High Speed The Way We Want It

The one thing we have learned over the current pandemic is that we need the internet, and the faster the better. Though cost is surely a hurdle, the amount of bandwidth available has its bottlenecks rooted from the underlying technology. Enter new technology from an Australian Research team who have claimed to have field tested internet speeds as fast at 44.2 terabits per second.

The breakthrough in bandwidth is attributed to a new optical chip that employs optical frequency combs or micro-comb, and has been published by [Corcoran et al] of Monash University. The team exploits the ability of certain crystals to create resonant optical fields called solitons and these form highly efficient optical transmission system. For the uninitiated, optical frequency combs are an optical spectrum of equidistant lines whose values if fixed, can be used to measure unknown frequencies. The original discovery earned Roy J. Glauber, John L. Hall and Theodor W. Hänsch the Nobel Prize in Physics in 2005, and though it is a relatively new field it has seen a lot of activity in the research community.

The experimental setup has a resonator with a free spectral range spacing of 48.9GHz, and from the generated optical fields or lines, 80 were selected. Using a side-band modulator the bands were doubled and eventually with 64 QAM modulation facilitated a symbol rate of 23 Gigabaud. Now at this point, the paper says that this experiment is still an under-utilization of the available resources. The extra connectivity speed may be helpful in gaming and streaming but we will be needing faster drives to get our emails attachments downloaded faster. If you are inspired and want to play with lasers and optical communications, check out the DIY Laser Optical Link.

Thanks [Anil Pattni] for the tip.

26 thoughts on “High Speed The Way We Want It

    1. Gah, bedtime and I got my teras and petas and how many feet per nanosecond mixed up, yeah, 100,000 feet. It’s a pretty big spool but I’ll get it in somewhere.

    1. Oh, believe me, state-of-the-art high-speed far-reach optical FEC is *bewildering amazing*. Also, constellation shaping, nonlinearity compensation, ADC architectures, polarization equalization…

      It takes *all* the classical digital comms tool box (and then some) to do this…

  1. “…we need the internet, and the faster the better.”

    Ehhh I think that’s a bit of a leap. I think a lot of our current problems flow from the internet, and of course as our cruel world often insists, the solution must be more of the problem.

    1. Most of our current problems have existed for a LONG time. What the internet has done is expose these problems to the world.

      The internet has created some problems but it’s been an exceptional for exposing the truth.

      1. > The internet has created some problems but it’s been an exceptional for exposing the truth.

        “Before internet I didn’t know there are so many idiots in the world.” Words attributed to polish futurologist and writer Stanisław Lem. It’s pretty accurate. I too thought that stupidity is because of lack of information access. That wasn’t it. When everyone can be heard, now idiots think their word matters and there are so many of them…

  2. I think it’s pretty clear that in the medium term it will not be the comms link that will be the problem, it will be whatever has to supply the data being the issue. I think it’s that way already, ie it’s a rare web site that can run as fast as even a 100mbit/s connection for all of it’s users…

    1. I think you’re underestimating the amount of traffic a city or a nation can put on an sea cable, or the amount of data flowing within data centers to stream netflix, youtube, … to you and everyone else at once at 9pm.

      Let’s do a rough order-of-magnitude calculation:

      Say, we have a large city with 10⁶ Netflix subscribers. 12.5% of them are online at prime time, each taking a 1080p, roughly 5 Mb/s stream on average. 5·10⁶ b/s · 1.25·10⁻¹ · 10⁶ = 6.25·10¹¹ b/s = 625 Gb/s, Netflix end-user load alone. Single city. Not a country. Netflix only, not counting Youtube, Amazon Prime, Magenta Home, “adult entertainment”, browsing, gaming, commercial usage, multimedia communications.

      These 44 Tb/s are only worth roughly 50 of these city netflix loads. And at least in Germany, we’re really still far away from offering pervasive broadband usage to every household.

      So, yeah, backbone capacity *is* a problem already: less over land, where it’s comparatively cheap to run a new fibre when you need it, and you can have an optical amplifier every couple of km, because there’s a power grid you can easily use, but for underwater applications, where the cost of running more cables is *really* prohibitive, and where it’s a real technical challenge to get power to the amplifiers you need, somewhere in the middle of the pacific ocean.

      Of course, you’re right, feeding these data water hoses is hard, too. But that’s a distributable problem, once you’ve got the fibre-speed comms ICs sorted out.

  3. Nek minit the Australian government starts implementing this, but only to each city. From there it gets distributed through a network of 60 year old + copper cables. The perfect compromise between speed, cost, and irritation to the end user. We can call it Fibre to the City (FTTC).

      1. NBN exists. 5G doesn’t in the places where NBN is already well established.

        The headline speeds that 5G promises won’t exist anywhere except in the most densely populated areas. Why? $$$$ Cost.

        Out in the suburbs, we’ll get the watered down 5G-NR version, which does not much different speeds to 4G now. ~200Mbps in “ideal” conditions, more realistically around 50-100Mbps.

        Where I am… yes, I can get 4G… a tiny little spec of a 4G signal which is mostly enough for a voice call, and passable for web browsing, but won’t set any speed records.

        Plus, 5G handsets crap out once ambient temperature exceeds 29.5°C. How many days have we had in this country where the _minimum_ exceeded this?

        No, 5G is a lame duck. Cellular certainly will be part of the solution, but it’s not the entirety of the solution, and never will be.

        1. Can happen, I’m sure there is people in these situations, but the statistics also include such cases; your link is from the UK where more than 90% of the users are well over 5 Mpbs.

          1. And the article is from Australia, where our govt ruined our lovely national fibre network to pander to the incumbent telco and use their rotten copper. It’s a bloody joke.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.