Integrated Sensing and Communications

• Integrated Sensing and Communications is a unified framework that merges radar sensing with data transmission to maximize spectral efficiency and reduce latency.
• Key waveform design strategies modify traditional radar signals and leverage dual-functional waveforms like OTFS to balance communication and sensing trade-offs.
• Integration with enabling technologies such as RIS, AI, and massive MIMO enhances resource optimization, interference management, and scalability in next-generation networks.

Integrated Sensing and Communications (ISAC) refers to the unification of wireless sensing (e.g., radar, environmental monitoring, localization) and communications (data transmission) functions into a single system, waveform, and often hardware platform. Emerging as a key paradigm for 6G and beyond, ISAC enables improved spectral efficiency, reduced hardware redundancy, and low-latency situational awareness by allowing the same radio resources to be exploited for high-capacity data transmission and precise environmental/target sensing. This article details the principles, architectural paradigms, waveform design strategies, integration with enabling technologies, performance trade-offs, and open research challenges for ISAC in next-generation wireless systems.

1. Integration Levels and System Architectures

ISAC system architectures can be classified by the depth of integration:

1. Application-Level Integration: Sensing and communication systems remain physically separate but exchange information at higher layers. Example: sensor-assisted communications, where sensing informs beamforming or resource allocation (Wang et al., 2022).

2. Spectrum-Level Integration: Both functions share spectral, temporal, or spatial resources, and may even share hardware components; the sharing can be dynamically adjusted for cost and energy efficiency.

3. Full Integration: Sensing and communications share the same waveform and spectrum simultaneously, necessitating merged transceiver hardware and tightly coupled signal processing. Approaches include: (a) embedding communication into radar signals (e.g., modulated FMCW), (b) embedding sensing into communication waveforms (e.g., OFDM, OTFS), and (c) co-designing new dual-functional waveforms.

Additionally, network architecture has evolved from single-cell (one base station) to multi-cell and cloud/cooperative designs. Multi-cell and network-level ISAC enables collaborative and distributed sensing/communication, with benefits in robustness, geometric diversity, and scalability (Li et al., 26 May 2024, Strinati et al., 18 Feb 2024).

2. Waveform Design for ISAC

The central technical challenge in ISAC is waveform co-design. Three main categories exist:

A. Sensing-Origin Waveforms Modified for Communications:

• FMCW (Frequency-Modulated Continuous Wave) radar “chirps” can be phase- or symbol-modulated (e.g., PC-FMCW, CPM-LFM, chirp-shift schemes) to embed data (Wang et al., 2022). However, such modifications often increase out-of-band emissions and may degrade radar accuracy or require higher ADC performance.

B. Communications-Origin Waveforms Augmented for Sensing:

• OFDM (Orthogonal Frequency Division Multiplexing) is robust to multipath fading and is standard in LTE/5G/Wi-Fi. Sensing is performed by processing received echoes (via DFT) to extract delay (range) and Doppler (velocity). Main challenges include the need for large bandwidth for fine range resolution (e.g., 1 GHz for 15 cm), high ADC sampling rates, and difficulty in full-duplex OFDM radar due to high PAPR (Wang et al., 2022).

C. Novel Dual-Functional Waveforms:

• OTFS (Orthogonal Time Frequency Space): Encodes information in the delay-Doppler domain (via SFFT and Heisenberg transform). Offers resilience to Doppler and channel dispersion—beneficial for both high-mobility communications and radar (Wang et al., 2022, Kaushik et al., 2023). However, introduces 2D intersymbol interference, increasing equalizer complexity.
• OCDM (Orthogonal Chirp Division Multiplexing): Uses orthogonal chirps (full bandwidth) via the Fresnel transform; supports full diversity but complicates OOB emission suppression and multiuser MIMO (Wang et al., 2022).
• AFDM (Affine Frequency-Division Multiplexing): Allows adaptive chirp slope, enabling full diversity in doubly selective channels and improved performance over OCDM in some scenarios (Wang et al., 2022).

