Vladimir Misljencevic

An important piece on AI on the battlefield came out today in The Times.

A crucial reminder that we need to accelerate autonomy — in deployment, procurement, grants, and more — so we don’t end up in the same situation as we did with fiber optics. ==
Last month, it saved the lives of an infantry unit that found itself in the path of 14 Russian armoured personnel carriers. Having tried and failed to destroy the column using conventional drones, it was the AI-powered ones that were able to penetrate their radio-electronic shield.  “We don’t have a lot of these [terminal guidance] drones, so when we do use them we use them as a last resort, when we really need them — and this was one of those instances,” said Andriy, 32, a sergeant in the brigade. He estimates that only one in 20 of the drones at his disposal are AI-enabled, all of which were bought with donations rather than supplied by the general staff. As is often a theme in this war, the Russians have more, he says. ==
For this article, we provided commentary from The Fourth Law and connected the correspondent with the 58th Brigade, who are using our technology with exceptional effectiveness. https://lnkd.in/dYnYjBxJ

40

1 Comment

🚀 AI-RAN Meets the I/Q Frontier: Unlocking RAN Intelligence Through Raw Signal Data 📶 In the era of AI-native RAN, we often talk about automation, intent-based orchestration, and LLM-driven RICs — but what if the real goldmine for intelligence is buried deeper? 💡 Enter I/Q data: the In-phase and Quadrature components of modulated RF signals. It’s the rawest, richest representation of the wireless environment — capturing everything from signal distortions to propagation nuances. And thanks to AI/ML, we now have the tools to decode its potential in real time. 🔍 AI/ML-based I/Q signal analysis is emerging as a game-changer in AI-RAN architectures by enabling data-driven optimization at the physical layer: 1️⃣ Channel Estimation & Prediction LSTM-based models can learn from historical I/Q sequences to predict Channel State Information (CSI), enabling smarter beamforming and scheduling before issues arise. 2️⃣ Modulation Recognition CNNs and SVMs can identify modulation types in real time, helping to mitigate interference and dynamically adapt to changing network conditions. 3️⃣ Anomaly Detection & Fault Diagnosis Autoencoders can flag deviations in I/Q patterns — from hardware degradation to malicious signal injection — for proactive maintenance and increased network security. 4️⃣ Beam Management Optimization AI agents trained on I/Q patterns can adjust beam direction and power dynamically, improving coverage, energy efficiency, and user experience in dense deployments. 5️⃣ Interference Identification & Cancellation Reinforcement learning enables real-time interference mitigation by learning adaptive transmission policies directly from I/Q observations. 6️⃣ Device Localization & Mobility Prediction Using spatial-temporal I/Q data, ML models can localize users and anticipate mobility even under NLOS conditions — critical for applications like autonomous transport and emergency services. 7️⃣ Intelligent Resource Allocation Predictive scheduling based on real-time I/Q data allows the RAN to allocate spectrum and processing resources where they’re needed most — maximizing throughput and minimizing latency. 📡 Bottom line: AI-RAN isn’t just about orchestration at the top. It’s about intelligence at the wavefront. By combining real-time I/Q analytics with AI-native RAN principles, we unlock a new layer of observability and control — paving the way for self-optimizing, resilient, and intent-aware 5G and 6G networks. 🧠 It’s time to bring machine learning closer to the signal. Let the RAN see and learn from what it transmits and receives. AI-RAN Alliance https://ai-ran.org #AI #RAN #AIRAN #5G #AIforRAN #IQData #RANIntelligence #5GOptimization #Beamforming #InterferenceMitigation #AIinTelecom #MLforRAN #TelecomAI #EdgeAI #SignalProcessing #AIEngineering #6GReady

166

7 Comments

This announcement caught my eye - two US vendors Eridan and Patero Inc. have integrated #postquantum #cryptography (#PQC) directly into an #OpenRAN radio unit for secure #5G networks. It looks optimised for private / local networks, especially in security-focused verticals such as critical infrastructure, logistics and defence. (I'm started to hear a lot about military-grade 5G O-RAN implementations recently). Like many governments , the US administation is taking quantum cybersecurity threats seriously, with Executive Order 14144 specifying PQC in some contexts. However despite the claims, it's not *quite* the first combination of O-RAN and PQC. Two years ago in 2023, I moderated at a panel for one of the UK's Department for Science, Innovation and Technology-funded projects. Called Secure 5G and part of the FRANC (Future RAN) programme, it was a prototype which combined Arqit's quantum-safe symmetric key-agreement technology with a Lime Microsystems O-RAN optimised SDR. That was more of an "overlay" approach to existing hardware, rather than Eridan's baked-in "deep-PHY" innovation. See https://lnkd.in/exp5ZtTg for my write-up. The Eridan / Patero approach is market-ready, and claims high spectral efficiency (a function of the RU's compound-semiconductor design), low power consumption and 3-microsecond fronthaul latency despite the new key-management approach. It also aligns with the US NIST post-quantum standards. I'm not aware of the Lime/Arqit solution being commercialised yet. (I've also seen a reference to a Telefonica quantum-safe P5G solution demo'd at this year's MWC) Reading the Eridan specs, the RU (branded "Miracle") is initially supporting the n48 CBRS band, but the design can deal with frequency ranges between 600MHz to 4.2GHz. It's also not immediately clear how easily this fits into an overall end-to-end quantum safe architecture, but presumably there are some CU/DU partners involved in this as well. That said, there is also a potential risk that deep integration of the PQC capabilities into the chips makes it a hostage to any future changes in the encryption standards or certification requirements. I guess that's the trade-off for lower overall latency. Overall, it's good to see PQC being taken seriously for #privatewireless, especially announced publicly. It's less advanced at the moment in public 5G networks, which may be a constraint on commercialisation of slicing or other solutions in some markets / verticals going forward. At the moment, it's mostly envisaged as part of #6G, likely covered in 3GPP Release 20/21 standards, although maybe there are some behind-closed-doors implementations as well. As a side-note, I'm intending to keep a close eye on #QuantumNetworks in future, as part of my wider interest in #PostDigital technologies. #Private5G #Cybersecurity #Telecoms #PostDigital https://lnkd.in/en62HQRu

