Page 1 / 2   >   >>
fleet_line 12/5/2012 | 12:19:48 AM
re: PONs: Passive Aggression Can someone explain for a dummy how the downstream 155 is split optically? Say, for example, does a 155 Sonet framed stream split it's optical power and get copied to two pipes?
There's no electrical stuff going on here, so what else is it?
Do the downstream ports get provisioned to grab their "slice" of the frame, say one DS3's worth?
PONnewbie 12/4/2012 | 11:07:06 PM
re: PONs: Passive Aggression Hello all,
What physical components in a PON constitute a bandwidth limitaion of 155 Mbps or 622 Mbps?

Why not a PON in the Tbps range with only db losses causing slight bandwidth reduction?
rjmcmahon 12/4/2012 | 11:07:04 PM
re: PONs: Passive Aggression I believe the 155/622 limitations come from using ATM technologies. Others believe FTTH should be based off Ethernet technologies.


From a physics perspective, I am not qualified to answer about bw limits imposed by the application of optical splitters. Hopefully somebody else will answer and we both will learn something new today.

PONnewbie 12/4/2012 | 11:06:39 PM
re: PONs: Passive Aggression I understand there are two different flavors; ATM and Ethernet and that bandwidth is determined by the standards for each. Provided that you had a physical switching medium capable of Terabits, could you implement a PON on that infrastructure? Would it be ATM or Ethernet? Something else?

I'd welcome anyone's articulation as to how bandwidth's are determined as a standard is being created. I'm in the dark.

xlh9504 12/4/2012 | 10:14:09 PM
re: PONs: Passive Aggression It's the difficulty of burst mode technology that cause the bandwidth limitation.

It's hard to design burst mode receiver at high speed.
optodunce 12/4/2012 | 8:38:25 PM
re: PONs: Passive Aggression Recently I was doing some research on PON's and wnated to know what fiber calbe spec are most of these companies using for the FTTH (i.e. optical solutions, etc.) Is it smf-28 type or other types does anyone know?
newguy 12/4/2012 | 7:40:46 PM
re: PONs: Passive Aggression "Moreover, the use of passive optical components - ones that aren't powered by electricity - make PONs future-proof. They won't need upgrading to support higher transmission speeds to reap the rewards of advances in technology."

Question - Don't PONs use laser diodes and photodiodes for OE and EO conversion? And don't these have data rate limitations? And so if you are buying a 622Mb PON is it future proof for 1.25Gb or 10Gb?

There are passive elements in a PON but there are active elements, too.
photonsu 12/4/2012 | 7:33:00 PM
re: PONs: Passive Aggression you need some number of photons per bit, so traditional "same wavelength PONs" must double power or decrease split ratios by factor of two to double user bandwidth. it's an old idea that has outlived it usefulness, and because of its added expense, tends to hold back the market, rather than promote it.
I remember the days when companies fell like flys trying to supply those labor intensive "cheap passive couplers" to the market. The coupler vendors that have survived, don't sell the cheap FTTH one any more.
nomad 12/4/2012 | 7:27:42 PM
re: PONs: Passive Aggression The basic rationale for PONS is - optics are too expensive to do a point to point system. That basic premise is now wrong. It is now less expensive to do point to point low cost optics with simple CPE than higher cost shared optics with complex CPE. Check out the offerings from a few new companies offering low cost point to point systems. World wide packets and PurOptix are both offering more bandwidth, more services, and easier provisioning. Check it out.
zman 12/4/2012 | 7:27:38 PM
re: PONs: Passive Aggression
> The basic rationale for PONS is - optics
> are too expensive to do a point to point
> system.

I like WWP and PurOptix home run fiber focus; heck, sign me up. But do you know why ILECs and MSOs are asking for xPON networks still today?

It is not the cost of the optics, it is the cost and maintenance of the infrastructure and limited fiber resources. Last mile networks are almost all shared: Satellite, Cable Modem, Cell Phones, SONET. *ALL* of these share a limited pipe, with all traffic reaching each node. None of them are point to point - think about it.

I have a cell phone and a cable modem: on each of these, I am hit with all my neighors traffic, and my box magically filters out mine. Shared infrastructure is no big deal; they can offer the right service at the right price. xPON is the exact same, but higher bandwidth.

Page 1 / 2   >   >>
Sign In