Thursday, 18 December 2025

Transport Networks Holding Modern Mobile Architectures Together

When people talk about mobile networks, the conversation almost always starts with the air interface. Spectrum, waveforms, MIMO, antennas and radios dominate conference agendas, white papers and training courses. After that comes the RAN, then the core, and occasionally backhaul is given a brief mention. What sits quietly in the middle of all this, often taken for granted, is the transport network. Yet without a well designed transport layer, even the most advanced radio and core architecture struggles to deliver on its promises.

Transport networks are the connective tissue of mobile communications. They carry traffic between radio sites, aggregation layers, edge and regional data centres, and the core network. As highlighted in the accompanying Mpirical video, transport is not a single homogeneous network but an end to end topology made up of multiple architectural domains, each with different performance, scale and resilience requirements.

At the core of the network, transport is typically built using highly resilient designs such as full mesh or spine and leaf architectures. These environments are already operating at hundreds of gigabits per second per link, with clear evolution paths towards terabit scale throughput. This part of the network rarely gets attention from mobile engineers, yet it underpins everything that follows. If the core transport layer cannot scale, the rest of the mobile network inevitably hits a ceiling.

Moving closer to the cell site, the transport network transitions into metro and aggregation domains. Here, spine and leaf or ring based topologies are commonly used, supporting large numbers of high capacity connections while also providing access to edge and regional data centres. This is where transport starts to intersect directly with mobile architecture decisions. The placement of edge computing platforms, local breakout, and centralised RAN functions all depend on the capabilities of this aggregation layer.

Closer still to the access network, transport designs often shift again. Ring, star or chain topologies are frequently used to connect clusters of cell sites, with capacities that reflect both traffic demand and economic constraints. Although fibre is the dominant medium, especially for 5G, it is rarely the only one. Microwave, integrated access and backhaul, and even non terrestrial technologies play an increasingly important role in extending coverage and improving resilience where fibre is impractical or unavailable.

The importance of transport becomes even clearer when viewed through the lens of disaggregated RAN and cloud based architectures. With gNodeB functions split into remote radio units, distributed units and centralised units, transport is no longer just backhaul. It becomes fronthaul and midhaul as well, each with distinct latency, synchronisation and bandwidth requirements. Centralised units may sit deep in the network, served by high capacity backhaul, while distributed units are connected via midhaul rings and radios are attached using star or ring topologies at the very edge.

This architectural shift exposes a common blind spot. Many performance issues blamed on the RAN are in fact rooted in transport limitations. Synchronisation accuracy, latency variation and resilience all depend heavily on transport design and operation. Packet based transport, while flexible and cost effective, places strict demands on timing and quality that cannot be treated as an afterthought.

As networks move towards 5G standalone, private networks and early 6G concepts, transport will become even more tightly coupled with service delivery. Network slicing, deterministic performance and edge driven applications all rely on a transport layer that can offer predictable behaviour rather than best effort connectivity. This pushes transport out of the shadows and into the critical path of mobile network design.

The 5G Transport Network Topology video as follows:

For mobile engineers, the message is clear. Understanding the air interface will always be essential, but it is no longer enough. Transport networks shape where functions can be placed, how services perform, and how networks scale over time. The video embedded alongside this post provides a useful visual reminder that mobile networks are not just radios and cores connected by invisible links. Transport is a network in its own right, and it deserves far more attention than it usually gets.

Related Posts

Tuesday, 25 November 2025

IET Lecture by Prof. Andy Sutton: Point to Point Microwave Radio Systems

Point to point microwave radio systems have been with us for more than eighty years, yet they rarely attract much attention in an era where fibre dominates network planning and satellite systems continue to develop at pace. At a recent IET Anglian Coastal Local Network event, Prof. Andy Sutton delivered an excellent lecture that brought these fixed radio links back into the spotlight. His talk explored the history, engineering and future of microwave and millimetre wave links, reminding us why they remain essential for transmission networks in the UK and around the world.

The story begins with the national microwave radio network of the 1970s, with the BT Tower at its centre. These early deployments supported long links across the country and laid the foundation for many of the design principles still used today. While the landscape has changed significantly, the fundamentals of fixed radio communication continue to be shaped by spectrum availability, propagation characteristics and careful engineering.

Microwave links depend on a wide range of bands, from the lower 6 GHz region through to 80 GHz E-band. The choice of frequency affects everything from link length to susceptibility to atmospheric absorption. As Andy explained, a link designer must consider not just free space path loss, but also Fresnel zone clearance, rainfall intensity and antenna characteristics. The slides included a worked example that showed the impact of frequency and distance on the radius of the Fresnel zone and highlighted the need for adequate clearance to maintain availability over time.

