Category Archives: Technology

The Risk of Space Collisions

A recent article in Scientific American looked at the likelihood of collisions between satellites and other devices in orbit today and in the foreseeable future. This is becoming an increasingly important question because of the increasing number of satellites. There are currently 18,000 satellites in orbit, and 29,000 total objects, including spent rocket bodies, dead satellites, and large pieces of space debris. Goldman Sachs research says there could be as many as 305,000 satellites in orbit by 2031 and 1.7 million by 2036. Even if the actual numbers are half that, it means a huge number of orbiting objects in space.

As we put more satellites in space, the chance of collisions increases. This matters because major collisions can be disastrous, as colliding satellites break up into smaller debris that can act as orbital bullets to smash other satellites. Scientists have postulated many scenarios where some parts of space become unusable. Space debris generated at the lowest orbits can fall out of space after a few years. However, the article cites a scientist who says that debris at higher altitudes stays in orbit a lot longer. He said that at 1,200 miles above Earth, debris can persist for 100,000 years.

One prediction in the article is that there is already a 29% chance of collisions per year at altitudes between 520 and 605 kilometers since there are numerous discarded rocket stages the size of school buses at those altitudes that can’t be maneuvered to avoid a collision. The danger  of collisions at those higher orbits is that the debris will rain down and endanger all other satellites below. This might not create dead orbital zones, but it would greatly increase the risk of operating a constellation and would compel satellite operators to send excess satellites into orbit in anticipation of losing some of them.

Perhaps the biggest concern of all is the process of monitoring satellites to avoid collisions. Until recently, this process was done manually, but this now requires automated tracking of satellites to predict collisions. In the last year, Starlink says it performed 300,000 adjustments to avoid collisions, which works out to 40 maneuvers per satellite per year. Scientists worry that this will become impossible at some point as the number of satellites grows. It’s particularly worrisome since many of the counties and companies that plan to put satellites in orbit may not interface well with rival companies and countries. There’s nothing that compels them to do so.

A really scary scenario is the aftermath of a major solar storm that knocks out the electronics of many satellites. We’ve had several such solar storms just in the century we’ve been tracking solar events. A satellite that is disabled by a solar storm would become a dead object that would not be able to maneuver to avoid collisions. The article cites an expert who predicts that a killer solar storm would lead to major satellite collisions within days.

The biggest concern of all for me is that nobody is governing satellite traffic safety. The International Telecommunications Union (ITU) in Switzerland, which is part of the United Nations, approves worldwide spectrum for satellites and loosely coordinates assigned orbital slots. However, the ITU doesn’t consider the capacity of a given orbit to handle satellites as part of its review process. With dozens of countries planning to launch satellites, space is becoming the Wild West. It might take a major event to get the world interested in doing the hard work needed to make this work.

Wooden versus Metal Poles

I recently saw a question on LinkedIn that asked why we don’t use more metal utility poles instead of wooden ones. It’s a good question, and there are a lot of issues to consider when choosing between the two.

One issue to consider is useful life. On paper, the useful life of a wooden pole is thirty to forty years. However, there are plenty of poles that have lasted fifty and even sixty years. The local conditions have a lot to do with pole life. Problems like rotting wood, constant moisture, fungi, and woodpeckers can lower the useful life of a wooden pole. Steel utility poles are expected to last fifty to sixty years and can last up to eighty years. The main problem that can shorten the life of metal poles is salt and high humidity in the environment.

Another important issue is cost. Wooden poles have gotten a lot more expensive in recent years. I see prices online today between $250 and $1,300 for a 40-foot pole, with the price varying by thickness and load-bearing ability. Steel poles are more expensive, with a typical price for a 40-foot pole between $2,000 and $3,000. It’s cheaper to ship wooden poles since they weigh less than steel poles of the same height. Installation costs are generally less for wooden poles. Construction contractors and utilities already have the equipment and experience to install them. In most applications, wooden poles can be directly buried in the ground. Steel poles routinely need a foundation, which adds time and cost to the installation process. From a pure cost/benefit perspective, when looking at the combination of the initial cost versus the life, wooden poles have a lower cost per year than steel poles.

