Monash University, Clayton Campus, Department of Electrical and Computer Systems Engineering

ECE4043/5143 Optical Communications

Design Project: Project Specifications and Planning

Calculations (to fulfil the Design Requirements/Spectifications)

0SNR: use 58 equation across whole link; this is approximate only, to give an indication of link performance and required power. Compare this with simulated required OSNR, and relate this to the provided data sheet(s).

CD: this needs to be for each node (i.e. show the residual CD at each node); find the maximum tolerance to CD from a data sheet.

PMD: Find a PMD tolerance of your system off the transponder datasheet and calculate the total PMD of your link using the parameters from the fibre datasheet and device datasheet(s). Is the accumulated PMD less than the receiver’s tolerance?

Tips for project structure

Plan out the system

What components are along point-to-point link?

What channel rates are needed?

How many channels are needed? What channel spacing is to be used?

How many spans are required, with how many EDFAs? This may be fewer than nine.

How close a match to the hardware test system in our lab (max. 10 × 80-km spans) can you get?

Find data sheets for the devices used in the link

Transceivers (e.g. 100 Gb/s dual pol. QPSK), EDFAs, fibres, WDM mux/demux, etc.

Do the data sheets that you have found contain all of the information that you need?

Can you calculate-out the design requirements from these (e.g. use equations to find numbers)?

Set up link calculations

What’s the launch power vs. OSNR like (plot it)?

What is the accumulated CD?

What is the accumulated PMD?

What can you tell from these calculations, in terms of compatibility with your transceivers?

Start simulating your link using MATLAB

What’s the tolerance of your system to different impairments?

What value of laser linewidth can your system tolerate?

What OSNR gets you to your pre-FEC BER threshold?

What’s the receiver’s tolerance do you have to residual chromatic dispersion?

How do the DSP parameters (window size in phase noise compensation, # taps and error step for equalizer) affect these?

Can you find an optimum setting for the DSP parameters, given the likely impairments?

Prepare experiments in the test link

What range of set powers do you want to sweep with the EDFAs?

Are you accounting for optical nonlinearity in the fibres?

Which DSP parameters do you want to sweep or tune over?

Run initial experiments on hardware

Did the results turn out as expected?

Identify the impact of fibre nonlinearity in your test sweeps.

Explain your results in terms of your simulations.

Plan further tests to get the data you need.

Iterate testing and get final data

After analysing your initial data, plan further tests to get the results for your report.

Explain your results: suggest future tests to justify your conclusions.

Explain differences between the hardware experiment and simulation.

Suggest possible differences between the test system and expected real-world performance.

Marking

The mark allocation for this project is split into three components:

calculations showing the system works for linear fibre;

tests showing the system working in real fibre; and the reporting, which includes justification for your design and datasheets of the components you used in your calculations and simulations, and

a comparison between the real world and the testing you have worked through.

Your pretend audience is a potential customer, like purchasing manager at a large telecommunications company, whom you are trying to sell your product too. You want to convince me that your product is good and to give me confidence that you can deliver the goods (not just an emulation/simulation of unrealistic components/system). Additionally, you are to keep it in a form that is quick and easy to read (so not too long).

Any results included should be explained.