This article comprehensively reviews recent developments and research on deep learning-based (DL-based) techniques for integrated sensing and communication (ISAC) systems. ISAC, which combines sensing and communication functionalities, is regarded as a key enabler for 6G and beyond networks, as many emerging applications, such as vehicular networks and industrial robotics, necessitate both sensing and communication capabilities for effective operation. A unified platform that provides both functions can reduce hardware complexity, alleviate frequency spectrum congestion, and improve energy efficiency. However, integrating these functionalities on the same hardware requires highly optimized signal processing and system design, introducing significant computational complexity when relying on conventional iterative or optimization-based techniques. As an alternative to conventional techniques, DL-based techniques offer efficient and near-optimal solutions with reduced computational complexity. Hence, such techniques are well-suited for operating under limited computational resources and low latency requirements in real-time systems. DL-based techniques can swiftly and effectively yield near-optimal solutions for a wide range of sophisticated ISAC-related tasks, including waveform design, channel estimation, sensing signal processing, data demodulation, and interference mitigation. Therefore, motivated by these advantages, recent studies have proposed various DL-based approaches for ISAC system design. After briefly introducing DL architectures and ISAC fundamentals, this survey presents a comprehensive and categorized review of state-of-the-art DL-based techniques for ISAC, highlights their key advantages and major challenges, and outlines potential directions for future research.

Low-profile foldable array antennas are becoming increasingly more important for a wide range of applications such as satellite communications and wearable electronic devices. The conventional arrays formed by patch-like antennas have been extensively studied on surfaces with a curvature but they have exhibited limited bandwidth and polarization performance. This study investigates a coupling enhanced crossed rings antenna array with two typical configurations for dual polarization, which inherently produces ultrawide bandwidth, dipole-like polarization characteristics and a fully curved array (FCA) eventually. The fractional bandwidth of the array is over 100% on a planar surface and expanded to approximately 140% on the curved surface. For the bent array of slant polarization, the beamwidth increases by over 20° compared to the planar array and cross polarization discrimination (XPD) maintains above 15 dB. The effects of curvature on the impedance matching and polarized radiation patterns for such arrays are investigated by measuring the performance of the fabricated prototype arrays. The results revealed that the tightly coupled crossed rings antenna array on a curved surface has a potential to form multiple beams on a limited aperture size through smaller subarrays which can yield ultrawide bandwidth due to concentrated mutual coupling mechanism. This characteristic is promising in applications where traditional flat panel arrays are difficult to implement such as in mobile stations, moving platforms and for satellite communication on-the-move.

Online calibration is desired in antenna arrays of ultrawide bandwidth. This study proposes a time domain calibration method based on the simultaneous perturbation algorithm. Two objective functions were established: power of the received signal at array output; or combination of power and correlation coefficient between the signal at array output and a target signal. For both criteria, the convergence settings require only two measurements at each iteration. One advantage of the method is that the entire signal operation for calibration is performed in the time domain. This is achieved by resolving the effects of distortion on time delay of each channel, which accounts for both amplitude and phase distortions at different frequencies. Therefore, the proposed method significantly increased the calibration efficiency for ultra-wideband antenna arrays. Since time delay coefficients for calibration associated with array elements were determined independently due to characteristic of the simultaneous perturbation, estimation accuracy of the method is tangential to the number of elements in the array, and is mainly dependent on the convergence conditions. This gives the method an additional distinct advantage for calibrating large-scale antenna arrays with ultrawide bandwidth. An estimation accuracy of 99% on time delay adjustments has been achieved and demonstrated.

Antenna arrays have gained significant interest in millimetre‐wave communication systems as an enabling technology to achieve higher capacity and mitigate the high propagation loss. Such arrays with a large bandwidth need to be efficiently calibrated to maximise their performance. An antenna array calibration method based on a stochastic approximation algorithm and simultaneous perturbation has been developed and the procedures to implement it in both frequency and time domains have been presented. The approaches to define objective functions and establish gradient approximations to fulfill a successful convergence for acquiring calibration coefficients in both domains have been explored. In the time domain implementation, only a fraction of the measurement time was required to calibrate an antenna array of ultrawide bandwidth compared with other methods using a perturbation technique. The effectiveness of the proposed method has been validated via numerical experiments in both domains.

The limitations of conventional sensors have made array antennas increasingly crucial for gathering information and communication applications in intelligent transportation and communication systems. Compact cylindrical arrays are particularly favored for their ability to achieve azimuth angle scanning. However, the substantial mutual coupling effect between the elements on curved surfaces and its implication for these arrays remain unclear, which is a key factor to consider when such arrays are used for multibeam applications. This study investigates the effect of mutual coupling in a dual-slant-polarized cylindrical array. The results showed that mutual coupling is predominantly observed among the closely located elements, and it is essential for achieving an ultra-wide bandwidth. The study also analyzes the impact of mutual coupling on the scan impedance and radiation characteristics for multibeam applications and reveals that these arrays exhibit robust multibeam capability, hence having great potential for use in sensing and communication applications.

Achieving reliable initial alignment of liquid crystal (LC) molecules on copper surface of microwave materials remains a key challenge, primarily due to a more significant surface roughness of radio frequency (RF) materials compared to smoother substrates traditionally used in optical devices, where LCs were initially tailored for. To address this problem, we studied two principle LC alignment strategies: (i) direct alignment of the LC layer on microwave device substrates via a polyimide (PI) coating, and (ii) integration of a conventional LC cell incorporating two glass substrates to provide well controlled surface anchoring. Both approaches are implemented in two distinct phase shifter designs and evaluated through full wave simulations and experimental measurements at the X- and Ku-bands respectively. The results demonstrated that the LC cell with glass substrates supports thinner LC layer, which is pivotal for obtaining a more rapid response. Devices based on the LC cell configuration exhibit a 35$\%$ improvement in response time, attributed to enhanced molecular orientation and reduced fringing field effects. However, the inclusion of extra glass substrates will degrade the phase shift range. The analytic study indicated that incooperation of glass layers reduced the differential phase shift ($\Delta \phi$) by 50$\%$ under the same LC thickness. These findings highlighted the advantages of glass substrate LC cells in enabling low loss, compact, and fast response phase shifters, making them well suited for reconfigurable antenna arrays and beamforming systems in next generation wireless communication networks.