The talk moved on to modern access radio systems, where compact rooftop nodes and all-outdoor radios have become common. These systems rely on careful use of vertical and horizontal polarisations, often enabled through XPIC technology. XPIC allows separate data streams to coexist on the same frequency using orthogonal polarisations, effectively doubling link capacity when conditions allow. This is paired with adaptive coding and modulation, which enables the radio to shift modulation schemes according to link quality. The result is a more resilient and efficient link compared to older fixed-modulation systems.

Capacity planning is a balancing act that involves radio channel bandwidth, modulation choice and the number of aggregated carriers. Wider channels and higher order modulation support multi-gigabit throughput, although this introduces penalties in transmit power and receiver sensitivity. The trade-offs are central to radio design and determine the type of equipment used, whether through a separate indoor and outdoor unit or an integrated all-outdoor system.

Andy also covered the practical elements of radio link planning, such as antenna selection, path profiling, waveguide losses and typical link budget calculations. A link planning example using a 32 GHz radio demonstrated the relationship between transmit power, antenna gain, free space loss and fade margin for a target availability of 99.99 percent. The discussion tied together the theoretical foundations with real-world engineering and illustrated how access radios are designed for street-level backhaul scenarios.

The lecture then moved to millimetre wave systems, particularly E-band radios that operate around 70 and 80 GHz. These links offer enormous capacity over shorter distances and are increasingly used for dense urban backhaul and enterprise connectivity. The slides included examples of network topologies showing how microwave and fibre can be combined to meet different deployment objectives.

A substantial part of the presentation focused on trunk or core microwave radio systems. These high-capacity, high-availability links support long distances and historically formed the backbone of national networks. Although demand for trunk links has reduced as fibre has spread, they still exist in challenging environments. In the UK, many trunk links remain operational in Scotland and island regions where terrain and geography limit fibre deployment. The lecture covered branching networks, duplexers, waveguide installations and space diversity techniques, all of which contribute to the reliability of long-haul links.

Looking ahead, research continues into new frequency bands, wider channels, higher modulation schemes and improved radio hardware. These advances will support even greater capacities, with millimetre wave links expected to reach 100 Gbps over short distances. Microwave radio may no longer be the headline technology it once was, but the field continues to push boundaries and remains an essential part of modern communication networks.

Andy’s lecture was a comprehensive tour of the past, present and future of point to point microwave systems. For anyone working in transmission, mobile networks or wireless engineering, it served as a valuable reminder of the depth of innovation in this area and its continued relevance in the broader ecosystem.

If you would like to explore the material in more detail, the slides from the event are available here and the video can be seen here. Both are well worth a look.

Related Posts:

Tuesday, 4 November 2025

AIoT and A-IoT

Our industry loves acronyms. In fact, sometimes it feels as if half our job is simply keeping up with them, while the other half is explaining them to everyone else. A recent example I saw referenced D2D for satellites, but expanded it as Device to Device instead of Direct to Device. Today, two similar acronyms are gaining momentum and are likely to become far more mainstream: AIoT and A-IoT.

Artificial Intelligence (AI) and the Internet of Things (IoT) are two of the key technological pillars of the modern digital world. IoT connects billions of devices, from sensors and cameras to industrial machinery, all producing vast amounts of useful data. AI enables these devices and systems to learn from this data, recognise patterns, predict outcomes, and act autonomously.

When these technologies come together, we get the Artificial Intelligence of Things, or AIoT. In simple terms, AIoT allows connected devices to analyse the data they generate and make decisions without always relying on central systems.

The intelligence in AIoT can sit in different places. Cloud based AI offers extensive processing power and the ability to leverage wider datasets. Edge AI processes data closer to where it is generated, enabling faster and more context aware decision making while reducing bandwidth use and protecting data privacy. Increasingly, lightweight machine learning models allow intelligence directly on devices themselves, enabling instant reactions without constant network access. This evolution transforms IoT devices from passive data collectors into proactive decision makers.

The benefits are significant. AIoT increases automation, improves efficiency, enhances reliability, and enables predictive maintenance, energy optimisation, autonomous navigation, and smarter logistics. It also supports sustainability initiatives, for instance by improving energy and water use monitoring or enabling more intelligent control of municipal utilities. In short, AIoT forms a key part of the digital transformation strategies emerging across industries.

To get a better sense of how AIoT could shape our everyday lives, I have embedded a couple of older Ericsson videos below that imagine a future where intelligence is seamlessly built into everything.

