Abstract The urban radio‐frequency (RF) noise generated by our cities continues to change with time. Although models exist to describe the RF noise as functions of frequency and urban land use types, very few models describe the spatial character or structure of the noise on the scales of city blocks (50–150 m). The goal of this work is to investigate the connection between urban morphology and the higher‐order spatial statistics of the noise field. To achieve this goal, a large measurement campaign was conducted in Boston, Massachusetts. Many spatial measurements allowed for calculation of spatial correlation functions of noise power in three different neighborhoods, which were used to quantify the spatial structure of the fields. A statistical point source model is then developed, with adjustable parameters relating to urban morphology. Good agreement between the model and the experimental correlation functions suggests the 25 MHz urban noise field is well described by a random network of fixed point sources, radiating with a 1/r power law behavior.

Abstract The Radio Frequency Sensor (RFS), a new radio frequency lightning detector, was launched into geosynchronous orbit in December 2021, and first collected data in January 2022. RFS is a specialized software‐defined radio receiver that detects, records, and reports impulsive broadband radio‐frequency (RF) signatures from lightning in the very high frequency (VHF; 30–300 MHz) range. Its vantage point from a Western hemisphere geosynchronous orbit provides unique opportunities to study evolution of RF lightning signatures over the durations of thunderstorms over the Americas and Pacific Ocean. Its overlapping view with the Geostationary Lightning Mappers (GOES‐16 & 17) enables additional comparisons between the sources of optical emissions and associated VHF emissions that were not possible with previous sensors. We find that RFS preferentially detects bright VHF signals called transionospheric pulse pairs (trans‐ionospheric pulse pairs (TIPPs)). It is estimated that more than 85% of the RFS‐detected lightning events are TIPPs. This paper presents initial results from the first year and a half of on‐orbit operation.

Abstract State‐of‐the‐art conformal array pattern synthesis techniques typically suffer from a significant computational burden. This article builds upon the auxiliary array pattern synthesis technique, allowing the technique to work on elliptical contours. This deterministic method is both computationally efficient and capable of handling complex design cases with additional design constraints. In order to highlight the versatility and applicability, the method has been applied to an elliptical arc array comprised of Rampart antennas and an elliptical ring array using dipole elements. The results have been validated through full‐wave electromagnetic simulations, as well as physical measurements. Both arrays presented show good agreement with the proposed synthesis technique. Highlighting the ability to satisfy additional design constraints, such as enforcing a minimal element spacing or having a controlled beamformer‐network complexity.

Abstract Large‐scale ionospheric gradients associated with the solar terminator can deflect high frequency (HF) radio waves to off‐great circle paths during the morning and evening, negatively impacting technologies reliant on HF radio wave propagation. For example, geolocation algorithms used by scientific and military over‐the‐horizon radars (OTHRs) generally assume on‐great circle propagation, and thus lateral deviations from the great‐circle path can lead to positioning errors. In this study, radio wave propagation is simulated via 3D numerical ray traces though an empirical, high‐latitude model ionosphere initialized for a variety of times of the day and year to explore and quantify high‐latitude off‐great circle propagation associated with the solar terminator. Analysis of these simulations show large scale east‐west ionospheric gradients due to the solar terminator can cause lateral deviations in north‐directed propagation paths exceeding 20° at sunrise and sunset depending on radio wave frequency, though the largest portion of received signal power tends to experience maximum deflections of 5°. An exploration of the dependence of propagation direction on deflection shows that propagation paths parallel to the solar terminator tend to experience the largest deflections. Since the solar terminator at high latitudes is at an angle with respect to north in the winter and summer, propagation paths oriented west or east of north can experience larger deflections than north oriented paths at sunrise and sunset during these times of year. Impacts of these diurnal deflections on the operation of OTHR and scientific radar are discussed, as well as possible strategies for mitigating them.

Abstract The practical applications within the domain of the fifth generation (5G) and the emerging beyond 5G network necessitate a high data transmission rate along with minimal achievable delay. With this objective in focus, the maximum capacity is extensively quantified through the utilization of the delay‐constrained effective capacity (EC) technique, which stands in contrast to Shannon's ergodic capacity. The current study is engaged in the analysis of EC within a delay‐limited wireless system operating in a double‐shadowed Rician (DSR) fading channel. Within this channel, only the Nakagami‐m distribution concept has been applied to both the dominant and secondary shadowing components of the proposed network model. A new exact closed‐form expression for EC within the DSR fading channel has been derived using the Fox‐H function. Furthermore, an analysis has been conducted for both high and low signal‐to‐noise ratios to provide further insights and explanations for the proposed model. It is worth noting that the results obtained from both simulation and analytical methods exhibit substantial similarity, revealing interdependence among various parameters present in the proposed model.

