The growing interest in terahertz (THz) circuits and devices for next-generation communication and sensing technologies, combined with the increasing need for novel nonplanar designs, necessitates accurate, and broadband characterization of advanced dielectric substrates. This study comprehensively characterizes industrial-grade Radix (TM) dielectric resins alongside conventional substrates from 50 GHz to 2.5 THz, with measurements conducted at both the National Physical Laboratory and University of Birmingham using quasi-optical systems run by vector network analyzers (VNAs) and time-domain spectrometers (TDSs). Five different collimated beam measurement setups are used, each using in-house developed material extraction algorithms that incorporate the influence of surface roughness into the extraction process, extending proprietary VNA and TDS-based extraction algorithms. The comprehensive interlaboratory measurement campaign demonstrates strong consistency, validating the robustness and reliability of the methodologies. Detailed analysis confirms that the additional surface roughness introduced by 3-D additive manufacturing of Radix (TM) resins has a negligible impact on electrical parameters.

This study examines the influence of surface-roughness-induced scattering losses on the dielectric parameter extraction process in terahertz time-domain spectroscopy. Our material extraction methodology, based on a ray optic transfer function, is tested on synthetic, natural wood and engineered wood samples. The results indicate that the proposed ray-optics–based scattering model remains valid for surfaces with low to moderate roughness, where the ratio of RMS roughness to wavelength is below approximately 0.09. The study further presents an uncertainty analysis addressing the effects of average sample thickness and surface roughness on the extracted dielectric parameters. Supported by detailed simulations, the results reveal that up to 10%uncertainty in both surface roughness and average sample thickness leads to variations of up to 21.94% in the loss tangent, whereas variations in the real part of the complex permittivity reach up to 8.71%.

Ray-tracing engines have been a cornerstone of deterministic radio propagation prediction since the 1990s and continue to be widely employed, particularly due to advancements in computational technologies such as high-performance computing, GPUs, and artificial intelligence. These engines are extensively applied across various domains within the electromagnetic community, including wireless propagation, optical antenna design, and radar cross-section estimation. Their computational efficiency in electrically large environments positions them as a favourable alternative to both extensive measurement campaigns and other deterministic simulation methods, such as FDTD, MoM, and FEM. Nevertheless, achieving the dual objectives of high accuracy and low computational load remains a significant challenge, particularly in scenarios involving complex electromagnetic phenomena, including diffraction, transmission, and scattering (Schweins et al., IEEE Open J. Antennas Propag., 5(2), 2024, Azpilicueta et al., IEEE Trans. Antennas Propag., 7(10), 2023, Seretis et al., IEEE Trans. Antennas Propag., 70(6), 2022.).

The lack of an accessible comprehensive reliable dataset of material properties at terahertz (THz) frequencies is a significant roadblock for the development of accurate propagation models for THz wireless systems. To fill this gap, in this article, we extract the complex permittivity of 75 different home and commercial furnishings categorized into fabric, leather, plastic, stone, and wood from quasi-optical measurements. Given that quasi-optical transmission measurement standards are not developed yet, we utilize both vector network analysis and time-domain spectroscopy under collimated and focused beam illumination to highlight the differences and the challenges that precise material extraction faces at THz, and to compute average complex permittivities with confidence intervals.

The sensitivity of the complex permittivity to thickness uncertainty of the materials under test (MUT) in a vector network analyzer-based (VNA) quasi-optical measurement system is investigated through theoretical and experimental analyses. The study covers five material groups in the frequency range 220-750 GHz. The results highlight that material groups such as fabrics and leathers, which are more susceptible to thickness measurement uncertainty than other materials, exhibit variations of up to (10 +/- 0.3)% and (10-255.8)% in the real, epsilon(r)', and imaginary part, epsilon(r)'', of the complex permittivity. Lastly, as in the plastic sample measurement, extreme deviations are observed, particularly in epsilon(r)'', when the loss tangent is near-zero value.

One remaining challenge in modern metasurface-based surface wave technologies is efficient in- and out-coupling of energy. This letter demonstrates launching efficiencies up to 82% by varying the position of the launcher relative to the unit cell and provides an equivalent circuit model to guide designers. This is an additional degree of freedom that has not been discussed previously and arises due to the discontinuity introduced by the launcher and its effect on the local surface impedance.

This article presents a novel reconfigurable feed network that allows the switch of antenna arrays across four different polarization states. Different from most previous multipolarization antennas, the reconfiguration is implemented solely in the feed network without disturbing the radiation element. The key enabler is a reconfigurable coupler that switches between two circuit functions—a branch-line coupler or a crossover—through altering the line widths of its parallel branches. This is only made possible by using liquid metals (LMs). A double-pole-double-throw (DPDT) switch, also enabled by LM, is employed to select the input port of the reconfigurable coupler. The switch and the reconfigurable coupler form the reconfigurable feed network, providing four states of excitation to the antenna. The reconfigurable coupler, the switch, and a 2 $\times$ 2 array have been designed and experimentally verified with the LM actuation. As a branch-line coupler, the reconfigurable coupler shows excellent amplitude and phase balance within 0.04 dB ( $\vert\vert S_{{21}}\vert$ – $\vert S_{{31}}\vert\vert$ $<$ 0.04 dB) and 90 $^{\circ}$ at 3.5 GHz, whereas, as the crossover, it has an isolation over 25 dB. The switch exhibits an isolation of 50 dB. The measured peak gain of the array is 14.7 dBic. At 3.5 GHz, the measured axial ratio (AR) associated with the right-hand circular polarization (CP) state is 1.0 dB. The speed of switching and power handling associated with the LM have been discussed. The use of the LM has enabled a unique way of circuit reconfiguration and a full polarization-reconfigurable multifunctional antenna.