For anyone interested in going deeper into this topic, Transforma Insights and Supermicro have good explainers. While 3GPP continues to work on AI, ML and IoT, AIoT as a concept is largely implementation driven rather than a standardised feature in itself.

In contrast, 3GPP is actively defining a different acronym: A-IoT, short for Ambient IoT.

Ambient IoT represents a major shift in connected device design. Instead of relying on batteries or frequent charging, Ambient IoT devices operate using energy harvested from their surroundings. This can include radio signals, light, heat, or motion. The technology supports both passive operation, where devices backscatter incoming RF signals, and active operation, where they harvest enough power to generate and transmit signals independently.

Unlike traditional IoT devices, Ambient IoT units are extremely low power, low cost, and very simple in design. They have a shorter range and lower data throughput than conventional wireless technologies, but they excel in scenarios where massive numbers of tiny, battery-free sensors can be deployed and left to operate with minimal maintenance.

This makes Ambient IoT well suited to applications such as environmental sensing, supply chain tracking, inventory monitoring, smart agriculture, and intelligent labelling. It also opens opportunities in consumer environments, from smart packaging to indoor positioning. With the right network support, these devices can operate indefinitely, enabling sustainable, large-scale sensing networks.

Ambient IoT is already included in 5G Advanced Release 19. For those interested in learning more, 3GPP has a detailed overview, Oppo has produced an excellent white paper, and LG Uplus has published a forward looking document exploring Ambient IoT in the context of 6G.

Both AIoT and Ambient IoT represent the next phase of connected intelligence. AIoT pushes computation and decision making closer to where data originates, while Ambient IoT removes power barriers and enables pervasive, maintenance-free connectivity. Together, they will support systems that are scalable, energy efficient and context aware.

As these technologies mature, we can expect a world where devices are not only always connected, but also constantly learning, adapting, and operating independently with minimal energy demands. The future of connectivity lies in this balance between intelligence and efficiency, and both AIoT and Ambient IoT will play a crucial role in shaping it.

Related Posts

Thursday, 16 October 2025

Evolving Communication Security Towards 6G at the ETSI Security Conference 2025

The annual ETSI Security Conference returned to the French Riviera from 6 to 9 October, once again bringing together the global cybersecurity community in the beautiful surroundings of ETSI headquarters. Over 250 participants from industry, government agencies, academia, global standards bodies, and open-source communities attended, making it one of the most engaging editions to date. The four-day event featured keynotes, panel discussions, technical sessions, poster presentations and live demonstrations, offering a holistic view of today’s security challenges and tomorrow’s opportunities.

The opening day provided a broad overview of the global cybersecurity landscape, setting the tone for the week ahead. Discussions highlighted emerging trends such as the growing influence of artificial intelligence and the rapid evolution of regulatory frameworks, including the European Commission’s Cyber Resilience Act. The sessions underscored the importance of collaboration between policymakers, researchers, and standards organisations. The afternoon focused on the cyber skills gap, a recurring theme across many sectors, stressing the need for education and training to build a security-aware workforce capable of safeguarding future digital systems. Standards were identified as key enablers in bridging policy and implementation, helping to transform regulatory intent into operational resilience.

The second day examined the paradox between AI as both a risk and a defence mechanism in cybersecurity. Experts discussed how AI-driven systems can expose new vulnerabilities if developed without strong security foundations, while also offering powerful tools for detection and response. Another session addressed fraud reduction and the convergence of security strategies to protect both networks and end users. A major highlight was the discussion on the global uptake of ETSI’s consumer IoT security standard, ETSI EN 303 645. Representatives from Germany, the UK, Singapore and Japan shared national experiences implementing consumer labelling schemes based on this standard, confirming its status as a globally recognised baseline for IoT security.

The third day was dedicated to the evolution of communication technologies and the emerging security landscape as the world moves towards 6G. Chaired by Dario Sabella from xFlow Research, the morning session explored how the journey from 5G Advanced to 6G requires a fresh approach to network security. The day began with an update from Alain Sultan of ETSI on the ongoing work within 3GPP SA3, focusing on strengthening frameworks for new architectures and deployment models. Bengt Salin from Ericsson outlined what should be considered in shaping security for 6G, emphasising that the next generation must be secure by design, not by adaptation. Nauman Khan from STC analysed the threat landscape surrounding 5G MEC and private networks, noting that as edge computing becomes more widespread, it introduces new vulnerabilities but also provides insights that can guide 6G security frameworks. Leyi Zhang from ZTE then presented on Secure Space-Air-Ground Integrated Networks, a concept uniting terrestrial, aerial, and satellite systems to provide ubiquitous connectivity. Ensuring trust, authentication, and data protection across such a heterogeneous environment presents one of the greatest challenges for 6G.