But there are other factors to consider. For example, wooden poles need to be inspected more often during their lifetime. One interesting issue to consider is that many poles don’t make it to the end of their useful life. Some are destroyed by storms or by vehicle accidents. Poles sometimes have to be relocated during road moves. A lot of poles have been replaced in recent years when they didn’t have enough room to add fiber, requiring a change-out to a taller pole. The extra useful life of steel poles is negated for poles that have to be replaced, and in that case, the lower upfront costs favor wooden poles.

With that said, there are a lot of places where a metal pole makes sense. If you’ve never noticed, poles that carry significant loading from cables usually are supported by guy wires that go from the pole to the ground (shown in the picture above). The guy wires support the poles in times of heavy wind. It often makes sense to place a metal pole in places where there isn’t room for the guy wires. Metal poles are also often preferred when really tall poles are needed.

An interesting issue to consider is emergency repairs on poles. Linemen use strap-on tools like spikes, gaffs, or climbers that allow them to climb a wooden pole. Working on a metal pole requires a bucket truck. After an emergency that affects multiple poles, it’s far more efficient for making repairs for linemen to be able to climb poles without having to wait for an available bucket truck for each pole.

When considering all of the factors, it’s pretty obvious that utilities still prefer wooden poles over metal poles. But you generally don’t have to search very hard to find places in urban markets where metal poles were the right solution.

Adoption of WiFi 7

Ookla recently published a research article that documents the implementation of WiFi 7 in the U.S. and around the world. WiFi 7 routers were introduced to the market in early 2024, and Ookla reports that deployment of WiFi 7 routers has already grown to 7.2% in the country. That’s a pretty big uptake in a year and a half, considering that broadband customers generally hang on to WiFi routers unless changing service providers.

It’s not surprising that WiFi is being adopted. WiFi 7 brings a big improvement over WiFi 6 and earlier generations of WiFi. WiFi 7 can theoretically reach much faster speeds, and in practice delivers faster speeds and performance inside the customer premise. Some of the specific benefits include wider data channels inside the home, particularly for customers who have devices that can use 6 GHz spectrum, which is 1,200 MHz of bandwidth between 5.925 GHz and 7.125 GHz. WiFi 7 has better data throughput to devices by the use of 4K-QAM, has better performance from preamble puncturing that can bypass interference, and has faster speeds for specific applications due to multi-link operations that let the router combine multiple frequency bands when needed. WiFi 7 has also doubled the number of simultaneous devices that can be connected from eight to sixteen.

As might be expected, the growing WiFi 7 adoption is mostly coming from ISPs. Charter is currently the biggest user of WiFi 7 in the world. In the U.S., the new technology is being deployed by other large ISPs like Comcast, Frontier, and CenturyLink. It’s likely that the upgrades to WiFi 7 will continue. For years, ISPs have been frustrated by a large percentage of customer complaints about broadband that are actually due to poor customer-provided WiFi modems.

Interestingly, the U.S. is leading the world in WiFi 7 adoption. Part of the reason for this is that China is not stressing the use of WiFi 7 inside the country since the nation has decided to use 6 GHz spectrum for cellular traffic instead of for public WiFi.

The impact of the WiFi 7 routers is dramatic. Ookla cites speed test statistics for Comcast that show that upload speeds have more than doubled for some customers strictly due to the installation of a new WiFi 7 modem.

WiFi technology isn’t sitting still. The first prototypes of WiFi 8 modems should be produced later this year, with commercial production is about two years. WiFi 8 adds even more features that will help the customer experience. The most interesting new feature is Dynamic Sub-band Operations (DSO) that will allow the WiFi router to assign tiny bandwidth channels instead of a full channel when connecting to devices that don’t need much bandwidth. WiFi 8 also has the ability to coordinate with neighboring WiFi routers to cut down on interference. There will be a new technology called Distributed Resource Units and ELR that will improve the upload path to cut down on upload stutters from devices like security cameras and smartphones.

There is even more improvement coming with WiFi 9. The official specifications should be released in early 2027, with commercial adoption likely coming in the early to mid-2030s.

Cable Companies Continue to Upgrade

Jeff Baumgarner of Light Reading wrote an article detailing increased spending by cable companies as they continue to upgrade networks. The article notes a 40% increase in spending for the deployment of distributed access architecture (DAA).