A ready-to-use numerical model has been developed for the atomic ladder (cascade) systems which are widely exploited in Rydberg Radio Frequency (RF) sensors. The model has been explicitly designed for user convenience and to be extensible to arbitrary N-level non-thermal systems. The versatility and adaptability of the model is validated up to 4-level atomic systems by direct comparison with experimental results from the prior art. The numerical model provides a good approximation to the experimental results and provides experimentalists with a convenient ready-to-use model to optimise the operation of an N-level Rydberg RF sensor. Current sensors exploit the 4-level atomic systems based on alkali metal atoms which require visible frequency lasers and these can be expensive and also suffer from high attenuation within optical fiber. The ability to quickly and simply explore more complex N-level systems offers the potential to use cheaper and lower-loss near-infrared lasers.

Surpassing 100-Gb/s data throughput is a key objective and an active area of research for sixth-generation (6G) wireless networks that can only be met by exploiting the terahertz (THz) frequency band (0.3–10 THz). THz channel modeling faces new challenges given the emerging relevance of scattering and molecular absorption in this frequency range as well as the lack of a reliable library of material properties. In this work, we address these challenges by measuring systematically the dielectric properties of 27 common building and office materials and reporting an in-house 3-D ray-launching (3D-RL) algorithm that uses the created material library and accounts for rough surface scattering and atmospheric attenuation. In order to validate the proposed algorithm, a channel sounder measurement campaign has been performed in a typical indoor environment at 300 GHz. Simulations and measurements show good agreement, demonstrating the need for modeling scattering and atmospheric absorption in the THz band. The proposed channel model approach enables scenarios at THz frequencies to be investigated by simulation, providing relevant knowledge for the development of ultrahigh-speed wireless communication systems.

We present a passive RF to optical data transfer without a local oscillator using an atomic “Rydberg” receiver. We demonstrate the ability to detect a 5G frequency carrier wave (3.5 GHz) and decode digital data from the carrier wave without the use of a local oscillator to detect the modulation of the RF signal. The encoding and decoding of the data are achieved using an intermediate frequency (IF). The rubidium vapor detects the changes in the carrier wave's amplitude, which comes from the mixing of the IF onto the carrier. The rubidium vapor then upconverts the IF into the optical domain for detection. Using this technique for data encoding and extraction, we achieve data rates up to 238 kbps with a variety of encoding schemes.

Indoor wireless communications need to move towards Terahertz (THz) frequencies in order to keep up with society's demand for data transmission, but this change is currently hindered by limited knowledge of material properties and propagation and scattering models at these frequencies. The dielectric properties of common household materials are investigated here with a twofold objective: (1) to extend the library of material properties at THz, and (2) to estimate and disentangle losses in scattering measurements in order to facilitate propagation, scattering and, ultimately, channel models.

The identification and quantification of scattering phenomena is essential for designing indoor wireless communications. From a combination of time domain spectroscopy, analytical modelling and ray tracing simulations, a novel scattering factor is proposed for terahertz frequency bands. These results aim to assist in incorporating scattering effects in ray tracing simulations of indoor environments.

As a typical railway Cyber-physical System (CPS), radio-based train control systems have been playing an increasingly important role in rail transit. The engaged network, Global System for Mobile Communications for Railway (GSM-R), which is an out-dated wireless communication technology, will be decommissioned due to diminishing support from industry, and a new generation successor, e.g. Long-Term Evolution (LTE), is urgently required to replace the current network. The radio-based train control systems must be safety critical, which relies on a high-security Data Communication System (DCS). In this paper, a novel wireless network migration methodology in DCS is proposed. By using this methodology, the high-security required DCS performance in radio-based train control systems is maintained and the network migration cost, e.g. the used number of base station (BS), is reduced when updating the GSM-R to LTE.

