Fierce Network TV

Advancing Connectivity: Innovations in mmWave, AI, and Network Architectures

Advancements in mmWave, AI-driven network management, and strategic architectures are redefining connectivity. In-house chipsets are playing a key role in boosting efficiency and meeting the demands of next-gen networks. Explore how these innovations are shaping a more seamless and intelligent future.


Diana Goovaerts:

Ulic, it's so great to be here with you on the show floor at MWC. So innovation is at the heart of technology and technology strategy. Could you share how your new in-house chipset designs are shaping the future of your products, and what advantages do those bring to customers?

Ulik Broida:

Okay. So even if I put aside the value of us controlling the components and not being dependent on component shortage or stuff like that when we buy components from other vendors, because now it's our own chip, the fact that we took many parts of a system like the digital part, analog part, the networking part, and we put them on one piece of silicon, allows us to have much better integration and therefore much better performance. So if, for example, if you take the E-band product that we're now demonstrating in Mobile World Congress, if you go to our competitors, you will see that usually E-band products are going up to 10 gigabit per second. Now, because we have our own chip with wider capabilities, we can do 25 gigabit per second in one box, one product. So it's more than double than our competition.

Diana Goovaerts:

Millimeter wave is playing an increasing role in high-capacity networks. Can you talk to me very briefly about how you're making the most, and making the spectrum most efficient possible?

Ulik Broida:

So as I said, our chip is more powerful. It's our own technology than other components. So this chip can today digest wider chunks of spectrum in the millimeter wave space. So if most component can digest up to two gigahertz, 2,000 megahertz of space, we can do four. So it's double the capacity. Now, because this trip has even better efficiency, it's more than double. As I said, it's 25 gigabit per second versus the 10 gigabit per second that most our competitors can demonstrate.

Diana Goovaerts:

With the evolution of networks, how do all outdoor and split-mount architectures fit into today's landscape, and what factors should customers decide between when considering them?

Ulik Broida:

So first of all, I would say, it's not this or that. It's both. I think customers and us, the vendors, we should master both use cases or both architectures. So if you have a simple link that you want to install, you will never be able to install faster and with lower cost than all outdoor. Very easy to deploy, less need of real estate on the ground, less need of air condition, power, right away, and so on, and so on. But if your link is a, I would say, a bit more complex, you have multiple carriers in one link, you have multiple directions in one site, then you would need a split-mount configuration, and you should get a, from your vendor, a good radio and a good indoor unit that can provide this solution.

Diana Goovaerts:

So AI is inescapable at this show. Can you talk to me a little bit about AI-driven network services, and network operations, and also how you guys are addressing that?

Ulik Broida:

Generally speaking, AI is good when you have many, many data points and you want to have a system that learns. So we believe AI, in our space, have two main use cases. The first one is, when you do network management and you want to do prediction, you want to do things before they happen. I would say, what is faster than real time? Anticipate. Anticipate what's going to happen. So we provide a few AI engines in our network management system, and in this way our customer can see that in two weeks or maybe two months, there will be a problem. So we can go to our customer and tell them, "Look, you don't have to wait until your customers start to complain about bad service, bad reception. We can tell you, 'Go ahead now, increase your capacity or solve your configuration problem because it will happen in two months.'" So this is one use case, network management.

The other one is root cause analysis. As networks became more complex, growing and become bigger, sometimes finding the root cause of a problem becomes a bit more complex. So our AI system really learn from the past, takes into consideration many data points, and able to tell the customer, look, most likely the problem is there, go and look for the problem. And by that, fix problems much faster with less investment.

Diana Goovaerts:

So one of the key themes here at the show is looking towards the future. So talk to me a little bit about how Ceragon plans to push the boundaries of connectivity in the future.

Ulik Broida:

So I think we should look at it at least in two folds. One is better products. So the IP-100E that we are now showcasing, as I said, more than double the capacity of anything else. Well, in the future, this is E-band, in the future, our customer will look for additional bands, W-band, D-band. These bands are not regulated yet, but as the capacity will grow in the future, most likely we will need them, and our customers. And our chip, the system of chips that we mentioned, is the best asset for this type of powerful products.

The second aspect to look into is not only products, we believe that many of our customers want to get from us, solutions. So not everybody can have the technical department that can take our products, deploy them in the optimal way, maintaining them in the optimal way, maybe even plan the network in the optimal way. So we are now creating an envelope of services that we can come to the customer, tell them, "If you want only our products, that's great. If you want us to deploy, plan, maybe even support on the ongoing basis, our products, we can provide these services to you. And by that, make the best solution for you."

The editorial staff had no role in this post's creation.