Target Detection with Tightly-coupled Antennas: Analysis for Unknown Wideband Signals
Erfan Khordad*, Peter J. Smith, Sachitha C. Bandara*, Rajitha Senanayake*, Robert W. Heath Jr*the Department of Electrical and Electronic Engineering, University of Melbourne, Melbourne, Australia † \dagger School of Mathematics and Statistics, Victoria University of Wellington, Wellington, New Zeal · 2025
Read Target Detection with Tightly-coupled Antennas: Analysis for Unknown Wideband Signals by Erfan Khordad*, Peter J. Smith, Sachitha C. Bandara*, Rajitha Senanayake*, Robert W. Heath Jr*the Department of Electrical and Electronic Engineering, University of Melbourne, Melbourne, Australia † \dagger School of Mathematics and Statistics, Victoria University of Wellington, Wellington, New Zeal (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-00779.txt.
Introduction
Erfan Khordad, Peter Smith, and their co-authors study how tightly coupled antenna arrays can spot targets from echo signals. Such arrays place their elements close together, so the elements affect each other strongly. The team asks whether the wide range of frequencies these arrays can use helps detection.
Their setup assumes a sensing receiver that gathers time and frequency samples of echoes whose form is not known in advance. They handle echoes that stay steady, drift slowly, or change fast across frequency. A new detector is offered for the slow case, and likelihood based detectors cover the other cases. Tests show the new detector is more robust, and coupled arrays beat weakly coupled ones.
Catalog record
| CREATOR | Erfan Khordad*, Peter J. Smith, Sachitha C. Bandara*, Rajitha Senanayake*, Robert W. Heath Jr*the Department of Electrical and Electronic Engineering, University of Melbourne, Melbourne, Australia † \dagger School of Mathematics and Statistics, Victoria University of Wellington, Wellington, New Zeal |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-00779
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Target Detection with Tightly-coupled Antennas: Analysis for Unknown Wideband Signals Erfan Khordad*, Peter J. Smith, Sachitha C. Bandara*, Rajitha Senanayake*, Robert W. Heath Jr*the Department of Electrical and Electronic Engineering, University of Melbourne, Melbourne, Australia † \dagger School of Mathematics and Statistics, Victoria University of Wellington, Wellington, New Zealand † † \dagger\dagger Department of Electrical and Computer Engineering, University of California at San Diego, San Diego, USA Abstract This paper presents analysis for target detection using tightly-coupled antenna (TCA) arrays with high mutual coupling (MC). We show that the wide operational bandwidth of TCAs is advantageous for target detection. We assume a sensing receiver equipped with a TCA array that collects joint time and frequency samples of the target’s echo signals. Echoes are assumed to be unknown wideband signals, and noise at the TCA array follows a frequency-varying correlation model due to MC. We also assume that the echo signals are time varying, with no assumption on the temporal variation. We consider three regimes in frequency as constant, slowly or rapidly varying, to capture all possible spectral dynamics of the echoes. We propose a novel detector for the slowly-varying regime, and derive detectors based on maximum likelihood estimation (MLE) for the other regimes. For the rapidly-varying regime, we derive an extended energy detector for correlated noise with frequency and time samples. We analyze the performance of all the detectors. We also derive and analyze an ideal detector giving an upper bound on performance. We validate our analysis with simulations and demonstrate that our proposed detector outperforms the MLE-based detectors in terms of robustness to frequency variation. Also, we highlight that TCA arrays offer clear advantages over weakly-coupled antenna arrays in target detection. I Introduction For beyond-5G wireless communications systems, using a substantial number of antennas in an array is imperative to achieve high spectral efficiency. Hence, antennas are packed together, resulting in antenna spacing below half-wavelength, and therefore mutual coupling (MC) in such arrays is no longer negligible [ 1 , 2 ] . Ultra-dense arrangements of the antennas can also form quasi-continuous arrays [ 3 , 4 , 5 ] . MC has been widely regarded as a cause of performance degradation and its effect has been mitigated, e.g. [ 6 ] . However, the pioneering works such as [ 7 , 8 ] showed that high MC is advantageous. Tightly-coupled antennas (TCAs) with high MC yield a broad operational bandwidth, at the expense of increased design complexity. The wideband gain was shown by implementations [ 9 , 10 ] and