Backscatter-based Integrated Sensing & Communication

• B-ISAC is a dual-function framework that exploits passive backscatter modulation to enable ultra-low-power sensing and communication in large-scale IoT networks.
• It leverages cascaded BS–tag–RX channels and advanced signal processing to mitigate double fading and geometric sensitivity, enhancing both link robustness and sensing accuracy.
• B-ISAC integrates techniques like MAS, MIMO, and meta-backscatter, employing joint optimization of power allocation and antenna design to balance communication rates with sensing precision.

Backscatter-based Integrated Sensing and Communication (B-ISAC) merges passive backscatter modulation with co-designed wireless sensing and communication functions. Leveraging passive tags as information-bearing scatterers in a cascaded transceiver–tag–receiver link, B-ISAC architectures realize ultra-low-power dual-functionality critical for large-scale Internet of Things (IoT) and next-generation wireless systems. The core challenge lies in the severe double fading and geometric sensitivity intrinsic to the backscatter channel, motivating recent advancements in system design, signal processing, resource optimization, and hardware architectures.

1. Fundamental Principles and System Architectures

B-ISAC exploits the physical mechanism whereby a passive tag, typically devoid of active RF chains, modulates incident electromagnetic waves from a base station (BS) or access point (AP) via its impedance-controlled reflection coefficient β(t), embedding both environmental state and digital data into the backscattered field. The received signal at the sensing/communication node is defined by a cascaded channel, combining BS-to-tag (forward, h_f) and tag-to-RX (backward, h_b) propagation:

$$y(t) = \sqrt{P_{\rm tx}} \, h_b^T \beta(t) h_f s\big(t-\tau\big) + n(t)$$

Here, $P_{\rm tx}$ is the transmit power, $s(t)$ is the dual-purpose ISAC waveform, and $n(t)$ is additive noise. The end-to-end channel exhibits harsh double fading—$$|h_c| = |h_f||h_b| = G_t G_{\rm tag} G_r (\lambda/4\pi)^2/(d_f d_b)$$—imposing substantial design hurdles for both reliable data transmission and high-fidelity sensing (Zhang et al., 27 Jan 2026).

Several B-ISAC architectures are established:

• Classical B-ISAC (static antennas): Fixed BS/AP and tag locations with monostatic (co-located TX/RX) or bistatic (separate TX/RX) configurations. Signal processing encompasses coherent demodulation, time-delay, Doppler, and angle estimation (Thomä et al., 2022, Abrudan et al., 16 Sep 2025).
• Movable-Antenna Systems (MAS): TX/RX arrays mounted on actuators capable of sub-wavelength repositioning, actively adapting link geometry in real-time to avoid fades, align angular axes, or maximize effective aperture (Zhang et al., 27 Jan 2026).
• MIMO and Ambient/Distributed Backscatter: Multi-antenna arrays at BS, RX, or both; distributed passive tags functioning jointly as sensors and relay scatterers, often operating in urban or industrial large-format deployments (Zhao et al., 2024, Zhang et al., 1 Mar 2025).

Block diagrams integrate the central DSP and RF front ends, power splitting, energy harvesting, load modulation, motion control (for MAS), and closed-loop channel feedback (Zhang et al., 27 Jan 2026).

2. Channel and Signal Models

B-ISAC models are characterized by a concatenated channel matrix $H_c = h_b h_f^T$ for point tags, or more intricate angular/temporal integrals for extended targets (Thomä et al., 2022). Path loss is the product of cascaded Friis links:

$$L_b(R_t, R_r) = \left(\frac{4\pi R_t}{\lambda}\right)^2 \left(\frac{4\pi R_r}{\lambda}\right)^2$$

For distributed, multi-link systems, each TX–target–RX trio yields an independent bistatic reflection measurement, improving sensing performance via spatial diversity (Thomä et al., 2022):

$$P_r = P_t G_t G_r \frac{\lambda^2 \sigma_{bi}}{(4\pi)^3 R_t^2 R_r^2}$$

Backscatter modulation is realized through time-varying reflection coefficients $\beta(t) \in \{\beta_0, \beta_1, \ldots\}$ or, in meta-backscatter realizations, via reflection coefficient $\Gamma(\theta)$ parametrized by sensed environmental states (e.g., gas concentration, temperature), with frequency-selective resonance (Liu et al., 2024).