Distributed Access Architecture is a network architecture that decentralizes cable networks by moving some of the brains and related functions to neighborhood nodes. Historically, cable company networks packed all of the network electronics at a centralized headend. There are significant benefits of moving broadband equipment into neighborhoods. The DAA upgrade is often accompanied by reducing the dnumber of customers on each neighborhood node, which alone increases the bandwidth distributed to the remaining customers. The upgrade to DAA generally means more overall bandwidth when cable companies upgrade to 10-gigabit bandwidth to feed each DAA node. In many cases, the transport reaching nodes still uses analog technology, and upgrading to a digital DAA improves bandwidth efficiency. Customers benefit from improved latency due to being closer to the core.

Part of the reason for the 2026 spending is that cable companies put network expansion plans on hold in 2025, waiting for the release of new Broadcom chips that enable the network to be expanded to 1.8 GHz of bandwidth. This higher bandwidth is enabling cable companies to significantly increase customer upload speeds by upgrading to symmetrical bandwidth with DOCSIS 4.0 or by using upgrades referred to as mid-splits to increase upload speeds on DOCSIS 3.1.

A lot of upgrades to DAA are restructuring cable networks for the future by using remote physical layer architecture (R-PHY) to move the modulation and demodulation functions to the neighborhood node. Remote MAC-PHY relocates both the PHY layer and the processing MAC layer to the node. This new configuration means the only thing left at headends are servers, switches, and routers, and opens the possibility of doing away with much of the headend and migrating the switching function to a regional data center.

Baumgartner notes other upgrades being made by cable companies. He notes that cable companies are integrating PON fiber technology into the network to serve new growth and rural markets. He quotes Jeff Heynen or Dell’Oro saying that sales of PON nodes to cable companies are up 71% year-over-year.

Baumgartner also provides some updates on DOCSIS 4.0 deployments. He says that Comcast is still leading the charge on DOCSIS 4.0 upgrades and has already upgraded millions of premises. He says Charter has plans to upgrade 35% of its footprint to DOCSIS 4.0 over the next few years and that it plans to accelerate the upgrades when it completes the merger with Cox. He also notes that Mediacom Communications is deploying D4.0 in some markets.

The one downside for vendors is that cable company spending on customer CPE is down 5% this year, probably reflecting the continuing loss of customers by every big cable company.

The reason for these upgrades is clear – customers still have more trust in fiber than in cable company HFC technology. PC Magazine recently conducted a nationwide survey that ranked customer satisfaction with ISPs. NextLight, a municipal network in Longmont, Colorado, got the highest rating of 9.7, followed close behind by GFiber at 9.4. The other big fiber ISPs ranged between 7.8 for Frontier to 8.3 for AT&T. The big cable companies ranged from 6.4 for Comcast to 6.8 for Optimum.

Is AI Changing Traffic Patterns?

Mitch Wagner of Fierce Network recently published an article that claims that AI is blowing up 30 years of traffic network assumptions. He claims that AI traffic is smoothing the daily peaks and dips in network traffic that all ISPs are familiar with. While every ISP is a little different, any ISP that serves a lot of end-user customers expects traffic peaks in the evening, smaller peaks during the daytime, and very low levels of network traffic at night.

Network engineers have always paid close attention to the peaks, which were the main factor in determining the size of network connections. Nobody wants to have a network that restricts bandwidth when customers want to use it the most. The cable companies learned this lesson the hard way during the pandemic when customers suddenly needed to work and school from home and found the broadband connections unable to meet their needs, particularly in homes where more than one person wanted to use the network at the same time. Every network engineer I know can cite the busy hour, busy day, and busy week on the networks they manage.

Wager says the peaks in traffic are evening out and that networks are seeing a more consistent demand throughout the day. Wagner cites Ed Fox, the CTO of MetTel, from an interview given for a Fierce Network Research report.

I have no reason not to believe Mr. Fox. However, MetTel is an ISP that operates in major urban centers and likely serves the kinds of businesses that have become heavy AI users. It’s not hard to imagine that an urban ISP serving businesses might be seeing a drastic change in traditional network traffic patterns. But I have to think that MetTel and other urban business-centric ISPs didn’t have the same traffic patterns as other ISPs before the advent of AI.

