have allready observed from your design…
Thanks and Regards
Sahil Singh
-
March 25, 2016 at 3:01 am #33796jyoti rainaSpectator
hi sahil singh.. thanx for ur suggestions.
i implemented all the required changes and thanx to ol u people here who gave me good sugggestions which helped me a lot in getting good results..
i hv achieved better results now by increasing power and changing othr parameters..it was very helpful indeed n thanx to ol of u.
-
-
March 24, 2016 at 1:47 pm #33743Ranjeet KumarParticipant
Hi Jyoti Raina,
I AGREE to points of Dhiman as he said Your results are as expected, because when the Q-factor will go high the BER will definitely become low. and BER of near to value 1 is not acceptable.
BER of value greater than 10^-07 is not acceptable .
I will also ask you to reduce transmitter power because 10 dbm is very high, You should reduce transmitter power to upto 4 dbm.
Here i am attaching 32 channel CSRZ WDM SYSTEM for long haul communication system.
Sometimes we could change the filter order and bandwidth to get better results.
Hope my response could help you.-
March 25, 2016 at 3:05 am #33797jyoti rainaSpectator
hi ranjeet sir..thanx for ur suggestions.
i appreciate ur help.
u said that we could change the filter order and bandwidth to get better results..can u plz tell me where we change the bandwidth of the mux..and can we choose order of the filter as we like.. i mean to any order..
i am using 4 order of the low pass bessel filter.. can i change it to any order i like or thr is sm limitation on dat.. -
April 3, 2016 at 9:28 am #34855Aabid BabaParticipant
Hello Ranjeet,
Nice to see you pulling up such efforts. (y)
Regards
-
-
March 24, 2016 at 3:32 pm #33753SAHIL SINGHParticipant
Hi Ranjeet Kumar,
You mentioned it correctly that changing the order of the filter as well as the bandwidth can give better results…. Hope it helps jyoti raina…
Regards
Sahil Singh
-
April 2, 2016 at 3:41 pm #34671jyoti rainaSpectator
thanx sahil singh sir fr ur suggestion.
-
-
March 24, 2016 at 5:03 pm #33758love kumarParticipant
hi jyoti rana .. as all said your results are quite good… if you BER make it to zero your system is consider to be a under utilized performance coz BER zero is only possible for ideal system which is not possible in practical aspect.. if still to want to get .. you can try by reducing fiber length ..increasing gain after fiber .. change photodector APD insated of PIN .. or by reduing noise in photodector .. linewidth can also be reduced ..
-
March 25, 2016 at 3:10 am #33799jyoti rainaSpectator
hi love kumar sir..thanx for ur suggestions.
i think i am using APDs only.. bt anyway i got good results bt to further improve on those i wil take ur suggestions and try to implement those on the design.. i hope the results wil be even better and very good eye diagram i ll get..
thanx agn
-
-
March 25, 2016 at 12:55 am #33765Ranjeet KumarParticipant
Hi Sahil singh,
Yeah, i am sure about changing the order bandwidth and cutoff frequency of different filter at the receivers end to get better results as per our requirement of design in WDM system.
usually changing using filter order of 3 or 4 gives better results then using order 2. -
March 25, 2016 at 1:21 am #33773SAHIL SINGHParticipant
Hi Ranjeet Kumar,
I totally agree with you that better results are obtained using filter order 3 or 4 as compared to the results obtained from using the filter order 2… I beleive that as we increase the order of the filter the more accurate the results are obtained…
Regards
Sahil Singh
-
March 25, 2016 at 2:07 am #33780Ranjeet KumarParticipant
Yeah, Sahil singh
as we increase the filter order upto a certain limit we would have better Min. BER and max. Q-Factor.
Because it filter more accurate and with high attenuation in stop band. -
March 25, 2016 at 2:13 am #33781SAHIL SINGHParticipant
Hi Ranjeet,
I agree with you that we as we increase the filter order upto a certain limit we would have Min. BER and max. Q-Factor…. Thanks for your reply…
Regards
Sahil Singh
-
March 25, 2016 at 2:47 pm #33869love kumarParticipant
hi ranjeet …. i am too agree with .. thank for making point of this..
-
April 2, 2016 at 3:43 pm #34672jyoti rainaSpectator
thank u ol fr ur help..
-
-
AuthorPosts
- You must be logged in to reply to this topic.