In wideband OFDM B-ISAC, full time- and frequency-domain models account for analog RF chain responses, nonlinear tag reflections, frequency/clock offsets, and multi-path propagation (Abrudan et al., 16 Sep 2025).

3. Joint Optimization and Trade-offs

B-ISAC is intrinsically governed by joint optimization of communication and sensing objectives under tight power, link budget, and geometric constraints. Weighted-sum and Pareto boundary formulations define feasible regions for communication rate $R_c$ and sensing SNR $SNR_s$, or more general information-theoretic metrics (mutual information, mean-square error, Cramér–Rao bounds) (Tian et al., 2024, Zhang et al., 12 Jul 2025):

$$\max_{p_{\rm tx}, p_{\rm rx}, \rho} \; w_c \log_2[1 + \gamma_c] + w_s SNR_s, \;\;\; \textrm{s.t. } \|p_{\rm tx}\|\leq\Delta_{\rm tx},\, \|p_{\rm rx}\|\leq\Delta_{\rm rx}$$

Alternating optimization, block coordinate descent, and semidefinite relaxation (SDR) are employed for beamformer design, power allocation, reflection coefficient tuning, and antenna motion planning (Zhang et al., 27 Jan 2026, Zargari et al., 2024, Zhao et al., 2024, Zhao et al., 2024). Resource element (RE) and power allocation in OFDM B-ISAC is addressed via successive convex approximation (SCA) and water-filling-type closed-form updates (Zhang et al., 12 Jul 2025).

A pervasive theme is the trade-off surface between sensing accuracy—often quantified via CRB for range and angle estimation, or mutual information—and communication throughput. Higher sensing accuracy typically incurs a rate penalty and vice versa, with the achievable boundary dependent on waveform design, antenna geometry, and device placement (Tian et al., 2024). For meta-backscatter, the discernibility of the physical parameter (e.g., humidity) is strictly inversely related to achievable bit-rate (Liu et al., 2024).

4. Algorithmic and Hardware Implementations

Feedback and Motion-Aided Adaptation

• Geometry-Aware MAS: Online measurement of backscatter strength, gradient-ascent channel reconfiguration, and sub-wavelength mechanical displacement yield highly robust, geometry-adaptive links (Zhang et al., 27 Jan 2026).
• Distributed Sensing: Spatial diversity in AP antennas or distributed massive tags enables robust RSS-based localization and target detection, often using maximum-likelihood trilateration (Zhang et al., 1 Mar 2025).

Beamforming and Resource Allocation

• Joint Transmission/Reception Beamforming: Beamformers for communication, tag activation, and deterministic sensing streams are co-designed under total power constraints. Quadratic transforms, SCA, and SDR convert intractable non-convex objectives into tractable programs (Zhao et al., 2024, Zhao et al., 2024).
• Meta-Backscatter and Sensor Optimization: Sensing–communication trade-off is directly linked to metasurface resonance depth/width and transmit power allocation via water-filling; gradient methods update metasurface geometry to maintain Pareto-optimal design (Liu et al., 2024).

Synchronization and Transceiver Imperfections

Digital compensation for clock/frequency offsets, channel estimation, and signal chain impairments is critical, especially in OFDM-based wideband B-ISAC (Abrudan et al., 16 Sep 2025). Hardware experiments demonstrate practical feasibility using RFID/OFDM platforms, USRPs, and metasurface prototypes (Abrudan et al., 16 Sep 2025, Zhang et al., 1 Mar 2025, Liu et al., 2024).

