On-chip high-Q Micro-Ring Resonators （MRRs） have emerged as essential building blocks for high-performance microwave photonic systems， thanks to their compact footprint， narrow resonance bandwidth， and excellent tunability. They are widely employed in applications such as ultra-narrowband Microwave Photonic Filters （MPFs） for channel selection in radio-over-fiber networks， low-phase-noise Optoelectronic Oscillators （OEOs） for high-frequency microwave signal generation， and broadband optical frequency combs for high-resolution spectroscopy and microwave-to-optical conversion. Recent advances have focused on pushing Q factors into the multi-million range or even higher to further enhance device performance. This paper reviews the fundamental principles， design strategies， and material considerations for on-chip high-Q MRRs， with emphasis on their integration and practical deployment in these representative microwave photonic applications.

The Q factor of a MRR characterizes the energy storage capability and loss level within the optical cavity. A higher Q indicates a longer photon lifetime， lower round-trip loss， and a narrower resonance linewidth. In microwave photonic system， this translates into a narrower filtering bandwidth of MPFs， a lower phase noise of OEOs， and enhances nonlinear optical effects. Accordingly， we present a detailed discussion of the fundamental structure， theoretical framework， and key parameters of high-Q MRRs， establishing a solid foundation for their design and optimization. From a coupling perspective， MRR behavior can be understood via two complementary models： power coupling and energy coupling， which correspond to frequency-domain steady-state analysis and time-domain transient analysis， respectively.

Subsequently， we elaborate on implementation strategies for high‑Q MRRs from three key aspects： material platform selection （including silicon， silicon nitride， lithium niobate， and lithium tantalate）， fabrication process optimization， and waveguide engineering. A cross‑platform comparison is carried out to evaluate material systems in terms of refractive index contrast， optical transparency window， nonlinear optical properties， and fabrication maturity. This analysis reveals the inherent trade‑offs among platforms—for instance， silicon offers high index contrast and mature CMOS compatibility but suffers from two‑photon absorption in the telecom band， whereas silicon nitride exhibits broad transparency and ultralow loss， albeit requiring more advanced processing to reach ultrahigh Q values. On the waveguide engineering front， techniques such as adiabatic width tapering， Matched bend， Euler bend， and Bezier bend are employed to minimize radiative and bending losses， enabling smooth mode transitions and supporting ultrahigh‑Q operation in compact footprints. Recent progress in high‑Q MRR development across these material platforms is also reviewed.

Furthermore， we examine the typical implementations of high‑Q MRRs in microwave photonic systems， with a particular focus on MPFs and OEOs. For MPFs， we begin by highlighting the potential of ultra‑high‑Q MRRs with large Free Spectral Range （FSR） in constructing ultra‑narrowband， widely tunable MPFs. We then review various system architectures developed to achieve ultra‑high out‑of‑band rejection and high reconfigurability. However， for real‑world RF applications， MRR‑based MPFs still require significant improvement in key RF performance metrics such as insertion loss， Noise Figure （NF）， and Spurious-Free Dynamic Range （SFDR）. Regarding OEOs， we discuss developments in on‑chip OEOs based on MRR， as well as emerging PT‑symmetry breaking schemes that eliminate the requirement for additional narrowband optical or electrical filters. In recent years， on‑chip integrated OEOs have made notable progress in frequency tuning range， Side‑Mode Suppression Ratio （SMSR）， and system compactness. However， the limited on‑chip effective delay， combined with additional noise introduced by active components， results in overall phase noise performance that remains considerably higher than that of discrete fiber‑based systems—particularly evident at intermediate and low frequency offsets. We also summarize recent progress in fully integrated on‑chip MPFs and OEOs， which is of great significance for achieving compact， lightweight microwave photonic systems.

Prospectively， high‑Q MRR‑based microwave photonic devices are poised to transition into practical use across a range of fields， including radar signal processing， 5 G/6 G wireless communications， and RF signal generation. By leveraging programmable architectures and adaptive control schemes， these devices can simultaneously achieve miniaturization， low power consumption， and high performance. This progress is expected to drive the broader adoption of microwave photonics in cutting‑edge domains such as telecommunications， national defense， aerospace， and intelligent sensing， while gradually narrowing the performance gap between on chip solutions and discrete high performance systems.

The ability to generate dynamically reconfigurable broadband arbitrary waveforms is of paramount importance for advanced radar imaging， next-generation wireless communications， integrated sensing and communication， and electronic warfare systems. These applications demand waveforms with large time-bandwidth products， complex modulation， and low phase noise to ensure high resolution， superior data throughput， and robust anti-jamming resilience. However， traditional electronic approaches are fundamentally constrained by the electronic bottleneck， which severely restricts instantaneous bandwidth， tuning agility and power efficiency， particularly at high frequencies. In this context， microwave photonics has emerged as a transformative paradigm by leveraging the intrinsic advantages of photonic technologies， including ultra-wide bandwidth， low transmission loss， and immunity to electromagnetic interference. In this field， opto-electronic oscillators are well known for generating microwave signals with ultra-high spectral purity， which is enabled by high-quality-factor resonators formed using long optical fiber delay lines. However， the inherently single-frequency nature of conventional opto-electronic oscillators renders them ill-suited for complex and wideband waveform synthesis.