Now more than ever, monitoring railway track geometry from in-service vehicles is an attractive proposition that ensures improved infrastructure performance without interrupting railway operations. Communicating the collected sensors-data to a central server has always been an issue due to the current GSM-R and LTE data-rate and spectrum limitations. The prospect of opportunistic access to an inefficiently utilised frequency spectrum, known as TV White Spaces (TVWS), is proposed to solve the spectrum scarcity problem that exploits desirable railway propagation characteristics. In order to provide full protection for the spectrum primary users, IEEE 802.22 standard sets strict policies on the mobile platforms. This research proposes a novel handover scheme that utilises a greedy algorithm to select the operational frequency channels. The scheme takes into account; the train’s trajectory, including the possibility of train delays, and coexistence issues between the spectrum’s secondary users. A case study of two trains reporting their collected maintenance data to 3 Access Points (APs) while travelling between Selly Oak and New Street Station, Birmingham, UK is presented. For high channels availability (≥40%), an average of 30 megabytes of extra track data can be transmitted using the new approach for an 8 min journey. In addition, a single channel can be used in the new approach for an average consecutive distance of 1.05 km compared with an average of 0.58 km for IEEE 802.22 standard. Both systems provide identical interference performance with more transmission power that can reach up to 42.2 dBm allowed under the new scheme.

The wideband frequency tunability of a two-port microstrip-fed patch antenna is achieved using injection matching. It is demonstrated that controlling the relative amplitude and phase shift between the excitation signals at port one and port two of the microstrip-fed patch antenna can tune its operating band to a lower wideband frequency range compared to the fundamental intrinsic resonance frequency of the corresponding one-port antenna. The resulting two-port antenna has an overall simulated efficiency in excess of 80% with fairly stable radiation pattern in the E-plane. The antenna is suitable for use in wireless communication applications in the C-band.

Wave attenuation through rain with different rainfall rates at millimeter wave ( $f = 77$ GHz) and low-terahertz (Low-THz) ( $f = 300$ GHz) frequencies is studied in this article. Rain has pronounced impacts on electromagnetic wave propagation and one of the well-known effects is attenuation of the transmitted wave. Attenuation at both frequencies and hydrometeor properties [rainfall rate and drop size distribution (DSD)] are measured simultaneously. The measured DSD is fit with gamma and Weibull distributions and is also compared to the frequently used distribution Marshall and Palmer (MP) model; Weibull is shown to be a better fit to the measured DSDs. Theoretical prediction of attenuation as a function of rainfall rate (up to about 20 mm/h) is determined using Mie scattering theory, and the fit gamma and Weibull, and MP distribution models; as well as using the International Telecommunications Union Radiocommunication Sector (ITU-R) recommendation. The calculations are evaluated by comparing them to the experiment. The measured results at 77 GHz best agree with the ITU-R recommendation whereas at 300 GHz, the calculation based on Mie scattering and the Weibull distribution exhibits the best fit to the measured data. The measured data that exceed the theoretical prediction are analyzed and interpreted based on their corresponding observed drop size properties, for the first time.

Communication-based train control (CBTC) systems have been playing a progressively significant role in metro signaling in recent years. As safety-critical systems, CBTC systems have very strict requirements on the wireless communication performance between train and wayside access points (AP), which is highly dependent on the deployment of the APs. In this paper, by customizing and adopting a decomposition-based multiobjective evolutionary algorithm, the proposed AP deployment optimization method has been implemented, verified, and its implementation accuracy has been assessed. A real-world case study is carried out in an integrated simulation platform, in which the optimized AP deployments are verified and show better performance than the original planning.

Experimental measurement results of automotive radar signal attenuation during various intensities of snowfall at the current automotive radar frequency (77 GHz) and low-terahertz (THz) (100-300 GHz) frequencies are presented and compared in this study. The attenuation is characterised by measuring the ratio of the received power from a reference target through snow precipitation of various intensities and through the same path with no precipitation. Statistical analysis of the attenuation is presented. Higher attenuation is measured at a higher frequency, and also attenuation increases as snowfall rate and liquid water content in snowflakes increases. This study is fundamentally important to investigate the effect of adverse weather conditions on low-THz radar performance in comparison with the current automotive radars operating in the traditional mm-wave band. In addition, the effects of low-THz wave attenuation and scattering due to typical contaminants formed on the radome of automotive radars and reflection from common objects on the road are presented.

In this paper, a wideband injection matched patch antenna is proposed for the wireless applications in the S and lower C bands using the injection matching theory. In this arrangement, the relative amplitude and phase shift between the excitation current signals at the two ports of a two-port antenna are used to match the antenna to a 50 Ohm source. A marginal reduction in efficiency from the corresponding one-port antenna is sacrificed to achieve a bandwidth ratio of 1.1:1 with a stable radiation pattern over the large operational bandwidth. The lower operational mode of the corresponding one-port antenna is also shifted to a lower frequency and the higher mode to a higher frequency by use of the injection matching. The realized antenna is compact in terms of thickness and is easy to fabricate.

We address the problem of admission control for wireless clients in WLANs taking into account collisions between competing access points and considering explicitly the effect of hidden terminals, which play a prominent role in optimised client association. We propose an efficient, distributed admission control algorithm, where the wireless client node decides locally on which access point it will associate with in order to maximise its link throughput. The client can choose to optimise either its uplink or downlink throughput, depending on the type of traffic it predominantly intends to exchange with the network. The proposed approach takes into account the full contention resolution of the RTS/CTS IEEE802.11 medium access control protocol and leads towards an increase of the total throughput for the whole network. Finally, an algorithm is proposed, which can serve also as the basis for the development of efficient traffic offloading protocols in heterogeneous 5G networks.