The identification and localization of large-range, wide-band electromagnetic interference (EMI) sources have always been both costly and time-consuming. The measurements at different times and places are often required before a typical system can locate a target. In this paper, we proposed a 2D electromagnetic imaging system to localize interference sources and identify the EMI frequency in real time. In this system, an offset paraboloid with a diameter of three meters is designed for large-range EMI imaging, while a multi-channel digital signal acquisition system is developed for wide-band EMI localization. The located interference source is segmented by the maximum entropy method based on particle swarm optimization, and the modified generalized regression neural network (MGRNN) is applied to identify the EMI frequency effectively by excluding misleading effects of outliers. The experiment which has been completed on our dataset indicates that our approach not only increases accuracy by 5% compared with the standard generalized regression neural network approaches for identification, but also exerts a large-range wide-band localization of the EMI source detection method.

An electric field measurement system with two optical pulse sources down-converts the high-frequency electric signal to the baseband through a probe exploiting the electro optic-effect. From that process, it is feasible to recover the measured frequency. However, the majority of the current studies focus on the case where the difference between the two repetition frequencies is quite small, while only a few involve larger frequency differences. Nevertheless, in a non-laboratory uncontrolled environment such as a field measurement, it is commonly unfeasible to guarantee a small frequency difference imposing further study on frequency recovery methods under larger frequency difference. Spurred by that, in this paper we propose a universal frequency recovery algorithm relying on the Remainder Matching (RM) technique, which is suitable for both large and small frequency differences. In our method, we employ the RM technique to determine the exact remainder combination in the first repetition frequency intervals and calculate the measured frequency. We also present the theoretical basis of RM, along with a summary of the system design rules, which have a significant impact on the performance of the measurement system. Finally, we challenge our RM-based solution on noise and analyze the corresponding false alarm rate metric. Our results demo nstrate that the proposed RM-based frequency recovery algorithm attains an appealing performance.

This paper introduces an under-sampling E-field measurement system involving three femtosecond pulse lasers, which can be used to measure a multi-frequency electric field (E-field) in a wide frequency range. Optical under-sampling technology utilizes femtosecond laser pulses to sample the electrical signals and convert the high-frequency signal down to the first pulse repetition frequency (RPF) intervals for measurement, i.e., tens to hundreds of MHz. However, previous work rarely studied the measurement of multi-frequency signals. Therefore, in this work, we use three RPFs and propose the Remainder Matching (RM) method of three RPFs to eliminate false-alarm-triggered frequencies. Moreover, to solve the challenging problem of grouping the down-conversion components of each measured frequency, we suggest the Eigenvalue Matching of Remainder’s Difference (EMRD) method to obtain the correct group accurately. Hence, this work combines the RM and EMRD methods to improve the system’s measurement accuracy. To evaluate the suggested measurement technique, detailed theoretical analysis and simulation are performed to demonstrate the universality of our method that can effectively guide the design of E-field measuring systems using optical under-sampling technology. The proposed scheme manages an error rate less than 0.1% without any measurement error of the down-converted components when six frequencies exist simultaneously and is less than 5% when the measured frequency deviation is ±25kHz. Finally, the proposed system is challenged on noise perturbations, with the results demonstrating that the suggested combined RM and RDEM scheme attains an appealing performance.

In the research of passive millimetre wave (PMMW) imaging, the focal plane array (FPA) can realize fast, wide-range imaging and detection. However, it has suffered from a limited aperture and off-axis aberration. Thus, the result of FPA is usually blurred by space-variant point spread function (SVPSF) and is hard to restore. In this paper, a polar-coordinate point spread function (PCPSF) model is presented to describe the circle symmetric characteristic of space-variant blur, and a log-polar-coordinate transformation (LPCT) method is propagated as the pre-processing step before the Lucy–Richardson algorithm to eliminate the space variance of blur. Compared with the traditional image deblur method, LPCT solves the problem by analyzing the physical model instead of the approximating it, which has proved to be a feasible way to deblur the FPA imaging system.

The optical-frequency-comb-based (OFC-based) sampling method is of better efficiency than a heterodyne receiver when measuring non-cooperative signals over large frequency range. However, under-sampling during the down-conversion process often results in missing frequency information, necessitating multiple OFCs for recovery. This paper proposes a frequency-phase-mapping method with single OFC is proposed to recover the missing frequency information. Firstly, an asymmetric two-branch electro-optical-modulating structure is designed to acquire the phase difference caused by the time delay. Then, the acquired phase difference is mapped into the signal frequency by complex unit circle without 2π-blur. A dimension-expanding procedure is proposed by introducing multi-measuring to enhance the robustness performance. The influence of frequency range, signal-to-noise-ratio (SNR) and measuring times are analyzed in detail. The experiment indicates that the correct ratio of frequency recovery can reach 93.5% when the SNR is 20dB, frequency range is 1G-12GHz and the measuring times are 2, demonstrating the proposed method’s efficiency.

Microwave camera is a passive focal-plane microwave imaging system which is utilized for large-scale, wide-band electromagnetic interference (EMI) imaging. Microwave camera images the EMIs with their amplitude and location by the reflector and sensor array, like taking a photograph. This camera however only receives the power density of the EMIs within a wide band and the frequency information is aliased and abandoned. To address this issue, in this paper, we propose a frequency estimation method to recover the frequency of the wide-band EMI from its amplitude image without a need to add new hardware. For the developed imaging system, the proposed method realizes the frequency estimation in 3-6GHz frequency range with the resolution of 100KHz consistently in most of the central part of the field of vision (FOV). Our proposed method is also applicable to other passive imaging applications such as passive millimetre radar imaging and infrared imaging, to extract the related information of the detected objects.