The incorporation of mode-locking mechanism into opto-electronic oscillator architectures has given rise to mode-locked opto-electronic oscillators， representing a major breakthrough that enables the coherent synthesis of broadband waveforms while maintaining low phase noise. This review systematically presents three representative mode-locked opto-electronic oscillator architectures： the Fourier-domain mode-locked opto-electronic oscillator， the wideband injection-locked opto-electronic oscillator， and the active mode-locked opto-electronic oscillator. The Fourier-domain mode-locked opto-electronic oscillator employs a high-speed tunable filter that is periodically scanned in synchronization with the cavity round-trip time， thereby creating a dynamic spectral storage mechanism that enables the sequential amplification of distinct frequency components. Initially demonstrated for the generation of wideband linear frequency-modulated waveforms， the Fourier-domain mode-locked opto-electronic oscillator has since evolved to produce dual-chirp signals， phase-coded linear frequency-modulated waveforms， and fully programmable arbitrary waveforms through precise control of the filter scanning trajectory. In contrast to the Fourier-domain mode-locked opto-electronic oscillator， the wideband injection-locked opto-electronic oscillator introduces an external broadband seed signal， typically generated by an arbitrary waveform generator， into a free-running opto-electronic oscillator cavity. The high-quality-factor resonator functions as a spectral purifier， substantially reducing phase noise and enhancing the signal-to-noise ratio while faithfully replicating the modulation profile of the injected seed signal. Alternately， the active mode-locked opto-electronic oscillator employs periodic gain modulation at a frequency matched to the cavity free spectral range or its harmonics， thereby actively phase-locking multiple longitudinal modes to form a coherent microwave frequency comb that manifests as a pulse train in the time domain. The active mode-locked opto-electronic oscillator facilitates precise repetition-rate control and has evolved to support tunable dual-band operation， multidimensional pulse coding （time， phase， and amplitude）， and fully programmable architectures for seamless switching between diverse waveform formats.

This review provides a comprehensive analysis of the operating principles and system architectures for each approach， alongside a comparative evaluation of critical metrics such as bandwidth， phase noise， and time-bandwidth product. The wideband arbitrary waveforms generated by mode-locked opto-electronic oscillators hold great potentials for various applications. In high-resolution radar and imaging， these waveforms provide the bandwidth necessary to achieve millimeter-scale spatial resolution while ensuring robust resilience against electronic interference. In integrated sensing and communication systems， the wideband microwave signals leverage a unified hardware platform to support high-throughput communications alongside high-resolution sensing， effectively eliminating the trade-offs between data rate and detection accuracy. Furthermore， in advanced LiDAR systems， such waveforms support high-fidelity 4D imaging， capturing 3D structural details and instantaneous Doppler velocity with millimeter-level accuracy.

Future research is expected to advance along several key frontiers multi-material hybrid integration using materials such as InP， Si₃N₄， and LiNbO₃ to realize compact， stable， chip-scale devices； extension of high-frequency and ultra-wideband operation into the millimeter-wave and terahertz regimes； and the realization of software-defined， intelligent reconfiguration enabled by artificial intelligence for cognitive waveform adaptation. In summary， mode-locked opto-electronic oscillators constitute a versatile and powerful photonic platform that effectively overcomes the electronic bottleneck. They establish a robust foundation for the generation of wideband software-defined microwave waveforms， which are essential for advancing next-generation radar， communication， and sensing systems.

A Microwave Photonic Filter （MPF） is a device that filters microwave signals by first modulating them into the optical domain， then processing them using optoelectronic components. Compared to traditional electronic microwave filters， they exhibit advantages such as superior frequency tuning range， flexible spectrum reconfiguration， and inherent immunity to electromagnetic interference， etc. These benefits make MPFs highly promising for widespread applications in wireless communication， radar， and electronic warfare systems. With the advancement of integrated optoelectronic technologies， MPFs are progressively evolving from discrete fiber-optic devices towards integrated solutions. This transition aims to substantially reduce their size， weight， power consumption， and cost. In this review， firstly the system architectures and working principles of both incoherent and coherent integrated MPFs are presented. Subsequently， recent research advances in these two categories of MPFs are reviewed.

For incoherent integrated MPFs， multi-tap configurations are typically adopted， utilizing Finite Impulse Response （FIR） digital filter architectures rooted in discrete signal processing algorithms. Some key integrated optical components enabling dispersion-delayed sampling in multi-tap MPFs are given， such as integrated high-dispersion chips， multi-wavelength light sources， and spectral shaping chips. By adopting these functional chips into microwave photonic filter links， some MPFs with good frequency tuning capabilities and out-of-band RF suppression have been reported. But the limited number of taps and unattainable complex tap coefficients resulting in insufficient spectral reconstruction capability of the achieved incoherent MPFs. To date， only aforementioned integrated chips have been integrated into microwave photonic filter links， while other critical components still rely on discrete fiber-based devices. Incoherent MPFs with higher integration levels remain unreported in the literature.

