Feng Qi

A system-level simulator for indoor mmW SAR imaging and its applications

Recently, the research interest in indoor active millimeter wave (mmW) imaging by applying the synthetic aperture radar (SAR) technique is increasing. However, there is a lack of proper computer-aided design (CAD) tools at the system level, and almost all the R&D activities rely on experiments solely. The high cost of such a system stops many researchers from investigating such a fascinating research topic. Moreover, the experiment-oriented studies may blind the researchers to some details during the imaging process, since in most cases they are only interested in the readout from the receivers and do not know how the waves perform in reality. To bridge such a gap, we propose a modeling approach at mmW frequencies, which is able to simulate the physical process during SAR imaging. We are not going to discuss about advanced image reconstruction algorithms, since they have already been investigated intensively for decades. To distinguish from previous work, for the first time, we model the data acquisition process in a SAR imaging system successfully at mmW frequencies. We show how to perform some system-level studies based on such a simulator via a common PC, including the influence of reflectivity contrast between object and background, sampling step, and antennas directivity on image quality. The simulator can serve system design purposes and it can be easily extended to THz frequencies.

LIMITATIONS OF APPROXIMATIONS TOWARDS FOURIER OPTICS FOR INDOOR ACTIVE MILLIMETER WAVE IMAGING SYSTEMS

To simulate imaging systems, Fourier optics has been applied very successfully to optics for decades. However, when simply moving to indoor millimeter wave imaging systems, some assumptions underlying the Fourier optics may break down, which contribute to the errors by applying Fourier optics. During the review of mathematical derivation of the Fourier optics, we point out how the errors are introduced by making the Fresnel approximation and omitting the phase factors. To distinguish from much literature, we discuss the accuracy of Fresnel approximation rather than plane wave. Moreover, we check the simulation results for millimeter wave imaging systems working in both pixel scanning mode and focal plane array mode and compare them to the results predicted by Fourier optics. It is shown that the difierence can be 28% for the speckle contrast when the object is with certain roughness. The optical routine is that when the lens is four times larger than the object, the imaging system can be considered as isoplanatic, thus Fourier optics can hold. Our simulation results imply that it may not be valid in indoor millimeter wave imaging systems. The goal of this paper is to draw some attention to the possibly large errors when modeling or designing the indoor millimeter wave imaging systems by Fourier optics directly. The mathematical discussions of the inaccuracies due to some approximations in Fourier optics can serve to understand and deal with aberrations.

Millimeter wave imaging: System modeling and phenomena discussion

This paper presents a simplified calculation method to model active millimeter wave imaging system based on Huygens principle. Though limited to its ability to do the calculation accurately, it still gives a good evaluation of the performance of practical 3D systems. Thus the simulator can act as an effectivel tool at the beginning of the system design. The inherent effects in coherent imaging such as ringing, speckle and glint will be shown. To deal with speckle, a Hadamard diffuser is suggested. Influence of the pattern and size of the Hadamard cell are discussed accompanied with the explanation from an array antenna point of view. Finally, mult-frequency and multi-angle illumination methods are shown with possible enhancement to the image.

10 aJ-level sensing of nanosecond pulse below 10 THz by frequency upconversion detection via DAST crystal: more than a 4 K bolometer

By using frequency upconversion detection of terahertz (THz) waves via 4-dimethylamino-N-methyl-4-stilbazolium tosylate (DAST) crystal with an optimized frequency conversion process, ultrahigh sensitivity has been achieved. Direct comparisons with a 4 K bolometer were implemented. By using a simple positive intrinsic negative (PIN) diode without either electrical amplification or optical amplification, frequency upconversion detection can compete with the commercial 4 K bolometer, while by replacing the PIN diode with an avalanche photo diode (APD), it performs more than three orders better than the 4 K bolometer. Based on power calibration, the minimum detectable THz pulse energy is in the order of 10 aJ (9-25 aJ) at 4.3 THz, with a pulse duration of 6 ns. Thus, the minimum number of THz photons that can be detected is down to the order of 10(3) at room temperature. The current THz detection system gives a noise equivalent power (NEP) in the order of 100u2009u2009fW/Hz(1/2) (50-128u2009u2009fW/Hz(1/2)). Moreover, by switching current optical detectors, the dynamic range is over six orders.