A panel discussion moderated by Dario Sabella brought together the morning’s speakers to reflect on security priorities toward 6G. The consensus was clear: while 6G is still in the early stages of standardisation, security must not be an afterthought. Lessons from 5G—particularly regarding openness, complexity, and trust—must inform the architecture and design principles of 6G from the outset. The afternoon sessions continued with broader discussions about digital sovereignty, fragmentation, and whether the internet is moving toward a “splinternet”. The day concluded with a deep dive into post-quantum cryptography, where real-world implementations provided valuable lessons for securing the next era of communication systems.

The final day of the conference shifted attention to geopolitics, cyber resilience, and the role of standards in shaping strategic responses to global challenges. Speakers explored how critical infrastructure security is increasingly influenced by geopolitical dynamics and how coordinated international standards can help mitigate risks. The Cyber Resilience Act remained a focal point, with experts emphasising the urgency of developing the 19 associated ETSI standards to support implementation. Harmonising global labelling schemes based on ETSI EN 303 645 was identified as an immediate priority, while in the longer term, education—both for future generations and C-level executives—was seen as essential to strengthen awareness of how standards underpin sovereignty, innovation, and competitiveness.

The 2025 edition of the ETSI Security Conference reaffirmed ETSI’s position as a central hub for cybersecurity dialogue and collaboration. From 5G and IoT to post-quantum cryptography and 6G, it showcased how security is now integral to every layer of the digital ecosystem. As the journey toward IMT-2030 continues, the message from Sophia Antipolis was clear: proactive, standards-based collaboration is the foundation of a secure connected future.

You can see the detailed agenda here. The presentations from the conference are all available here.

Related Posts

Thursday, 9 October 2025

Seamless UE Context Recovery (SUECR) in 3GPP Release 18

3GPP Release 18 introduces a wide range of enhancements across the 5G system, from energy efficiency and XR optimisation to AI-powered features. Among these developments is a practical but important addition known as Seamless User Equipment Context Recovery (SUECR), designed to handle situations where a device temporarily goes offline.

When a device such as a smartphone or IoT unit undergoes an operating system upgrade, a modem reset or a software update, it may become unavailable for a period of time. If this happens without informing the network, the operator’s core functions and connected application servers may continue to treat the device as available. This can result in wasted signalling, unnecessary retries and disruptions to critical operations that depend on the device’s availability.

SUECR provides a solution by allowing the device to notify the network of an unavailability period, which is a defined window of time during which it cannot communicate. Both the device and the core network retain important session and mobility information so that once the device returns, service can continue smoothly without unnecessary procedures.

The feature works in two ways depending on the device’s ability to store context information. If the device can preserve its mobility and session management contexts in non-volatile memory or on the SIM, it executes a registration procedure before going offline. The unavailability period is included in this request, and the Access and Mobility Management Function (AMF) records the duration and recognises the device as unreachable until it re-registers. If an application function has subscribed to receive updates on device availability, the AMF also forwards this information so that application servers can adapt accordingly. If the device cannot save its context, it instead executes a deregistration procedure to notify the AMF of its unavailability, with similar treatment by the network until the device performs its next registration.

Once the update or reset is complete, the device re-registers with the network and resumes normal service. If the planned downtime is delayed, cancelled or extended, the device repeats the procedure to keep the network and applications accurately informed. This ensures that network functions and application servers no longer waste resources attempting to reach devices that are temporarily offline.

By introducing SUECR, Release 18 strengthens service reliability and efficiency. It prevents unnecessary signalling and enables critical applications to maintain accurate awareness of device availability. The figure above, from NTT Docomo’s Technical Journal, illustrates how the unavailability period is managed depending on whether the registration or deregistration procedure is used.

Seamless User Equipment Context Recovery may appear as a small enhancement in the context of all the new Rel-18 features, but it addresses an important gap in 5G operations. As networks continue to evolve towards automation and support for mission-critical services, this function will play a key role in making device management more predictable and dependable.

Related Posts

Tuesday, 23 September 2025

5G+ and 5GA Icon (Pictogram) in New Smartphones

As 5G matures, new icons are appearing on smartphones to distinguish faster or more advanced connections. Some of the latest 5G smartphones around the world have started showing new icons such as 5G+ and 5GA. Interestingly, in Japan these are referred to as pictograms.