For coherent integrated MPF， typically microwave signals are modulated onto a single-wavelength laser source， where various integrated optical filters are adopted to spectrally shape modulated optical signal， followed by down-conversion at a photodetector to convert optical domain filtering response into microwave domain. Various integrated optical components to construct coherent MPFs such as Micro-Ring Resonator （MRR）/Micro-Disk Resonator （MDR）， nonlinear As2S3 waveguide with Stimulated Brillouin Scattering （SBS）， photonic crystal cavity， microsphere， phase shifted waveguide Bragg grating， Surface Acoustic Wave （SAW） based on suspended waveguide have been reviewed. Based on these integrated optical components， various MPF link architectures to realized band-stop and band-pass MPFs have been introduced. Band-stop MPFs can be realized through optical-to-microwave mapping based on Optical Single Sideband （OSSB） modulation. Furthermore， by employing Radio Frequency （RF） cancellation techniques via Unbalanced Optical Double Sideband （UODSB） modulation， this approach can overcome the limitation imposed by optical filter’s extinction ratio on RF suppression performance and enables band-stop MPF with RF rejection ratio larger than 60 dB. On the other hand， Band-pass MPFs can be realized through either by optical carrier separation and re-entry techniques enabling direct one-to-one mapping of optical band-pass responses to the microwave domain， or phase-to-intensity conversion schemes that transform optical band-stop filtering into microwave band-pass responses. In addition， some technologies to enhance MPF's frequency resolution have been introduced， such as improving quality factors of MRR/MDR with mode manipulation， incorporating on-chip narrow-band SBS gain/loss spectra， adopting SiO₂ microsphere， etc. Up to date， frequency resolution down to about ten MHz have been demonstrated for band-pass and band-stop MPFs. Building on recent advances in integrated optical filters for MPF links， this work futher surveys highly integrated microwave photonic filter chips implemented on Thin-Film Lithium Niobate （TFLN）， Silicon-On-Insulator （SOI）， and indium phosphide （InP） platforms， providing a comparative analysis of their respective advantages and limitations.

Finally， quantitative comparative analysis of key performance metrics of the reported incoherent and coherent integrated MPFs have been given. Some key technical challenges in integrated MPFs， including the trade-offs between filter's bandwidth and frequency tuning range， insufficient noise figure/out-of-band RF rejection ratio/frequency stability， immature heterogeneous integration， and broadband photonic packaging solution with low crosstalk and high RF integrity. Moreover， an outlook on the future development trends of integrated MPF is presented.

Phased array antennas， renowned for their capability to flexibly control spatially radiated electromagnetic waves， are widely used in fields such as communications， detection， and electronic warfare. As system demands for higher gain and broader operating bandwidth continue to increase， phased array antennas are evolving toward larger scales and higher operating frequencies.

Photonics technology exhibits notable advantages including strong resistance to electromagnetic interference， low transmission loss， and large operating bandwidth， offering new possibilities for further enhancing the performance of phased array antennas. By integrating optical and electromagnetic waves and leveraging the flexible control of optical parameters such as delay and intensity， it has become an important direction for the development of microwave and millimeter-wave phased array antennas. Within the framework of optoelectronic integration， relevant technical approaches can be categorized into three typical implementations： optical true-time-delay phased arrays， laser-space-fed phased arrays， and optically controlled metasurface antennas. The first two focus on providing high-precision time delay in the optical domain or through optical links， while the latter employs optical means to dynamically control the electromagnetic response of metasurface units.

Specifically， optical true-time-delay phased arrays utilize the large tuning range and low-loss characteristics of optical delay lines to achieve real-time delay control in the optical domain， effectively mitigating beam squint issues during wideband， large-angle scanning， thereby supporting beamforming and steering of wideband signals. Laser-space-fed phased arrays leverage the low-crosstalk nature of laser carriers to transmit optically carried microwave and millimeter-wave signals to radiating elements via free-space optical links， offering the potential to simplify complex electrical interconnections in large-scale arrays and providing a new technical pathway for implementing high-frequency， large-scale phased arrays. Optically controlled metasurface antennas retain the advantages of electronic metasurfaces， such as low cost and easy integration， while introducing light as an additional control dimension to enhance flexible beam manipulation. Moreover， thanks to the good penetration of light in underwater environments， this technology shows potential for future integrated space-air-ground-sea systems. It is worth noting that the rapid development of optoelectronic integration technology is driving continuous improvements in the performance， integration level， and cost-effectiveness of photonic devices. This lays a solid foundation for the miniaturization， integration， and weight reduction of optoelectronically integrated phased array systems， and also points to important technical directions for addressing the limitations in array scale.

This paper focuses on optoelectronically integrated microwave and millimeter-wave phased array antenna technology， with emphasis on the three typical architectures mentioned above： optical true-time-delay phased arrays， laser-space-fed phased arrays， and optically controlled metasurface antennas. It outlines their key technologies and development trajectories， reviews representative research progress， and， on this basis， analyzes and prospects the future application potential of optoelectronic integration technology in the field of optoelectronically converged phased arrays.

Magnetic field sensing is essential in applications such as power equipment monitoring， geomagnetic exploration， biomedical diagnostics， and intelligent sensing infrastructures. Although electrical magnetic sensors， including Hall-effect， magnetoresistive， fluxgate， and superconducting quantum interference device sensors， have achieved technical maturity， their deployment in complex or distributed environments is often constrained by electromagnetic interference， bulky shielding requirements， and limited scalability. By contrast， microwave photonic approaches exploit the immunity， flexibility， and low cost of optical fibers together with the high resolution and wide bandwidth of microwave-domain processing， offering a promising route toward high-performance magnetic field sensing. This paper reviews the fundamental principles and recent progress of microwave photonic magnetic field sensing technologies， with emphasis on their demodulation mechanisms and system architectures.