I work with a number of ISPs and I have not heard anybody talking about a big change in traffic patterns. My clients work in a variety of markets, from rural to urban, but none concentrate on the urban market business that MetTel is experiencing in places like New York City.

However, I appreciate the article, because if it’s true, then most ISPs, except for fully rural ones, might eventually see some shift in traffic patterns due to AI. I am curious to get feedback from this blog from IPS to hear if anybody is seeing anything like the big changes being experienced by MetTel.

The Wagner article also claims that upstream bandwidth is growing faster than downstream, reversing a 30-year design assumption built around heavy downloads. This is something that has been well documented by OpenVault. They’ve been shown that upstream usage has been growing faster than download usage starting in 2024. In the first quarter of 2024, national average upload usage increased by 13.2% compared to 7.7% for download. In the first quarter of this year, average upload growth was at 19.8% while download growth was at 7.8%. OpenVault credits most of the growth in upload usage to computer software synching with data centers.

Wagner claims the upsurge in upstream usage comes from AI inference traffic moving towards the edge, particularly for video processing. What he means by that is big growth coming from uses of video cameras for functions like AI-driven video surveillance at retail locations, camera-equipped wearables, and cloud-based operational technology in applications like oil and gas asset management. I hope that the big national companies that monitor traffic begin tracking this issue. I’d love to hear more about the trends in specific traffic, like video surveillance and wearables.

Network engineers all understand that upload traffic is usually a tiny fraction of download usage, with average download usage a dozen times more than upload. With the possible exception of cable companies, upload usage is largely an afterthought for network engineers, who barely consider it when sizing and designing networks.

I have no doubts that there are localized situations where AI traffic is making a big difference in network traffic. But for ISPs that mostly serve residential and small business customers, I still have to think it’s a tiny, possibly unnoticeable blip.

Network Challenges from AI Traffic

Cisco recently published a report that looks at the impact of AI traffic on networks. It’s an interesting paper because Cisco found that AI traffic does not operate the same as most other web traffic. While the volume of AI traffic is small today, Cisco predicts that we’ll have to make changes to the web over time to accommodate growing AI traffic volumes. Cisco predicts that by 2035, one fourth of all web traffic will be AI agents and AI models in data centers.

Cisco notes that we’ve spent decades optimizing a web that delivers burst traffic, like video. When a video is viewed from the web, the data stream doesn’t have to be delivered evenly in real time. Instead, all that is needed is for the transmission of the video to reach the viewer before they are ready to watch it. Anybody who has watched a video can see that the streamed video is always working to to stay ahead of what you are watching.

AI traffic is very different. Cisco uses the term AI inference traffic to mean the real-time transfer of AI traffic between the AI models operating in data centers and users. AI inference traffic is delivered at what Cisco calls software speed, meaning the receiving end is ready to digest and use the data as it is delivered, quite different than streamed video that is only trying to stay ahead of a viewer.

The difference between AI traffic and normal web traffic is significant. The typical burst of AI traffic lasts twice as a typical burst of video data. While individual bursts of video data are smaller, the flow rate for video, which means the actual delivery time, lasts ten times longer than AI traffic, since video bursts are spread over time in multiple small bursts.

AI traffic is also two-way and requires a good upstream connection. In fact, Cisco found that 9% of AI traffic requires more upstream traffic than downstream traffic. Cisco believes the need for network upload speeds will increase as AI agents mature.

Current network latency is not a bottleneck for AI traffic, but Cisco says latency will become a problem as the volume of AI traffic increases. This will require a major rework of web architecture when latency becomes an issue.

Cisco found that tasks performed by AI generate 450% more traffic than the same task performed in a more traditional way. In Cisco’s vocabulary, AI agents act as power users and use a lot of network resources.

The bottom line is that AI traffic is different from current web traffic and will not only increase traffic volumes on networks, but it will also change the shape, symmetry, and needed priority of traffic.

There have already been discussions of creating a private web to connect between AI data centers. But that would only solve part of the problem, because AI traffic is eventually delivered to users throughout the web. AI traffic is going to create an interesting new set of challenges for network engineers, something that nobody envisioned just a few years ago.