The trade-off in waveform design is balancing the conflicting requirements of communication (prefer random/amplitude signals, maximizing mutual information) and sensing (prefer constant modulus and low-correlation, maximizing parameter estimation accuracy), as analyzed using unified information models (Yu et al., 4 Jul 2024, Ouyang et al., 2022).

3. Mutual Information and Performance Trade-Offs

A mutual information (MI)-based analytical framework provides a unified metric for both communication and sensing (Ouyang et al., 2022, Yu et al., 4 Jul 2024). Central definitions:

• Communication MI: $I(X; Y \mid H)$, quantifying data throughput.
• Sensing MI / Sensing Mutual Information (SMI): $I(G; Y_S \mid X_S)$, quantifying the information extractable about environmental/target parameters.

Key results:

• ISAC systems achieve a strictly larger achievable (sensing rate, communication rate) region than conventional frequency-division architectures (FDSAC), due to concurrent resource use (Ouyang et al., 2022, Yu et al., 4 Jul 2024).
• At high SNR, ISAC gains degrees-of-freedom (rate slope) advantages because both functions span the entire resource block, whereas FDSAC systems are penalized by partitioning.
• Trade-offs must be managed via dual-functional beamforming and optimal time-frequency resource allocation; increasing resources for one function may diminish the other, but a linear proportionality exists between resource allocation and (communication MI, estimation MSE) performance (Yu et al., 4 Jul 2024); see Table 1.

Resource	Communication MI	Sensing MSE
+Δ time/freq	Increases linearly	Decreases ~50%

The MI–MMSE connection allows explicit Pareto front characterization and optimization, e.g., maximize $\rho\, I(X; Y_c|H_c) + (1 - \rho)I(H_s; Y_s|X)$, with $\rho \in [0,1]$ (Lu et al., 2023, Zhang et al., 9 Apr 2025).

4. Integration with Enabling Technologies

A. Reconfigurable Intelligent Surfaces (RIS):

RISs are passive (or nearly passive) planar metasurface arrays that can manipulate EM waves via programmable reflection coefficients (Liu et al., 2022, Chepuri et al., 2022, Kaushik et al., 2023). In ISAC:

• RIS assists by creating additional propagation paths for both communication and sensing—enhancing SNR, coverage, and enabling virtual line-of-sight links at mmWave/THz frequencies.
• Joint optimization of transmit beamforming and RIS parameters is essential but nonconvex, typically addressed with alternating optimization, penalty dual decomposition, or AI-driven learning (Liu et al., 2022).

B. AI:

AI/ML methods are increasingly essential for beamforming/waveform design, channel prediction, and adaptive control, compensating for incomplete models and high-dimensional nonconvex optimization (Vaezi et al., 17 Apr 2025, Liu et al., 2022).

• Deep neural networks can learn mappings from channels to waveforms, perform low-complexity online adaptation, or unroll iterative optimizers to guarantee real-time operation for large MIMO systems (Vaezi et al., 17 Apr 2025).
• AI supports robustness against model mismatch, non-stationary channels, and coordination among distributed agents.

C. Distributed and Networked ISAC:

Future 6G ISAC systems will be distributed and networked, leveraging cooperation among multiple user devices and network nodes for environmental sensing (computational imaging, collaborative target tracking). Distributed learning and semantic-aware data fusion enhance calibration, resolution, and robustness against link failures (Strinati et al., 18 Feb 2024, Li et al., 26 May 2024).

D. Multi-Beam and Massive MIMO:

Multi-beam ISAC leverages beam-space separation: communication beams are stable and targeted, while sensing beams are dynamically steered/scanned (Zhuo et al., 31 May 2024). Hybrid architectures may superimpose beams or approximate target patterns using alternating optimization, underpinned by highly directional mmWave/THz massive MIMO arrays.