Two microwave photonic sensing routes have emerged as representative approaches. The first is based on the Optoelectronic Oscillator （OEO）， in which the fiber-optic magnetic field sensing unit is embedded within the OEO feedback loop. In this configuration， magnetic-field-induced optical perturbations are translated into shifts in the oscillation frequency or phase of OEO. Owing to the intrinsically high-Q and low phase noise of OEOs， even minute wavelength shifts from fiber Bragg grating Fabry-Perot cavities bonded to magnetostrictive materials can be amplified into measurable frequency shifts at the MHz to GHz level. Furthermore， Mach-Zehnder interferometer assisted OEO architectures enable alternating magnetic field sensing by converting magnetostrictive phase modulation into frequency or phase variations of the OEO oscillation signal， allowing sensitivities approaching the sub-nT/Hz^1/2 level in the kHz magnetic field band. Responsivity enhancement strategies， such as the Vernier effect in dual loop OEOs and prestressed magnetostrictive packaging， have been demonstrated to improve the scale factor by more than an order of magnitude without modifying the single loop OEO configuration.

The second route is based on Microwave Photonic Filters （MPFs）. In this scheme， an optical carrier is modulated by an electro-optic modulator and subsequently passes through a magnetic field sensing unit. After photodetection， the magnetic field information is retrieved by tracking the shift of the filter passband， typically using a vector network analyzer. MPF-based magnetic field sensing systems naturally support multi-tap and reconfigurable architectures， and can be applied to both static and alternating magnetic field measurements. Advanced implementations incorporating Fabry-Perot interferometers， polarization-maintaining fibers and cascaded fiber rings， MPF-based schemes enable enhanced responsivity， dual-parameter sensing， and quasi-distributed or vector magnetic field measurement capabilities.

To render microwave photonic magnetic field sensing systems practical for engineering applications， advanced post-processing techniques are essential to mitigate nonidealities originating from both magnetic transducers and microwave photonic components. Magnetostrictive and magnetic fluid elements inherently exhibit hysteresis and temperature-dependent behavior， while fiber packaging may introduce mechanical vibrations that manifest as excess phase noise. In addition， instabilities in lasers and photodetectors can induce power and wavelength fluctuations. Model-based hysteresis compensation approaches， including Preisach and Prandtl-Ishlinskii families and their dynamic or separated loop extensions， have been shown to substantially suppress nonlinear errors and restore near-linear relationship between magnetic field and frequency. In parallel， data-driven techniques based on recurrent or feedforward neural networks， support vector machines， and hybrid model-network frameworks enable online learning of the complex coupling among magnetic field， temperature， and packaging-induced stress. These techniques enable effective suppression of nonlinear errors and adaptive decoupling of magnetic field， temperature， and packaging-induced disturbances， thereby enhancing long-term stability and measurement fidelity.

In summary， microwave photonic magnetic field sensing technology combines the low cost and lightweight of fibers with the high accuracy of microwave-domain signal processing. At the hardware level， magnetic field demodulation can be achieved through microwave photonic architectures， including high-Q OEOs and tunable multi-tap MPFs. At the digital level， hysteresis can be effectively compensated using advanced signal processing and algorithms. Most reported systems still rely on discrete optoelectronic components. This reliance increases susceptibility to optical path loss and environmental perturbations， which in turn degrades demodulation accuracy and long-term stability. Future efforts should therefore focus on monolithic or hybrid photonic microwave integration to realize low loss， low cost， and compact on-chip sensing and demodulation architectures. Through coordinated control of the optical and microwave domains， this technology offers a versatile route toward high sensitivity magnetic field measurement. With continued advances in integrated microwave photonic platforms， intelligent signal processing algorithms， and novel magnetic transduction materials， microwave photonic magnetic field sensing is expected to evolve toward higher sensitivity and stronger environmental adaptability. These developments will also promote improved system integration and support future applications in power equipment monitoring， geomagnetic exploration， and biomedical magnetic detection.

The temporal Talbot effect represents the time domain counterpart to the classical spatial self-imaging phenomenon， strictly governed by the fundamental physical principle of space-time duality. This coherent interference effect occurs when a periodic optical pulse train propagates through a dispersive medium， such as a standard single mode optical fiber or a chirped Bragg grating， where the group velocity dispersion acts as a quadratic phase filter on the optical signal. Depending on the precise relationship between the pulse repetition period and the accumulated dispersion， the system exhibits two primary behaviors： integer order self imaging， where the original waveform is perfectly reproduced， or fractional order self imaging， which results in the generation of pulse trains with multiplied repetition rates. Furthermore， the dispersive propagation inherently functions as a real-time Fourier transform， mapping the frequency spectrum of an imposed modulation directly into its temporal intensity profile at the output. This review systematically explores these mechanisms， emphasizing their utility as passive， linear， and energy efficient tools for high speed signal processing in the field of microwave photonics.