Life Left in HFC Networks

There was a time when it seemed certain that cable companies would have to bite the bullet and spend the money to upgrade to fiber. While there have been some upgrades by cable companies like Cox and Altice, most cable companies seem to be deciding that there is still good life left in DOCSIS cable networks. As you might expect, CableLabs has been quietly working behind the scenes to improve existing HFC technology.

DOCSIS 3.1 networks have become standard across the industry, and it’s now rare to see older technologies except for some small cable companies. Cable companies have been using DOCSIS 3.1 networks to deliver gigabit or faster download speeds, with the top speed depending on the overall size of the bandwidth being utilized in the internal radio network that controls the signal.

The whole cable industry got a shock during the pandemic when it became obvious to many of the millions of students and employees who began working out of homes that the slow upload speeds on DOCSIS 3.1 were a bottleneck. I think the inadequacies of this technology and slow upload speeds are what gave a big jumpstart to the public perception that fiber is far superior to cable technology.

CableLabs and vendors responded to the upload speed bottleneck by introducing two solutions that can add to the upstream portion of the cable network. Labeled as midsplit or highsplit, both solutions require some upgrades in the outside plant electronics, along with upgraded cable modems in homes. The midsplit upgrade is accomplished by increasing the frequencies used to support upload from 5-42 MHz to 5-85 MHz. The highsplit upgrade allocated even more frequency to uploading, as much as 204 MHz. Both of these upgrades increased upload speeds to 100 Mbps or faster, which eliminated the bottleneck for the average customer. In 2025, CableLabs offered an even better version of the midsplit upgrade by offering a new cable modem that can handle up to four more channels of bandwidth.

CableLabs is also improving DOCSIS 4.0 technology. This is an upgrade that became available for cable companies in early 2024 that can provide symmetrical broadband speeds. While the upgrade can deliver speeds up to 5 Gbps, most cable companies are using it to offer symmetrical 2 Gbps broadband. This upgrade makes it practical for a cable company to say it can match fiber speeds – or at least it did in 2024. There are now fiber ISPs offering residential broadband at speeds up to 10 Gbps.

CableLabs has demonstrated an upgrade to DOCSIS 4.0 that can mimic the faster advertised speeds of fiber providers. CableLabs recently released a new standard it is calling DOCSIS 4.0 Optional Annex. This standard works by increasing the network bandwidth inside the coaxial cable to 3 GHz. Cable networks operate by using radio frequencies inside the coaxial wires. Most DOCSIS 3.1 networks use 1.0 to 1.2 of total frequency. Some companies have upgraded to 1.6 GHz for DOCSIS 4.0. This new optional Annex, double that bandwidth and will supposedly support speeds up to 25 Gbps. CableLabs is also looking at a version of the new technology that would increase total network bandwidth to 6 GHz, which might support broadband speeds up to 50 Gbps.

These new options will give pause to any cable company thinking about upgrading to fiber. These new technologies provide a realistic alternative to fiber with DOCSIS 4.0.

Who Needs Moore’s Law?

I ran across a reference to Moore’s Law the other day. This was named after Gordon Moore, an engineer who later became one of the founders of Intel. In 1965, Moore observed that the number of transistors that could be squeezed into a given area of a circuit board was doubling every two years. He predicted this trend would last for perhaps another decade, but the microchip industry kept fulfilling his prediction for over fifty years, and the general consensus is that the Moore’s Law prediction died somewhere between 2016 and 2018. Chip density has continued to improve, but at a slower rate, and there is general consensus that we are getting close to reaching the maximum possible density of transistors, limited by the law of physics.

For a number of years, there were a lot of predictions that the end of Moore’s Law would mean the end of faster computing. For decades, our devices became obsolete every few years as the next generation of faster chips hit the electronics market. But chip makers have discovered a wide variety of ideas and technologies that continue to improve the speed of computing.