Millimeter wave imaging system modeling: spatial frequency domain calculation versus spatial domain calculation

Active millimeter wave imaging systems have become a promising candidate for indoor security applications and industrial inspection. However, there is a lack of simulation tools at the system level. We introduce and evaluate two modeling approaches that are applied to active millimeter wave imaging systems. The first approach originates in Fourier optics and concerns the calculation in the spatial frequency domain. The second approach is based on wave propagation and corresponds to calculation in the spatial domain. We compare the two approaches in the case of both rough and smooth objects and point out that the spatial frequency domain calculation may suffer from a large error in amplitude of 50% in the case of rough objects. The comparison demonstrates that the concepts of point-spread function and f-number should be applied with careful consideration in coherent millimeter wave imaging systems. In the case of indoor applications, the near-field effect should be considered, and this is included in the spatial domain calculation.

DISCUSSION ON VALIDITY OF HADAMARD SPECKLE CONTRAST REDUCTION IN COHERENT IMAGING SYSTEMS

Hadamard speckle contrast reduction (SCR) is considered to be an efiective approach to deal with speckle problems in coherent imaging systems. A Hadamard SCR system is divided into two sub- systems, which implement phase patterns projection and re∞ected waves imaging respectively. The performances of both sub-systems are discussed with numerical simulations and linked to certain parameters so as to give more insights of this approach. For generality, both optical and millimeter wave imaging systems are discussed. To distinguish from former literature based on Fourier optics, the simulation is implemented via wave optics, which is more physical and more accurate. Moreover, considering the fact that the Hadamard method originates from statistics, the efiectiveness of Hadamard SCR is in the flrst place linked to the texture of the objects surface. Statistical optics is also adopted during qualitative analysis of the results. It is shown that the ratio between the dimension of a resolution cell and the granular size of the objects randomly rough surface is closely linked to the performance of Hadamard SCR. Difierences in the roughness model in imaging cases of optical and millimeter waves are discussed, which would help to evaluate the validity of the Hadamard SCR approach in practice. The purpose of this paper is to clarify the misunderstandings of Hadamard SCR in previous literature and to give a guideline to apply this approach.

A novel EM-based MAP channel estimation for MIMO-OFDM systems

For MIMO-OFDM system, expectation maximum (EM) is used to decrease the complexity of maximum a posteriori probability (MAP) channel estimation algorithm. However, EM-based MAP algorithm generates the low mean square error (MSE) performance at high signal noise rate (SNR) because of the convergent feature of EM algorithm. According to this issue, a modified EM-based MAP (MEM-MAP) channel estimation algorithm is proposed. The proposed algorithm improves convergent property of EM algorithm at high SNR by introducing an equivalent model. Then, according to the idea that MIMO channels in angle domain can be assumed to be independent, most significant taps (MST) technique in angle domain is used to improve the MSE performance of MEM-MAP algorithm. Simulation results indicate the effectiveness of the proposed algorithm.

Possible improvements of exisiting indoor MMW focal plane imaging systems

In this paper, we are going to touch some design essentials in terms of system considerations by taking a practical indoor millimeter wave (mmW) imaging system as a benchmark. Nowadays, system designers follow the guideline from classical imaging radars, which have a long history and aim at far-field operations. However, we are going to emphasize the difference when we focus on indoor mmW systems, which do not work in the far-field region. Basically, we think current indoor mmW imaging systems can be improved by two steps: first, proper sensor selection; second, optimum optical system design. We will illustrate how the resolution of imaging systems can be improved and discuss some design tradeoffs. The frequency of interest is around 100 GHz. However, the discussions can be extended to either microwave or THz imaging systems.

Low-complexity H-infinity channel estimator for MIMO-OFDM systems

Although maximum a posteriori probability (MAP) channel estimator has an optimal performance for MIMO-OFDM systems, its high complexity could be in the order of O(K3), where K is the number of subcarriers. So, in this paper, we introduce an H-infinity estimator which has good performance, but with much less complexity. Next, lots of matrices inversions and multiplications are reduced by the iterative expectation maximization (EM) algorithm. It is shown that the proposed EM-based H-infinity estimator has obviously improvement over optimal least square (LS) estimator with good convergence and the H-infinity estimator has almost the same performance as optimal MAP estimator with much less complexity.