physically-consistent modeling [ 1 , 11 , 2 ] . Physically-consistent models use circuit theory to link physics with communication theory [ 12 ] . Target detection is one of the areas that can significantly benefit from the merits of TCAs, as we show in this paper. Additionally, enhanced target detection using TCAs can also be beneficial for integrated sensing and communications (ISAC). Target detection was conducted for ISAC setups, e.g. in [ 13 , 14 ] , however without exploring the advantages of high MC. Target detection, as an independent problem, was studied for a range of system models, for example in [ 15 ] by assuming unknown MC, yet considering MC as an adverse factor. Target detection with unknown wideband signals was performed using multiple energy detectors in sub-bands [ 16 ] . However, energy detection is susceptible to noise uncertainty and is sub-optimal for correlated signals [ 17 ] . Wideband detection was also considered in [ 18 ] and in [ 17 , 19 ] detection was investigated by taking into account signal correlation models. For the multiple-antenna receiver in [ 17 ] only independent and identically distributed noise was considered. Also a single antenna receiver was considered in [ 19 ] along with the presentation of asymptotic results. Furthermore, [ 19 , 18 ] focused solely on white noise. All these methods performed detection in one domain. Target detection using time and frequency samples was performed in [ 20 , 21 , 22 ] , however without detection performance analysis. Despite the above efforts, the merits of TCAs in enhancing target detection are still unexplored. In this paper, we show the detection gains resulting from employing TCAs. We assume that the signals reflected from the target are unknown and wideband. Thus our approach is applicable to any wideband waveform and can be utilized in both communications-centric and radar-centric ISAC waveform designs. We assume time-varying echo signals that fall into three regimes in the frequency domain as constant, slowly or rapidly varying, with no specific assumption on the rate of time variation. Also, we factor in the frequency-varying noise correlation caused by MC. We propose a novel detector for the slowly-varying regime and derive MLE-based detectors for other regimes. Our detector for the rapidly-varying regime is an extended energy detector for frequency-varying correlated noise, which collects time and frequency samples. We further derive a detector that gives an upper bound on detection performance. We present rigorous performance analysis by calculating the distribution of all the detectors under null and alternative hypotheses. We show perfect accuracy of our analysis by simulations and demonstrate that our detector has greater robustness than MLE-based detectors. We also show that TCA arrays exhibit clear benefits over conventional weakly-coupled antenna arrays in target detection. II System Model and Problem Formulation Following [ 1 ] , we present a physically-consistent system model using circuit theory, assuming canonical minimum scattering (CMS) antennas, which are suitable for wideband applications, to model TCAs [ 23 ] . A CMS antenna has an equivalent resistor-inductor-capacitor (RLC) circuit, based on Chu’s theory. We call a CMS antenna modeled based on Chu’s theory a Chu’s CMS antenna. The radius of a sphere that encloses the Chu’s CMS antenna is denoted by a_{R} . Impedance parameters of the RLC circuit are a function of a_{R} . Equivalence between Hertz dipoles with Chu’s CMS antennas, using radiated power, gives an expression for the mutual impedance of two Chu’s CMS antennas [ 1 , Eq. (5)] . This expression considers the antenna spacing denoted by \delta and the arrangements of the dipoles. We use a colinear arrangement in which the dipoles are placed along a straight line [ 1 , 11 ] . We consider a transmitter that transmits wideband signals. These signals reflect from a target and the echoes return to a sensing receiver with a uniform linear array modeled by Chu’s CMS antennas. We assume small antenna spacing to have a TCA array. Wideband signals are used with the purpose of using the broad operational bandwidth of TCAs for sensing. We assume echoes returning to the sensing receiver are unknown and wideband. Thus, our approach is valid for any wideband waveform. We denote the number of antennas at the sensing receiver by N_{R} . Since antennas at the sensing receiver are modeled as Chu’s CMS antennas, the antenna array at the sensing receiver is modeled by RLC circuits. Thus, the signal received at the sensing receiver is represented by a voltage vector denoted by \mathbf{v}_{k,t} , with k and t being the k th frequency sample and t th time sample respectively. Define K and T as