A long time ago, we looked at how Swisscom described its 5G rollout as 5G-wide and 5G-fast. Today, Swisscom uses the 5G+ icon to represent what it previously called 5G-fast. In its annual report, Swisscom explained:

5G (and 5G+) is the latest generation of mobile technology. Compared to 3G and 4G, it provides even more capacity, very short response times, and higher bandwidths. 5G technology plays a major role in supporting the digitalisation of the Swiss economy and industry. Swisscom differentiates between 5G-fast (narrower coverage up to 2 Gbit/s and more) and 5G-wide (Switzerland-wide 5G coverage with up to 1 Gbit/s). 5G-fast is also known as 5G+. Both variants are more efficient than their predecessor technologies with respect to energy consumption and use of electromagnetic fields.

Japan has only recently transitioned to using 5G+. A Google-translated page from NTT Docomo explains it as follows:

In areas where 5G communication is possible, the RAT display on standby will be "5G." On the other hand, during communication, the RAT display will be "5G+" for 5G communication using wideband 5G frequencies (3.7 GHz, 4.5 GHz, 28 GHz), "5G" for 5G communication using 4G frequencies, and "4G+" for LTE communication.

There are also footnotes clarifying that the display depends on the device, the bands supported, and the area of use.

From this, my understanding is that in newer devices the 5G+ icon is primarily used to indicate speed and capability, regardless of whether the connection is Standalone (SA) or Non-Standalone (NSA) 5G. KDDI is following the same approach, as explained on its own support pages.

Last year we looked at what iPhone icons meant. In iOS 18, 5G+ indicated that the phone was connected to mmWave. In iOS 19 this hasn’t really changed, although I have been told that it depends on the operator whether they choose to display 5G+ when the device is camped on higher-speed mid-band 5G.

Samsung Galaxy smartphones display two or three types of icons, as shown in the picture at the top. While the meanings are not entirely clear, Samsung’s user guide for Android 15 explains them as:

  • Filled square: “5G network connected”, which I interpret as being connected to a 5G Standalone network.
  • Transparent or outlined square: “LTE network connected in LTE network that includes the 5G network.”, which I interpret as 5G NSA.  
  • I did not find a reference to the unboxed 5G icon in this manual.

Finally, the OnePlus 13 in India has started displaying the 5GA icon. Since Jio only operates a 5G Standalone network, it is possible they have upgraded the network and device to use the Release 18 ASN with some new features. This allows them to market it as 5G-Advanced, thereby justifying the 5GA icon.

If you have noticed something different in your country or region, or have another interpretation, I would love to hear more.

Related Posts:

Thursday, 11 September 2025

Dummy Loads in RF Testing for Dummies

I have spent many years working in the Test and Measurement industry and have also worked as a hands on engineer testing solutions, and as a field engineer testing various solution pre and post deployment. Over the years I have used various attenuators and dummy loads. It was nice to finally look at the different types of dummy loads and understand how they work in this R&S video.

So what exactly is a dummy load? At its core, it is a special kind of termination designed to absorb radio frequency energy safely. Instead of letting signals radiate into the air, a dummy load converts the RF power into heat. Think of it as an antenna that never actually transmits anything. This makes it invaluable when testing transmitters because you can run them at full power without interfering with anyone else’s spectrum.

Ordinary terminations are widely used in test setups but they are usually only good for low power. If you need to deal with more than about a watt of power, that is where dummy loads come in. Depending on their design, they can handle anything from a few watts to many kilowatts. To survive this, dummy loads use cooling methods. The most common are dry loads with large heatsinks that shed heat into the air. For higher powers, wet loads use liquids such as water or oil to absorb and move heat away more efficiently. Some combine both air and liquid cooling to push the limits even further.

Good dummy loads are not just about heat management. They also need to provide a stable impedance match, usually 50 ohms, across a wide frequency range. This minimises reflections and ensures accurate testing. Many dummy loads cover frequencies up to several gigahertz with low standing wave ratios. Ultra broadband designs, such as the Rohde & Schwarz UBL100, go up to 18 GHz and can safely absorb power levels in the kilowatt range

Some dummy loads even add extra features. A sampling port allows you to monitor the input signal at a reduced level. Interlock protection can shut down a connected transmitter if the load gets too hot. These touches make dummy loads more versatile and safer in real-world use.

In day-to-day testing, dummy loads help not only to protect transmitters but also to get accurate measurements. By acting as a perfectly matched, non-radiating antenna, they give engineers confidence that they are measuring the true transmitter output. They can also be used to quickly check feedlines and connectors by substituting them in place of an antenna.

Rohde & Schwarz have put together a useful explainer video that covers all of this in a simple, visual way. You can watch it below to get a clear overview of dummy loads and why they matter so much in RF testing.

Related Posts: