Electronics Guide

Articles in This Category

Spectrum and Propagation

The electromagnetic spectrum is the scarce resource that shapes every wireless generation, and access to new bands has driven each step forward. The 5G New Radio standard organizes spectrum into Frequency Range 1 (FR1), spanning roughly 410 megahertz to 7.125 gigahertz, and Frequency Range 2 (FR2), the millimeter wave bands. FR2 originally covered 24.25 to 52.6 gigahertz; Release 17 added FR2-2 to extend the upper edge to 71 gigahertz. Research toward 6G targets two distinct regions: the upper mid-band between roughly 7 and 15 gigahertz, which balances bandwidth against coverage, and the sub-terahertz region from 100 to 300 gigahertz for short-range, extremely wideband links. Frequencies above 300 gigahertz are conventionally described as terahertz.

Spectrum identification proceeds through the ITU World Radiocommunication Conference. WRC-23 identified additional mid-band spectrum for IMT in several regions, and WRC-27 agenda item 1.7 studies the 4400 to 4800 megahertz, 7125 to 8400 megahertz, and 14.8 to 15.35 gigahertz ranges as candidates. These bands are heavily used by incumbent satellite, fixed-link, radar, and scientific services, so the studies concentrate on sharing and compatibility rather than on clearing spectrum outright.

Physics sets the terms of the trade. Free-space path loss rises with the square of frequency, so a link at 28 gigahertz starts about 18 decibels behind an otherwise identical link at 3.5 gigahertz. Atmospheric effects add further penalties in specific bands: molecular oxygen absorbs strongly near 60 gigahertz, and water vapor produces prominent absorption lines in the sub-terahertz range. Building materials, foliage, and even the human body attenuate millimeter wave signals severely, and blockage by a hand or a passing vehicle can interrupt a link within milliseconds. Practical high-band systems therefore depend on dense deployment, high-gain steerable antennas, rapid beam recovery, and fallback to lower bands when the high-band link fails.

Air Interface and Radio Technologies

5G New Radio uses cyclic-prefix OFDM in both directions, with DFT-spread OFDM available on the uplink to reduce peak-to-average power ratio and extend coverage. Unlike LTE's fixed 15-kilohertz subcarrier spacing, New Radio defines a scalable numerology in which spacing doubles through 15, 30, 60, and 120 kilohertz, with wider spacings defined for FR2-2. Each doubling halves the slot duration, so higher numerologies trade frequency-domain granularity for shorter transmission intervals and lower latency. Channel bandwidths reach 100 megahertz per carrier in FR1 and 400 megahertz in FR2, and carrier aggregation combines several carriers into a much wider effective channel. Data channels use low-density parity-check codes while control channels use polar codes, and downlink modulation extends to 256-QAM, with 1024-QAM added for favorable FR1 conditions.

Massive MIMO is the defining antenna technology of the era. Arrays with tens to hundreds of elements serve multiple users simultaneously through spatial multiplexing while concentrating energy on each user through beamforming. Mid-band macro deployments commonly use 64-transmit, 64-receive arrays. At millimeter wave frequencies the short wavelength lets a panel a few centimeters across host well over a hundred elements, producing the array gain that makes those bands usable at all. Fully digital beamforming, with one radio chain per element, is impractical at these element counts and bandwidths, so mmWave systems typically adopt hybrid architectures that combine a modest number of digital chains with analog phase shifters. Beam management then becomes a first-class protocol function: the network sweeps synchronization and reference signals across candidate beams, the device reports the strongest, and both ends track and switch beams as conditions change.

Coverage and device diversity receive dedicated treatment as well. Release 17 introduced non-terrestrial network support, allowing New Radio and narrowband IoT to operate over satellites and high-altitude platforms with timing and Doppler compensation suited to fast-moving orbital links. The same release defined reduced capability devices, often called RedCap, which trim bandwidth, antenna count, and processing to serve wearables, sensors, and industrial equipment that need more than narrowband IoT but far less than a smartphone.

Network Architecture

The 5G core replaced the point-to-point interfaces of earlier generations with a service-based architecture in which network functions expose HTTP-based APIs to one another. Control and user planes are separated, so the user plane function can be placed close to the radio while control functions remain centralized. Because the functions are cloud-native software rather than dedicated appliances, operators can scale them independently and deploy new features without replacing hardware.

Network slicing builds on that foundation. A slice is a logically isolated end-to-end network, identified by a slice selection assistance identifier, with its own performance and isolation guarantees. One slice may serve consumer broadband, another a factory's control traffic, and a third a utility's metering fleet, each drawing on shared physical infrastructure while remaining insulated from the others' load. Multi-access edge computing complements slicing by hosting applications at aggregation sites or cell sites, cutting the round trip that would otherwise reach a distant data center.

The radio access network has disaggregated in parallel. The O-RAN Alliance specifications split the base station into a radio unit, a distributed unit, and a centralized unit, connected by an open fronthaul interface based on a functional split that places the lower physical layer in the radio unit. Two RAN intelligent controllers supervise the result: a near-real-time controller running xApps on control loops between roughly ten milliseconds and one second, and a non-real-time controller running rApps on slower loops for policy, configuration, and model training. The intent is a multi-vendor supply chain and programmable optimization; the practical challenges are integration effort, fronthaul transport capacity, and demonstrating that a disaggregated system matches the energy efficiency and performance of an integrated one.

Service Categories and Applications

The ITU-R framework for IMT-2020 organized 5G around three usage scenarios, and the minimum technical requirements attached numbers to them. Enhanced mobile broadband targets a 20-gigabit-per-second peak downlink rate and a 100-megabit-per-second user-experienced rate, supporting high-resolution video, cloud applications, and fixed wireless access. Ultra-reliable low-latency communications target one-millisecond user-plane latency at very high reliability, enabling motion control, remote operation, and safety-related automation. Massive machine-type communications target a connection density of one million devices per square kilometer, prioritizing long battery life and deep coverage over throughput. These are ceiling values for the technology under defined conditions, not the rates a typical subscriber sees on a loaded commercial network.

IMT-2030 expands the picture to six usage scenarios. Three evolve from 5G as immersive communication, hyper-reliable and low-latency communication, and massive communication. Three are new: ubiquitous connectivity, which folds satellite and other non-terrestrial access into the mainstream service model; artificial intelligence and communication, which treats distributed learning and inference as network services rather than applications riding on top; and integrated sensing and communication, in which the same waveform that carries data also detects, ranges, and images objects in the environment. Four overarching aspects apply across all six: sustainability, security and resilience, connecting the unconnected, and ubiquitous intelligence.

Concrete deployments already track these categories. Fixed wireless access has become one of the most commercially significant 5G applications, using mid-band and millimeter wave capacity to substitute for wired broadband. Private networks serve ports, mines, factories, and campuses where controlled latency and on-premises data handling matter more than peak speed. Extended reality headsets, cloud gaming, and remote-operated equipment exercise the combination of bandwidth and latency together. Longer-term proposals such as holographic communication, digital-twin synchronization, and network-as-a-sensor remain research subjects whose feasibility depends on advances still under study.

Challenges and Research Directions

Energy efficiency is the most persistent constraint. The radio access network dominates mobile network power consumption, and every lever that raises capacity, including wider bandwidths, larger arrays, and denser cells, tends to raise energy use as well. Work on network energy saving in recent 3GPP releases addresses cell and carrier sleep modes, adaptive antenna muting, and signaling that lets equipment idle without stranding devices. Sustainability is an explicit IMT-2030 design goal rather than an afterthought, which marks a change from earlier generations.

Security and trust grow harder as the attack surface expands. Billions of low-cost connected devices, disaggregated multi-vendor infrastructure, virtualized functions sharing physical hosts, and supply chains spanning many jurisdictions all widen exposure. 5G improved on its predecessors by concealing subscriber identifiers over the air and by adding a security edge protection proxy for interconnect, but open interfaces and cloud deployment introduce new categories of risk that require continuous assurance rather than one-time certification. Migration to post-quantum cryptography is now part of that planning.

Several research directions could reshape the next decade. Reconfigurable intelligent surfaces would turn walls and facades into passive or nearly passive arrays that steer reflections, extending coverage into shadowed areas without a full relay. Integrated sensing and communication would let networks perceive their surroundings using the signals they already transmit, though spectrum sharing, privacy, and waveform design questions remain open. AI-native air interfaces, in which learned components replace hand-designed receivers or channel state feedback, promise gains but raise hard problems of interoperability and predictable behavior between equipment from different vendors. Terrestrial and non-terrestrial integration demands new mobility, timing, and interference-management techniques as low-earth-orbit constellations expand.

Taken together, these efforts describe a field where fundamental advances remain possible and where the hardest constraints are increasingly practical rather than theoretical: cost per bit, watts per site, spectrum availability, and the discipline of integrating many vendors' equipment into one dependable network. The subcategories above examine each of these frontiers in greater depth.