Abstract High latitude ionospheric density structures such as polar cap patches and arcs are capable of deflecting high frequency (HF) radio waves to off‐great circle paths, and are likely detrimental to technologies dependent on HF radio propagation. In this study, nearly 2.5 years of 4.6–14.4 MHz data from a multi‐frequency HF radio link between Qaanaaq, Greenland and Alert, Canada are used to investigate high‐latitude off‐great circle propagation in the polar cap. After an example of HF radio propagation affected by polar cap patches is shown in detail, a statistical analysis of the occurrence and impacts of off‐great circle deflections in the polar cap is presented. Off‐great circle propagation is shown to be increasingly common with increasing frequency up to 11.1 MHz, such that averaged over 1 year, received 11.1 MHz signals experienced deflections >30° from the great circle direction 65.6% of the time. The occurrence of these deflections across the year is shown to be at a maximum in the winter, while occurrence across the day varies with season. Trends across both time of day and time of year for 11.1 and 14.4 MHz deflections are consistent with polar cap patch occurrence trends. Off‐great circle deflections are shown to be associated with increased time‐of‐flights, a larger range of positive and negative Doppler shifts, increased Doppler spreads, and lower signal‐to‐noise ratios. These results are discussed in the context of ionospheric phenomena in the polar cap, and implications for over‐the‐horizon radars operating at high latitudes.

Abstract This work studies variations of ionospheric total electron content (TEC) during four distinct solar eclipse events over the Ethiopia region. Dual‐frequency global positioning system (GPS) data obtained from UNAVCO over Addis Ababa (9.036°N, 38.76°E) and Bahir Dar (11.6°N, 37.34°E) stations are used to examine the ionospheric variability during two annular solar eclipses on 15 January 2010 and 1 September 2016, a partial solar eclipse on 4 January 2011, and a hybrid solar eclipse (the eclipse path starts out as annular but later changes to total) on 3 November 2013. The results show a significant decrease in TEC values during the occurrence of the solar eclipses. Specifically, the TEC values are reduced to −20% and −10% during the annular eclipse on 15 January 2010, −33% and −38% during the partial solar eclipse on 4 January 2011, −26% and −24% during the annular solar eclipse on 1 September 2016, over the Addis Ababa and Bahir Dar stations, respectively. There is only minimal change in TEC of −8% and −9% at Addis Ababa and Bahir stations, respectively, during the 3 November 2013 solar eclipse even if the obstruction rate is high over the study area. Furthermore, the study shows that the spatial gradient of TEC reduction varies at different locations, which is attributed to the distinct amount of reduction in solar radiation reaching the Earth's surface, resulting in reduced photo‐ionization. Overall, this study provides insightful information about the behavior of the ionospheric TEC during solar eclipses over Ethiopia and emphasizes the use of dual‐frequency GPS data in tracking the variations of the TEC.

Abstract Presented here are the results of an investigation conducted to determine the effectiveness of deep learning (DL)‐based systems utilizing the power of transfer learning for detecting breast cancer in histopathological images. It is shown that DL models that are not specifically developed for breast cancer detection can be trained using transfer learning to effectively detect breast cancer in histopathological images. The outcome of the analysis enables the selection of the best DL architecture for detecting cancer with high accuracy. This should facilitate pathologists to achieve early diagnoses of breast cancer and administer appropriate treatment to the patient. The experimental work here used the BreaKHis database consisting of 7909 histopathological pictures from 82 clinical breast cancer patients. The strategy presented for DL training uses various image processing techniques for extracting various feature patterns. This is followed by applying transfer learning techniques in the deep convolutional networks like ResNet, ResNeXt, SENet, Dual Path Net, DenseNet, NASNet, and Wide ResNet. Comparison with recent literature shows that ResNext‐50, ResNext‐101, DPN131, DenseNet‐169 and NASNet‐A provide an accuracy of 99.8%, 99.5%, 99.675%, 99.725%, and 99.4%, respectively, and outperform previous studies.

Abstract VHF and UHF are widely used frequency bands, and accurate path loss prediction in these bands is of great significance. In order to improve the prediction accuracy and stability of radio wave propagation, we propose a combined modeling method based on the entropy weight method. This method assigns appropriate weights to the predicted values of sub‐models using entropy, and calculates the final predicted value through weighted aggregation. To validate the feasibility of our proposed method, we conduct modeling using data from Denver, Colorado. We compare our method with the ITU‐R models and a model established using the coefficient of variation method. The results demonstrate that our proposed model outperforms the compared models. Specifically, the root‐mean‐square error (RMSE) of our proposed model is reduced by 5.94 and 5.51 dB compared to the ITU‐R P.1546 and 2001 models, respectively. Moreover, our proposed model exhibits better stability than the ITU‐R models. We also use the improvement percentage (γ) to evaluate the degree of improvement of our model over the sub‐models, which is 49.13% and 47.86% for the ITU‐R P.1546 and 2001 models, respectively. Additionally, by comparing the RMSE and absolute error, we find that the entropy weight‐based method outperforms the coefficient of variation‐based method. This research demonstrates the superiority of the entropy weight‐based modeling method, providing a valuable channel modeling approach for path loss prediction and supporting the regional assimilation of radio propagation prediction methods.