A significant portion of this review addresses the manipulation of pulse repetition frequencies. While early approaches relied on fixed dispersion values， recent advancements allow for flexible multiplication factors. We discuss techniques such as phase pre-coding， which modifies the input phase to transition between fractional Talbot orders without energy loss， and binary intensity modulation， which offers a simpler implementation but suffers from optical power loss. A comparative analysis highlights that complex amplitude pre-coding scheme offers a balanced trade-off between energy efficiency and control complexity. By utilizing a Mach-Zehnder modulator to sculpt the spectrum into a flat profile， this method overcomes the bandwidth limitations of arbitrary waveform generators required for phase coding and the high extinction ratio constraints typical of intensity modulation schemes.

In contrast to multiplication， the paper examines passive amplification and weak signal recovery via the inverse temporal Talbot effect. By reversing the dispersion process， distributed energy from a continuous wave or a high-rate pulse train can be coherently focused into high peak power pulses. This concept， realized as a temporal Talbot array illuminator， effectively converts continuous wave light into high-quality pulse trains with tunable parameters. Crucially， this enables noiseless amplification where random， non-coherent noise does not sum up constructively， thereby significantly enhancing the signal-to-noise ratio. Experimental results have demonstrated amplification gains exceeding one hundred and ten times， allowing for the recovery of weak signals that are otherwise submerged below the noise floor of the photodetector.

The review further details the application of the real time Fourier transform in real-time spectrum analysis. Standard real-time spectrum analysis faces an inherent trade-off between measurement bandwidth and frequency resolution. To address this， the authors introduce a resolution enhanced architecture using the inverse temporal Talbot effect to temporally stretch the output observation window， effectively magnifying the resolution. Conversely， a bandwidth-enhanced architecture utilizing fractional-order multiplication is proposed to increase the effective sampling rate and measurement range. These innovations successfully decouple the resolution bandwidth constraint， enabling gap free analysis of wideband dynamic radio frequency signals. Additionally， the paper explores arbitrary waveform generation and pulse position control. By manipulating the spectral content via multi-tone modulation， complex temporal waveforms can be synthesized through frequency to time mapping. Similarly， precise control over pulse timing is achieved by modulating the carrier with a frequency modulated signal. This technique effectively converts instantaneous frequency shifts into temporal delays. This capability is applied to non-uniform sampling systems， such as the photonic Nyquist folding receiver， to significantly extend signal reception bandwidth beyond the traditional aliasing limits.

Finally， the review outlines emerging applications in other domains， such as temporal cloaking which hides events within dispersion induced time gaps， and prime number factorization via Gaussian sum calculation. The paper concludes by identifying critical challenges for future implementation. These include the precise management of high order dispersion to achieve perfect self imaging and the mitigation of Kerr non-linearities at high optical power levels. Furthermore， the review emphasizes the need for on-chip integration， noting the engineering difficulties in realizing sufficient dispersion on photonic integrated circuits while maintaining a compact Size， Weight， Power， and Cost （SWaP-C） footprint. Despite these hurdles， the temporal Talbot effect remains a versatile and promising foundation for next generation all optical signal processing networks.

Integrated Sensing and Communication （ISAC） has become a pivotal technology for 6G networks， addressing the increasing demands for data transmission and environmental awareness in advanced applications such as autonomous vehicles and smart IoT homes. However， conventional Radio Frequency （RF）-based ISAC systems face inherent challenges， including limited bandwidth， susceptibility to electromagnetic interference， and frequency-dependent losses， which constrain their ability to generate and process high-frequency broadband signals. Microwave Photonics （MWP） has emerged as a transformative solution that leverages optical technologies to overcome electronic bottlenecks， offering three key advantages： high-bandwidth signal processing capability， inherent immunity to RF interference， and a compact architecture with low power consumption. In communication systems， traditional multiplexing methods—such as Time Division Multiplexing （TDM）， Frequency Division Multiplexing （FDM）， and hybrid approaches—have been widely employed to enhance channel efficiency. Nevertheless， these methods exhibit limited resilience to interference. Code Division Multiplexing （CDM）， a well-established wireless technology derived from spread spectrum communication， uses unique codes to differentiate user data， enabling simultaneous transmission over the same frequency band for multiple users. CDM provides three significant benefits： robust multiple-access capability， intrinsic resistance to multipath interference， and improved signal security. Although prior research has demonstrated the potential of CDM， the integration of photonic-based Code Division Multiple Access （CDMA） with ISAC systems remains an underexplored area that warrants further investigation.

To address this research gap， we introduce a photonics-based ISAC system that combines spread spectrum coding， CDMA， and chirp waveform modulation. The proposed scheme features two main innovations： First， it achieves orthogonalization of user signals through specialized spread spectrum codes and scrambling sequences， effectively reducing random noise， inter-user interference， and multipath effects. This significantly enhances interference resistance and enables multi-user scalability. Second， the DC-biased spread spectrum encoding applied to Linear Frequency Modulation （LFM） signals generates integrated waveforms with unique chirp-phase characteristics and improved cross-correlation performance. This approach not only supports synchronization of communication sequences but also facilitates radar sensing via simplified photonic de-chirping processing， substantially lowering the Analog-to-Digital Converter （ADC） sampling rate requirements in radar receivers. For 3D positioning， we employ the Time Difference of Arrival （TDOA） method combined with a Chan-Taylor hybrid algorithm， enabling accurate localization with smaller network overhead. This integration of communication， sensing， and measurement systems has been successfully realized. Simulation results demonstrate the system's robustness： even under negative Signal-to-Interference-plus-Noise Ratio （SINR） conditions， spread spectrum phase coding sustains simultaneous multi-user communication and sensing. The system achieves decimeter-level distance resolution and high-precision near-field positioning. Parameter analysis evaluates the impact of DC bias， Voltage Peak-to-Peak （Vpp）， Received Optical Power （ROP）， and gain on communication performance in terms of Bit Error Rate （BER） and Signal-to-Interference Ratio （SIR）.