Consider some of the techniques that continue to improve computing power:

  • ASICs (Application-Specific Integrated Circuits): These are specialized chips designed for one specific task. For example, there are ASIC chips in data centers that are specifically designed to handle specific tasks related to processing masses of data.
  • GPUs (Graphics Processing Units): These chips take the opposite approach of ASICs and are designed to handle parallel tasks and tackle multiple functions at the same time. First designed for gaming, a GPU breaks a task into many independent threads and executes each thread using a different small core. GPUs keep improving as chip makers and software designers find better ways to coordinate and control the many threads.
  • 3D Stacking. Another new technique is stacking transistors and memory vertically as well as horizontally. This provides a huge boost to computing power. One of the more widely used 3D stacking technique is the use of chiplets, which are small chip components that can be stacked and fused to create a multi-layer chip. Chiplets benefit from Through-Silicon Vias (TSVs) that enables fast low-power communications between layers.
  • Memory Integration. One of the biggest bottlenecks for any chip is the process of moving data into and out of the chip core during the computing process. Memory integration creates temporary memory directly on the chip to store data that is still needed for the specific calculations being handled. Pulling needed information out of the chip’s own memory bypasses the normal data transfer issue.
  • Optical Computing. Optical computing uses light instead of electrons to transfer data around a chip. The primary benefit of light computing is that different colors of light can be used to allow for multiple streams of data transfer at the same time, instead of the single stream that comes from electrons. I recently wrote a blog that talked about a technology that can generate multiple wavelengths of light directly on a chip, which eliminates the need for bulky external lasers.
  • Optimized Algorithms: Some of the biggest improvements in chip speed come from rewriting software to be “hardware-aware”, meaning it perfectly aligns with a chip’s architecture.
  • Reconfigurable Computing. This is an architecture where portions of the chip can be programmed to change function or spatial configuration during the computing process.
  • Quantum Computing. Quantum computing increases computing power using qubits and the principles of superposition and entanglement. Qubits exist in multiple states instead of the two states of 1 and 0 for digital computing, which provides an exponential increase in calculation power.

We’ve just barely begun exploring many of these ideas, and there are likely many breakthroughs still to come. Picture what something like a reconfigurable architecture using multiple colors of chip-generated light waves in a quantum computer might mean.

We’re Drowning in Data

The analytic company IDC says the U.S. economy will be generating 394 trillion zettabytes of data annually by 2028 (a zettabyte is a trillion gigabytes). The majority of the energy used in data centers today is for storing some of this data in an accessible format. We don’t try to make all data available, and about 20% of the data we generate today is considered to be “hot data” that AI systems might want to draw on quickly. The remaining 80% of data is “cold data”, which we don’t put in data center storage, but which we also don’t discard, since it might still be of use in the future.

Today, hot data is largely stored on hard drives in data centers. This storage for quick retrievable uses a lot of electricity to operate the hard drives, and additional energy is used to cool the data center to offset the heat generated by the electronics. There is a growing trend of storing cold data on magnetic tapes, which also require energy for heating and cooling, since tapes are best stored at temperatures between 61 and 77 degrees. Tapes must also be replaced every 15-20 years by transferring the data – an intensive work effort.

The need to keep so much data at our fingertips to support AI means that we are literally drowning in data, and the problem is growing quickly every year. The solution to this is to find other ways to store massive amounts of data that don’t require a lot of electricity. There are several potential data storage methods on the horizon, and we’re going to need more.

One interesting possibility comes from Peter Kazansky, working at the University of Southampton in the UK. Back in 1999, while working with scientists at Kyoto University, Kazansky encountered a physical phenomenon that might provide the future for long-term data storage. The team at Kyoto found that when writing on glass with ultrafast femtocell lasers (a light pulse every quadrillionth of a second), the light traveling through the glass scattered in a way they could not explain.

It turns out that the researchers had discovered hidden nanostructures within silica glass created by micro-explosions from the lasers. The lasers had created tiny holes 1,000 times smaller than a human hair throughout the glass. The eureka moment came when researchers realized they could take advantage of this phenomenon by using lasers to print complex patterns inside the glass. After many years of research, Kazansky found that he could etch patterns in the glass that could store data in 5-dimensions – the normal x,y, and z coordinates, plus two additional coordinates related to voxels, or the scattering pattern of light.

This allows for storing massive amounts of data on a piece of etched silica glass. A 5-inch glass platter (slightly larger than a music CD) can store up to 360 terabytes of data. Unlike tape or hard disk storage, it looked like this technology creates forever memory that can be stored for the future. While energy is needed to etch the glass and encode the stored data, the process of reading the data uses light and is not energy-intensive. Kazanky founded SPhotonics in 2024 to commercialize the new storage method. Currently, the data can be retrieved at a speed of 30 MB per second, but he sees a path to reach 500 MB per second, which is faster than retrieving data from tapes.

Of course, storing data on etched glass is not without peril. A disc can break, and a fire or other disaster at a storage facility could destroy massive amounts of data, so most data will have to be stored at multiple sites. But at least the raw materials for silica glass are cheap and readily available. Probably the bigger issue facing the world is deciding how and when to ditch data that is no longer useful. Data scientists are already tackling this question today, but they are generally cautious and side with storing rather than destroying data if there is even a slight chance that it might be useful later.

Technology Shorts April 2026

The following topics discuss some interesting technologies that might someday influence the broadband industry.

Chip-level Photonics. Researchers at the CUNY Graduate Center have developed a thin, flat chip that can convert infrared light into precise frequencies of usable light that can be focused into a narrow, precise beam. The surface of the chip is patterned with tiny structures smaller than the wavelength of light. When hit with an infrared laser, tiny patterns convert the incoming light into a higher color as a narrow beam that can be steered by changing how the incoming light is polarized.

Scientists now envision a stack of different metasurfaces that could each be used to develop a different wavelength of light to use inside a chip to carry data. Effectively, this could create multiple laser light beams that were generated inside the chip without the expensive apparatus needed to inject external laser signals into each chip. Having a range of locally generated light signals could solve the problem of trying to move massive amounts of data into and out of the chip core – which is currently the biggest bottleneck to fast computing.

Dirt-Powered Fuel Cells. A team at Northwestern University has developed a device that can generate electricity by harvesting the power created by microbes that naturally reside in the soil and naturally break down organic matter. The fuel cell is about the size of a paperback book. The device has a disc-shaped anode that is buried in the soil with a second anode poking out near the surface. The device is large enough to tap the natural moisture in the soil at the bottom of the device. In testing, the device works across various soil conditions. The devices tested so far are creating 68 times more power than needed to operate the fuel cell, meaning there is a lot of power available to power other devices like agricultural sensors. The beauty of the technology is that it should work for many years without any need to replace batteries or other components like is needed for other power sources that could be used for similar applications. A fuel cell should work as long as there is enough carbon and moisture to fuel the natural microbes there.

New Laser Technology. Researchers at Tianjin University have created a new kind of optical device that can generate a light phenomena called skyrmions. The research shows that two skyrmions can be created, which are donut-shaped light patters that hold their shape – one that can be controlled by electric energy and the other by magnetic energy. The skyrmions are highly stable and resist interference, making them a good candidate for storing and transmitting data.

These new light sources could open up the use of terahertz wavelength lasers that could actively switch between light and magnetic mode, enabling a huge increase in the amount of data included in a laser transmission. This could provide the control of data flow needed to take full advantage of using a terahertz light source for data transmission. The key to making this work will be perfecting the constant flux between the light and magnetic pulses.

Forever Batteries? Scientists at the CSIRO Royal Melbourne Institute of Technology, and University of Melbourne, Australia, have developed a technology they are calling a quantum battery that can be recharged in a quadrillionths of a second, and that can work with six orders of magnitude of stored energy, making it practical for real-life applications. Unlike traditional batteries that use a chemical reaction to store and release energy, a quantum battery transfers energy using quantum coherence and collective interactions, rather than chemistry.

Normal batteries take a long time to recharge, and eventually the chemicals must be replaced, usually meaning replacing the battery. With traditional batteries, the larger the battery, the longer the time needed to recharge. Quantum batteries are the opposite, and the larger the battery, the faster it can be recharged. This is due to a phenomenon of collective effects between particles, which causes all of the storage units in a quantum battery to behave collectively.

The downside of quantum batteries is that the battery can also discharge all of its power quickly, and the challenge has been to find a way to control the output. The Australian team has demonstrated a full battery cycle from light absorption to storage to of electrical power output at room temperature and steady-state operation, showing that the quantum batteries have potential for real-life applications. The batteries can also be charged wirelessly, which opens up the possibility for remote charging.