3. Sharing the Surface Between Communication and Sensing

System Model in Words

The BS transmits a known waveform carrying data. A user decodes that data; the BS or a separate sensing receiver processes target echoes. A RIS changes the effective channel for both tasks.

Specify the receiver first. A BS receiver using a reciprocal RIS path has a four-hop budget. A receiver mounted on the RIS does not have that same return route. Neither case should inherit the other’s path loss by assumption.

References for this section3

• M. A. Haider and Y. D. Zhang, "RIS-Aided Integrated Sensing and Communication: A Mini-Review", 2023.
• M. Rihan et al., "Passive vs. Active RIS for ISAC", IEEE Network, 2023.
• M. Rihan et al., "Active RIS-Aided ISAC Systems: Beamforming Design and Performance Analysis", IEEE Transactions on Communications, 2024.

Communication Path

Let $h_d$ denote the complex direct channel and $h_r$ the complex RIS-assisted channel. A scalar user model is:

$$y=(h_d+h_r)x+n,\qquad \gamma_c=\frac{P_t|h_d+h_r|^2}{I+N}.$$

• $x$: transmitted symbol.
• $y$: received sample.
• $n$: receiver noise sample.
• $P_t$: transmit power.
• $I$: interference power.
• $N$: noise power.
• $\gamma_c$: signal-to-interference-plus-noise ratio.

Channels add as complex amplitudes: a reflected path can help or hurt depending on its phase.

Ideal spectral efficiency is $R=\log_2(1+\gamma_c)$ in bit/s/Hz. Actual throughput also depends on bandwidth, coding, scheduling and overhead.

References for this section3

• M. A. Haider and Y. D. Zhang, "RIS-Aided Integrated Sensing and Communication: A Mini-Review", 2023.
• M. Rihan et al., "Passive vs. Active RIS for ISAC", IEEE Network, 2023.
• M. Rihan et al., "Active RIS-Aided ISAC Systems: Beamforming Design and Performance Analysis", IEEE Transactions on Communications, 2024.

Sensing Path

• Direct echo: BS → target → BS.
• Reciprocal RIS echo: BS → RIS → target → RIS → BS.
• Mixed echo: only the outgoing or return route uses the RIS.

Separate these echoes by their expected delay and spatial response when possible. Unresolved paths can interfere. A communication SINR target does not establish a sensing detection probability.

References for this section3

• M. A. Haider and Y. D. Zhang, "RIS-Aided Integrated Sensing and Communication: A Mini-Review", 2023.
• M. Rihan et al., "Passive vs. Active RIS for ISAC", IEEE Network, 2023.
• M. Rihan et al., "Active RIS-Aided ISAC Systems: Beamforming Design and Performance Analysis", IEEE Transactions on Communications, 2024.

Surface Partitioning

A simple baseline dedicates a fraction $\alpha$ of the physical aperture to communication and $1-\alpha$ to reciprocal sensing. Assume fixed illumination, coherent steering, independent subarrays and no useful cross-beams.

$$\gamma_c=\gamma_{c,\mathrm{full}}\alpha^2,\qquad \frac{P_s}{P_{s,\mathrm{full}}}=(1-\alpha)^4.$$

The squared communication factor comes from coherent one-way aperture gain. The fourth-power sensing factor comes from using the sensing subarray on both passes. At a 50/50 split the factors are 1/4 and 1/16. The lab fixes full-aperture user SNR at 100 (20 dB); it shows a baseline, not the best possible trade-off.

How much surface can each task afford?

1. Formula and symbols

$$R=\log_2(1+\gamma_0\alpha^2)$$

$$P_s/P_{s,\rm full}=(1-\alpha)^4$$

• $\alpha$: surface fraction for communication
• $1-\alpha$: sensing fraction
• $\gamma _{0}$: full-surface linear communication SNR
• $R$: bit/s/Hz
• $P_s$: echo power

The fourth power follows two traversals of the sensing sub-aperture.

Area allocated to communication

50%

2. Work with the current inputs

$$R=\log_2(1+100\alpha^2)=\log_2(1+100\times0.50^2)=4.70$$

$$P_{s}/P_{s,\mathrm{full}}=(1-\alpha)^4=(1-0.50)^4=0.0625$$

3. Result

Communication spectral efficiency4.70 bit/s/Hz

Echo / full sensing aperture6.25%

Sensing echo loss-12.0 dB

References for this section3

• M. A. Haider and Y. D. Zhang, "RIS-Aided Integrated Sensing and Communication: A Mini-Review", 2023.
• M. Rihan et al., "Passive vs. Active RIS for ISAC", IEEE Network, 2023.
• M. Rihan et al., "Active RIS-Aided ISAC Systems: Beamforming Design and Performance Analysis", IEEE Transactions on Communications, 2024.

Joint Optimization

A joint design can instead choose all element phases and BS beams together. The variables are coupled: improving one beam can move energy away from another.

• Objective: for example, maximize echo SNR subject to minimum user SINR.
• Constraints: total BS power, surface losses, available phase states and, for active panels, amplifier power.
• Baseline: compare against partitioning using equal aperture, power, scan time and receiver assumptions.

A feasible algorithm is not automatically globally optimal. Report both the achieved sensing metric and communication metric, including their assumptions.

References for this section3

• M. A. Haider and Y. D. Zhang, "RIS-Aided Integrated Sensing and Communication: A Mini-Review", 2023.
• M. Rihan et al., "Passive vs. Active RIS for ISAC", IEEE Network, 2023.
• M. Rihan et al., "Active RIS-Aided ISAC Systems: Beamforming Design and Performance Analysis", IEEE Transactions on Communications, 2024.

Takeaway

Use partitioning to understand the cost of sharing, then evaluate whether a joint design earns its additional complexity. Next, examine geometry and phase control, because a stronger echo is only one part of localisation.

References for this section3

• M. A. Haider and Y. D. Zhang, "RIS-Aided Integrated Sensing and Communication: A Mini-Review", 2023.
• M. Rihan et al., "Passive vs. Active RIS for ISAC", IEEE Network, 2023.
• M. Rihan et al., "Active RIS-Aided ISAC Systems: Beamforming Design and Performance Analysis", IEEE Transactions on Communications, 2024.