31

2 Comments

⚡ White Paper: Power Electronics HIL – Test Smarter, Scale Faster When you're building embedded controllers for inverters, motors, or converters—real-time validation is non-negotiable. Hardware-in-the-Loop (HIL) testing lets you simulate switching dynamics and fault conditions before risking real hardware. 🔍 What’s inside: > How to emulate high-frequency PWM and wide bandgap semiconductors > Why FPGAs are essential for fast-switching control > Real-world use cases for electric drives and power converters 🔗 Download the white paper: https://spr.ly/60484tM6k #MATLAB #MathWorks #HIL #Simulink #PowerElectronics #FPGA #EmbeddedSystems #ModelBasedDesign #RealTimeTesting #Speedgoat #ElectricDrives

46

1 Comment

Did you know you can now build and operate a 3GPP-compliant LTE network in the 902-928 MHz license-free spectrum—the same band used by technologies like WiSUN, LoRaWAN, and RF Mesh? After three years of innovation, Ubiik has achieved a groundbreaking milestone with freeRAN™ LTE Base Station and Pyxis LTE Router, now FCC-certified in the US, as well as certified in Canada and Australia. Why is this revolutionary? --Unmatched Cybersecurity: LTE's 3GPP-standard security provides far superior protection compared to traditional LPWA technologies. --Higher Throughput: Enables data-intensive IoT applications that go beyond the capabilities of existing LPWAN solutions. --Roaming Capabilities: Seamlessly integrates with public cellular networks to bridge coverage gaps, making it ideal for warehouses, factories, and remote sites. --License-Free Operation: Deploy cost-effective, private LTE networks without the need for expensive spectrum licenses. This breakthrough positions freeRAN™ to complete the puzzle, paving the way for LTE-M and LTE to become the dominant technologies for future LPWAN networks. Whether it's smart cities, utilities, or industrial IoT, freeRAN™ redefines connectivity with unparalleled performance and flexibility. 🚀 Visit us at CES 2025, Booth #10278 in the IoT Zone, and discover how this game-changing product line is transforming the IoT landscape. https://lnkd.in/gh8AVbnN #CES #LPWAN #LoRaWAN #pLTE #smartcity #tracker #ubiik #IOT

51

1 Comment

From Silicon to Software: Week 12 The Software BOM: When the Build System Becomes the Product In hardware, a Bill of Materials (BOM) is table stakes. You don’t ship a board without knowing every resistor, capacitor, inductor, and IC. Every part has a number. Every vendor is documented. In software? Too many teams still treat the BOM as optional, a “nice-to-have.” Something compliance can worry about later. Here’s the problem: code doesn’t just run. It’s built. Toolchains, compilers, libraries, dependencies, environment variables, build scripts, CI/CD pipelines, all of it is part of the product, whether you admit it or not. Lose track of that, and you’re not managing a product anymore. You’re managing a gamble. I once had to update a shipping product so old it could’ve qualified for a museum exhibit. The CI/CD system had been shut down to “save resources.” Years later, a small regulatory change required a simple tweak. Easy, right? Just change a constant and rebuild. Except no one could. The pipeline was gone. The scripts were gone. The compiler only ran on an operating system that now needed a virtual machine to boot. We dug up old docs, patched drivers, and found someone who still remembered how to install the tools, like summoning a retired wizard for one last spell. The code change took five minutes. The archaeology took three weeks. That’s not maintenance. That’s engineering by hope, not design. And this is why a Software Bill of Materials (SBOM) isn’t just about open-source compliance or checking a regulatory box. It’s about survival. Knowing what’s in your code is only half the battle, you also need to know how it was built, with what, and whether you can rebuild it when the next regulator, auditor, or customer comes knocking. The smartest teams I’ve seen treat the build system itself as part of the product: • Capture compilers, libraries, and dependencies in containers. • Version CI/CD scripts alongside the code. • Automate reproducible builds so no one has to remember ancient flags. • Archive toolchains and build artifacts as first-class deliverables. Because hardware fails loudly: you can smell a burnt resistor, you can see the magic smoke wafting in the air. Software fails silently. You don’t realize you’ve lost the ability to build until the moment you need it and by then, the cost is measured in weeks, not hours. An SBOM isn’t a file. It’s the difference between a product you can support for decades and one you hope no one ever asks you to touch again. If your product’s longevity depends on the software, then your dependencies and build system are part of the product. Treat them that way. How does your team handle this? Do you keep your build environments alive, or do they quietly fade away after launch day? #FromSiliconToSoftware #embedded #software #sbom #maintenance #cicd

22

4 Comments