5. Performance Analysis and Metrics

Quantitative metrics for B-ISAC include:

• Communication Rate ($R_c$): Bits/s/Hz, typically Shannon-style after SINR estimation (Zargari et al., 2024).
• Sensing Accuracy: Cramér–Rao lower bound (CRB) on transmission delay, DoA, or range (Tian et al., 2024, Abrudan et al., 16 Sep 2025).
• Detection Probability ($P_D$): For tag presence or target detection (Neyman–Pearson, CFAR) (Zhao et al., 2024).
• Received SNR ($\gamma_c$, $SNR_s$): For communication and estimation subtasks.
• Weighted Throughput Capacity (WTC): Weighted sum of radar and backscatter communication rates for joint computation offloading (Xu et al., 2022).
• Symbol/Bit Error Rate (SER/BER): Closed-form expressions under Rayleigh/Rician fading for practical error-rate analysis (Zhang et al., 1 Mar 2025, Zhang et al., 1 Mar 2025).

Illustrative system-level results:

System	Max $R_c$ (bps/Hz)	$SNR_s$ (dB)	$\Delta P_{tx}$ (dBm)	Link Robustness
Active ISAC	4.5	12	+5	Medium
Static B-ISAC	1.8	6	0	Low
MAS-B-ISAC	3.0	10	0	High

MAS yields a $\sim$67% rate gain and 4 dB sensing SNR improvement over static B-ISAC in canonical line-of-sight links (Zhang et al., 27 Jan 2026).

6. Applications, Use Cases, and Deployment Guidelines

B-ISAC supports a sweeping range of practical scenarios:

• IoT, 6G, and Massive-Scale Sensor Networks: Battery-free tags for logistics, asset tracking, environmental monitoring, and smart factory deployments (Abrudan et al., 16 Sep 2025, Zhang et al., 1 Mar 2025).
• Vehicular and Urban Sensing: MAS-based B-ISAC for V2X, with sub-wavelength lateral/elevation motion compensation for vehicular tags (Zhang et al., 27 Jan 2026).
• Industrial Automation and Warehousing: Mobile robot tracking via MAS and distributed AP arrays, with rapid antenna adaptation to dynamic trajectories (Zhang et al., 27 Jan 2026).
• Meta-Backscatter for Physical Sensing: Sensing environmental parameters (humidity, temperature) via resonance-based metasurface reflection, optimizing sensing accuracy and communication capacity (Liu et al., 2024).
• Health Monitoring, Human-Centric Sensing: Wearable tags and tomographic imaging via backscatter RSS for motion and activity detection (Zhang et al., 1 Mar 2025).
• RF-Chain-Free Computation Offloading: IRS-enabled B-ISAC for joint data collection and radar-based environmental sensing without user RF transmitters (Xu et al., 2022).

Deployment is constrained by channel coherence times, mechanical actuation limits (for MAS), energy harvesting vs. reflection trade-off, antenna placement strategy, and regulatory transmit power (Zhang et al., 27 Jan 2026, Zhang et al., 1 Mar 2025, Zhang et al., 1 Mar 2025).

7. Open Challenges and Future Directions

• Extreme-Scale and Cell-Free Architectures: Coordinated backscatter sensing and multi-tag/multi-AP fusion under limited backhaul and synchronization budgets (Liu et al., 2024, Thomä et al., 2022).
• Waveform and Hardware Codeign: Joint design of hardware (meta-materials, IRSs), waveform (OFDM, chirp, OTFS), and MAC for optimal comm-sense trade-offs (Zhao et al., 2024, Liu et al., 2024, Zhang et al., 12 Jul 2025).
• Super-Resolution and Multipath-Robust Sensing: Bistatic, multi-band, and multi-node super-resolution algorithms to surpass CRB limits in dense multipath environments (Abrudan et al., 16 Sep 2025).
• Resource-Efficient SDR and Online Solutions: Scalable, real-time optimization for joint beamforming, power allocation, and actuator control (Zargari et al., 2024, Zhao et al., 2024).
• Net-Zero and Energy-Autonomous Systems: Maximizing battery-free, maintenance-free deployments by balancing harvesting, backscatter, and sensing objectives (Zhang et al., 1 Mar 2025).

B-ISAC provides a unified framework for integrating passive wireless sensing and low-power communication, demarcating a distinctive pathway toward sustainable, scalable, and context-aware future networks (Zhang et al., 27 Jan 2026, Zhao et al., 2024, Tian et al., 2024, Abrudan et al., 16 Sep 2025, Zhang et al., 1 Mar 2025).