5. Practical Implementations and Prototyping

Proof-of-concept ISAC systems have demonstrated the feasibility of simultaneous high data-rate communications and radar sensing on commercial 5G hardware (Wild et al., 2023). Notable aspects:

• OFDM radar processing chains can be repurposed from existing digital baseband processing.
• Synchronization using PTP/SyncE achieves jitter as low as 40 ps—critical for accurate time-of-flight and Doppler estimation.
• Clutter removal and KF-based target tracking algorithms can isolate moving objects (e.g., pedestrian detection) within typical communication frames, with practical range resolution and centimeter-level tracking accuracy.
• Such integrated platforms can extend sensing range well beyond laboratory setups, indicating scalability for urban or industrial contexts.

6. Security, Privacy, and Standardization

ISAC introduces new information-theoretic and physical layer security risks: sensing operations can reveal environmental or sensitive user information (e.g., through channel state information or echo patterns), and adversaries may exploit this for eavesdropping, spoofing, or privacy attacks (Lu et al., 2023). Countermeasures include:

• Artificial noise, secure beamforming, adaptive scheduling, and cryptographic protections aligned with emerging standards (Kaushik et al., 2023).
• Standardization efforts by ITU-R, 3GPP, and ETSI are ongoing to define use cases, interoperability, air-interface requirements, and waveform/library protocols for ISAC (Kaushik et al., 2023, Zhang et al., 9 Apr 2025).

7. Open Challenges and Future Research Directions

Significant challenges remain to fully realize ISAC systems:

• Air Interface Design: Flexible air interface protocols that accommodate the requirements of both high-rate data and high-resolution sensing for diverse use cases (Wang et al., 2022).
• Hardware Integration: Achieving stringent isolation, mitigating self-interference—especially in full-duplex setups, and reconciling divergent ADC/DAC requirements.
• Resource Allocation and Interference Management: Non-orthogonal sharing of spectrum, dynamic cross-layer scheduling, and mitigating multi-user and multi-beam interference (Wild et al., 2023, Xu et al., 2023, Lu et al., 2023).
• Algorithmic Scalability: High-dimensional and nonconvex optimization (for beamforming/RIS/adaptive waveforms) requires efficient AI/ML-driven solutions, especially as dimensionality grows in massive MIMO and distributed scenarios.
• Practical Channel Modeling: Accurately modeling wideband, time-varying, and near-field propagation, including the EM effects of metasurfaces/RIS, remains an open area (Liu et al., 2022).
• Standardization Gaps: Unified global frameworks for protocols and security, cross-layer designs, and interoperability with heterogeneous modalities and legacy equipment.
• Impact of Physical Constraints: At extremely high frequencies (mmWave/THz/optical), challenges such as pathloss, hardware nonidealities, Doppler, and beam-squint must be addressed for robust ISAC deployment (Zhuo et al., 31 May 2024, Zhang et al., 9 Apr 2025).

References to Key Formulas and Figures

• FMCW chirp: $S_T(t) = \exp\{j2\pi [f_c t + 0.5\, a\, t^2]\}$
• OTFS mapping: 2D SFFT and Heisenberg transform in transmitter (see Fig. 4, (Wang et al., 2022))
• RIS-assisted channel: $$y = H_\text{direct}\,x + H_\text{RIS}\, \Phi\, G\, x + n$$ (Liu et al., 2022)
• MI–MMSE relationship: $$\frac{\partial I(\gamma)}{\partial \gamma} = 0.5 \cdot \text{MMSE}(\gamma)$$
• Pareto optimality: maximize weighted sum $\,\rho\, I(X; Y_c|H_c) + (1-\rho)I(H_s; Y_s|X)$
• Sensing–communication performance: explicit $($CMI–SMI$)$ and $($CMI–MSE$)$ regions in (Yu et al., 4 Jul 2024)

Conclusion

ISAC represents a critical paradigm shift for 6G and future networks, unifying wireless communication and sensing into a coherent, resource-efficient, and adaptable framework. The practical realization of ISAC requires multidisciplinary advances across waveform design, distributed and AI-based optimization, integration of emerging technologies such as RIS and digital coding metasurfaces, robust security and privacy models, and standardized protocols. The performance bounds, architectural strategies, and open challenges outlined above chart the research agenda for the next era of wireless systems.