This research provides valuable insights for designing high-performance， interference-resistant integrated sensing and communication systems applicable to areas such as UAV control and management， smart homes， and intelligent transportation.

With the explosive growth of next-generation information technologies， such as ultra-high-speed 5G/6G wireless communication networks， ultra-wideband radar systems， and quantum information processing， the requirements for the generation， transmission， and manipulation of high-frequency microwave signals have reached unprecedented levels. Facing the escalating challenges of bandwidth， speed， and energy efficiency， Integrated Microwave Photonics （IMWP） has emerged as a transformative solution. By leveraging the ultrawideband characteristics of photons propagating within integrated circuits， IMWP achieves high-precision modulation， distribution， and processing of microwave signals through interference and multiplexing across diverse optical paths， thereby surmounting the performance bottlenecks inherent in traditional electronic systems. Furthermore， on-chip programmable photonic-microwave signal processors provide a robust hardware foundation for next-generation high-performance microwave systems by reconfiguring their internal architectures to achieve multi-functionality and flexible configuration. Although various architectures based on interferometer arrays， such as finite impulse response filters with delay-line arms and cascaded ring resonators， have been demonstrated， a critical bottleneck hindering their engineering transition is the difficulty in balancing architectural reconfigurability with system scalability in complex application scenarios. Conventional designs often suffer from fixed interconnection topologies and a lack of directness in functional implementation， which constrains their practical deployment in diverse and dynamic environments.

Addressing these limitations， this work proposes and experimentally demonstrates an adaptive programmable integrated photonic signal processor based on a hexagonal waveguide mesh topology. The processor utilizes Mach-Zehnder Interferometers （MZIs） as fundamental tuning units， featuring an optimized physical-layer design to ensure high-fidelity functional mapping across the network. The device was fabricated on a 220 nm Silicon-on-Insulator （SOI） platform， with a compact footprint of approximately 3 mm×1 mm， integrating 32 tunable MZIs， 32 thermo-optic phase shifters， and 24 optical I/O ports. Comprehensive characterization of the MZI test structures reveals excellent performance， including extinction ratios exceeding 20 dB and a half-wave heating power of 4.41 mW. A central innovation of this study is the implementation of a self-adaptive control framework driven by the Crested Porcupine Optimizer （CPO） algorithm. The operational logic of the CPO-based configuration is divided into three distinct phases： initialization， closed-loop evaluation， and strategy evolution. By generating a random population within the predefined voltage range and iteratively calculating fitness values based on real-time experimental spectral feedback， the system autonomously converges to the optimal voltage configuration. During the evolution process， the algorithm dynamically toggles between exploration and exploitation modes using four distinct evolutionary strategies， ensuring both global search coverage and local refinement precision. This self-adaptive framework enables the processor to achieve immediate operational readiness with minimal control complexity， establishing a highly stable hardware-software co-design for universal signal processing.

Experimentally， the fabricated processor was configured to demonstrate two representative microwave photonic functionalities： topological filtering and frequency measurement. In the topological filtering experiments， the mesh’s high degree of reconfigurability was leveraged to synthesize diverse spectral responses， including both Finite Impulse Response （FIR） and Infinite Impulse Response （IIR） filters. By precisely controlling the power splitting ratios and phase shifts within the hexagonal units， we successfully realized Asymmetric MZIs （AMZIs） with tunable extinction ratios exceeding 25 dB and Free Spectral Ranges （FSR） ranging from 0.17 nm to 0.64 nm. Furthermore， Optical Ring Resonators （ORRs） with a 6-BUL （Basic Unit Length） cavity and two-stage Coupled Resonator Optical Waveguides （CROWs） were implemented， demonstrating the mesh's capability to synthesize complex recursive structures with flattened transmission spectra. For frequency measurement， a broadband frequency identification system was constructed using an Amplitude Comparison Function （ACF） based on a Carrier-Suppressed Single-Sideband （CS-SSB） modulation scheme. The system utilized the mesh's ability to switch between multi-stage frequency discrimination channels with different FSRs. By synergistically combining the results from a wideband coarse-measurement channel and a narrowband high-precision channel， the system achieved a broad frequency identification range from 0.01 GHz to 30 GHz with a significantly improved Root Mean Square Error （RMSE） of 131 MHz. This collaborative measurement scheme effectively eliminates the periodic ambiguity typical of narrowband discriminators while maintaining high resolution across the entire spectrum.

In summary， this work provides a viable and scalable pathway for realizing high-performance， low-complexity integrated microwave photonic systems. The hexagonal mesh architecture demonstrates exceptional functional adaptability and structural scalability， marking a paradigm shift from Application-specific Integrated Circuits （ASPICs） to software-defined universal photonic processors. Beyond the filtering and frequency measurement demonstrations， the architecture serves as a general-purpose hardware platform capable of supporting diverse tasks， including tunable delay lines for wideband beamforming， arbitrary waveform generation， and reconfigurable optoelectronic oscillators. Theoretical analysis and experimental validation suggest that this self-adaptive framework can be extended to larger-scale networks， providing a clear evolutionary path for multi-functional， adaptive measurement systems in complex electromagnetic environments. The successful implementation of these functionalities on a single programmable chip paves the way for the future deployment of integrated photonics in advanced radar systems， 5G/6G communication infrastructures， and high-dimensional optical signal manipulation.

To meet the demands of future high-performance radar systems with superior resolution， agility and adptability， a reconfigurable radar waveform generation method based on the period-one （P1） oscillation dynamics of optically injected semiconductor laser is comprehensively investigated. The underlying principle involves perturbing the oscillation state of semiconductor laser through the injection of external continuous wave light. This perturbation alters the intracavity photons and carrier densities of laser， thereby exciting various nonlinear dynamics. Among these， P1 oscillation state exhibits asymmetric sideband modulation， which provides a theoretical basis for microwave signal generation.

Firstly， the characteristics of P1 oscillation， including the frequency tunability and fast frequency-switching capability， are numerically analysed by solving the nonlinear rate equations. Simulation results demonstrate that the transition between distinct oscillation frequencies can be achieved with a switching speed on the order of ns level， highlighting the inherent agility of the optical injection system. The evolution routing of the nonlinear dynamics is also explored through the dynamical map as a function of injection strength and detuning frequency， which provides a guide to achieve and maintain the desired P1 oscillation state. Based on this， the experimentally generation of broadband radar signals are demonstrated. To generate broadband radar signal， a dynamical control unit of injection parameters is proposed to manipulate the instantaneous frequency of generated wideband waveform. By programming the control signals， various broadband radar signals are generated successfully， including linear frequency modulated signal， triangular wave signal， stepped-frequency signal， frequency-codded signal， frequency-codded linear and frequency modulated signal. The key waveform parameters of broadband signal such as bandwidth， time period， operating band and duty cycle， can also be flexibly adjusted by tuning the injection parameters， which further validate the feasibility of the proposed system. However， the optical injection system in application falls short in performance due to a challenge induced by the intrinsic noise of semiconductor lasers. to address this， approaches for enhancing the system performance are proposed and demonstrated. For single-frequency signal generation， a dual-loop optoelectrical feedback structure incorporating balanced photodetection is employed to improve the Side-Mode Suppression Ratio （SMSR） and phase noise of generated signals.

Compared with conventional optical injection system， the linewidth of generated signal based above method is significantly narrowed by three-orders of magnitude. The SMSR is enhanced by about 65.5 dB due to the Vernier effect. In addition， the system maintains excellent frequency tunability while achieving a phase noise of below -124.10 dBc/Hz at 10 kHz offset. For broadband radar waveform generation， pre-compensation of the injection intensity and Fourier Domain Mode Locking （FDML） mechanism are incorporated， through which the in-band Signal-to-Noise Ratio （SNR） and frequency accuracy of generated radar waveforms are significantly improved. In the experiment， a broadband signal with a bandwidth of 6.3 GHz is generated， of which in-band SNR is improved by about 47 dB. In addition， limited by the transient properties of semiconductor laser， when the P1 oscillation frequency is fast changed with a large frequency step， strong damping oscillation of the output frequency occurs， resulting in deterioration of the frequency stability and accuracy. The application of the FDML mechanism is shown to effectively suppress the damping oscillation， once again underscoring the advantages of the proposed signal generation method.

To validate the practical utility of the proposed radar waveform generation method， both Single-Input Single-Output （SISO） radar and a Multiple-Input Single-Output （MISO） radar systems are established. Firstly， based on the proposed SISO radar system， high resolution radar ranging is successfully realized， clearly distinguishing two closely spaced targets. The measured result shows excellent agreement with the actual distance. For MISO radar system， in the transmitter， the generated signal is emitted by means of time-division multiplexing across multiple transmitter channels. In the receiver， broadband de-chirp processing is performed via IQ mixing. When detecting a drone in an experimental scenario， the target is precisely located， which verifies that the proposed radar waveform generation system is a promising solution to construct efficient and low-complexity microwave photonic radars.

This research is driven by the critical need to overcome the inherent limitations of conventional Radio Frequency （RF）-based Over-The-Horizon （OTH） communication systems. Traditional systems， which rely on the propagation of microwave signals through the atmosphere or via ionospheric reflection， face significant challenges that constrain their performance in modern， high-demand scenarios. These challenges include severely limited bandwidth capacity， which restricts data throughput； vulnerability to both intentional jamming and unintentional electromagnetic interference， compromising reliability； and a lack of operational flexibility， making real-time reconfiguration and multi-channel parallel processing difficult. Furthermore， achieving wideband， cross-frequency operation often necessitates complex， bulky， and power-intensive electronic components. The primary objective of this work is to conceptualize， design， and validate a next-generation OTH communication system architecture that fundamentally transcends these electronic bottlenecks. By strategically integrating the field of microwave photonics—which utilizes light to generate， process， and distribute microwave signals—this study aims to create a system prototype capable of delivering unprecedented levels of bandwidth， robust anti-interference performance， and dynamic reconfigurability. The ultimate goal is to establish a concrete technological foundation for future OTH platforms that can meet the escalating demands for secure， high-capacity， and resilient long-distance tactical and strategic communication links.

To achieve the stated objectives， a comprehensive system design and simulation-based research methodology was employed， centered on a novel architecture that synergizes three core microwave photonic subsystems. The system's front-end is a microwave photonic phased-array antenna. This antenna comprises 64 independent elements designed to simultaneously capture weak wireless signals from free space. The received RF signals from each element are first amplified by a dedicated Low-Noise Amplifier （LNA） to mitigate front-end noise. Subsequently， each amplified RF signal is used to directly modulate the intensity of a continuous-wave optical carrier generated by a Directly Modulated Laser （DML）， thereby translating the electrical signals into the optical domain. This process creates 64 parallel optical RF channels. A critical innovation lies in the beamforming network. The 64 optical signals are fed into a tunable optical delay and attenuation module. A centralized digital control unit precisely and independently adjusts the time delay and attenuation for each of the 64 optical paths. This optical True-Time-Delay （TTD） approach is frequency-independent， enabling wideband， squint-free beam steering and shaping. The individually processed optical signals are then coherently combined via a Wavelength Division Multiplexing （WDM） stage， effectively synthesizing the desired radiation pattern in the optical domain before detection. The combined optical beam is converted back into a consolidated electrical RF signal using a high-speed Photodetector （PD）. This aggregated RF signal is then routed to the second core subsystem： the microwave photonic frequency conversion unit. The heart of this unit is a coherent dual Optical Frequency Comb （OFC） setup. One OFC serves as a multi-wavelength Local Oscillator （LO）， while the other is used for signal modulation. Electro-Optic Modulators （EOMs） are used for precise signal imprinting and frequency shifting. A key technique implemented is Carrier-Suppressed Single-Sideband （CS-SSB） modulation， achieved by carefully biasing the EOMs， which eliminates LO leakage and provides excellent channel isolation. This setup allows for the simultaneous up-conversion or down-conversion of multiple RF channels across a ultra-wide bandwidth. The final subsystem is an integrated communication terminal， which performs demodulation， decoding， and protocol processing on the recovered baseband signals. The feasibility and performance of this integrated architecture were rigorously validated through detailed theoretical modeling and extensive link-level simulations using the industry-standard Optisystem software， which models the complex interactions between optical and RF components.

The simulation and analysis of the proposed microwave photonic OTH system yielded highly promising results across multiple performance dimensions， confirming the effectiveness of the chosen architectural approach. Core Communication Performance： The system successfully demonstrated stable and high-fidelity signal transmission across the entire 4~16 GHz operational bandwidth. Within this spectrum， a sustained communication data rate of 200 Mbps was achieved， representing a significant improvement over typical narrowband OTH links and validating the system's high-bandwidth capability. Reconfigurable Channelized Processing： A major accomplishment was the implementation of reconfigurable microwave photonic channelized frequency conversion. The system possesses the ability to dynamically process signals across 13 independent wavelength channels. This channelization allows for simultaneous， parallel handling of multiple communication streams or signal sub-bands within the wide RF aperture， greatly enhancing spectral efficiency and multi-user capacity. Phased Array Link Performance： For the 64-element microwave photonic phased array link， detailed simulation confirmed that multi-channel parallel processing was feasible with minimal signal quality degradation. Critically， the complex beamforming and signal combining processes were achieved with a signal-to-noise ratio （SNR） degradation of less than 3 dB， indicating highly efficient optical processing and minimal noise introduction from the photonic beamforming network. Optical Frequency Comb Conversion Performance： The dual-OFC frequency conversion subsystem exhibited exceptional precision and spectral purity. Using an array of 23 finely spaced optical comb lines， the system realized the reconfigurable switching functionality for the 13 channels. Specifically， it accomplished the up-conversion of a 336 MHz Intermediate Frequency （IF） signal to the target 4~16 GHz RF band. The converted spectrum showed excellent characteristics： a well-defined channel spacing of 1 GHz， a spurious suppression ratio better than 20 dB （indicating minimal unwanted harmonic generation）， and a high in-band amplitude flatness with fluctuations not exceeding 4 dB across the entire 12 GHz range. This suite of metrics confirms the system's ability to perform precise， MHz-level frequency control and management with high spectral integrity.

In conclusion， this research has successfully established the technical viability and superior potential of a microwave photonic approach to revolutionizing over-the-horizon communication. The proposed and analyzed system architecture， integrating a photonic phased array， a dual-optical-frequency-comb-based channelizer， and an integrated terminal， directly addresses and overcomes the fundamental bandwidth， interference， and flexibility constraints inherent in traditional electronic OTH systems. The simulation results provide compelling evidence that such a system can deliver wideband （4~16 GHz）， high-data-rate （200 Mbps）， multi-channel （13 channels）， and reconfigurable communication links with high spectral purity and efficient beamforming. Beyond the immediate performance metrics， this work lays a critical and substantial foundation for the future engineering and deployment of advanced OTH communication systems. It demonstrates a clear pathway toward building more secure， resilient， and high-capacity strategic communication infrastructures. Future work will naturally focus on the practical implementation of this architecture， including the development of integrated photonic chips to reduce size， weight， and power consumption （SWaP）， experimental validation in real-world propagation environments， and the exploration of advanced signal processing algorithms to further enhance performance. This study marks a significant step forward in the convergence of photonics and wireless communications for national security and civilian long-range connectivity applications.