⚡ electrical engineering

Joint RIS Phase and Power Allocation for a Hardware Impaired OTFS Waveform in Dense Vehicular Integrated Sensing and Communication

This paper proposes RAHI-OTFS, a joint optimization framework for Reconfigurable Intelligent Surface phase configuration and power allocation that explicitly incorporates hardware impairments to significantly enhance both sensing accuracy and communication reliability in dense vehicular networks.

Original authors: B P Vidyashree, Punith Kumar MB, Purushottama Lingadevaru

Published 2026-10-14

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: B P Vidyashree, Punith Kumar MB, Purushottama Lingadevaru

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the bustling arteries of a modern city, the air is thick with invisible signals. Cars talk to traffic lights, to each other, and to the roadside infrastructure, all while the same radio equipment is increasingly expected to act as a set of eyes, constantly measuring the speed and distance of every vehicle to prevent accidents. This dual role, where a single device both communicates data and senses the physical world, is known as integrated sensing and communication. It is a cornerstone of the next generation of wireless networks. However, designing these systems is like trying to tune a radio while driving through a canyon; the signals bounce off buildings, the vehicles move at high speeds, and the radio waves themselves are distorted by the very hardware that sends and receives them. For years, engineers have tried to solve the problem of signal blockage by using "smart mirrors" on buildings—surfaces that can be programmed to reflect signals around corners to reach vehicles that have no direct line of sight. At the same time, they have developed special waveforms, or signal patterns, that are robust enough to handle the rapid changes caused by fast-moving cars.

The challenge that has remained largely unaddressed is that most of these advanced designs assume the hardware is perfect. They imagine a world where the radio components are flawless, where the signals are never distorted by the electronics inside the transmitter or receiver. In reality, the inexpensive radio units that cities would actually deploy are imperfect. They suffer from phase noise, where the timing of the signal drifts slightly; they use digital converters that chop the signal into steps rather than a smooth flow; and their power amplifiers distort the signal when it gets too loud. This paper, titled "Joint RIS Phase and Power Allocation for a Hardware Impaired OTFS Waveform in Dense Vehicular Integrated Sensing and Communication," asks a critical question: does the strategy of using these smart mirrors and special signals still work when we stop pretending the hardware is perfect? The researchers, working with a team from institutions in India, set out to design a system that treats the imperfections of the hardware not as a separate problem to be fixed later, but as a fundamental part of the design from the very beginning.

The team focused on a specific type of signal called OTFS, which organizes information in a grid based on time delays and speed shifts, making it naturally suited for fast-moving vehicles. They combined this with a Reconfigurable Intelligent Surface, a large panel of passive elements that can be tuned to steer radio waves. Instead of optimizing the surface to work with a perfect signal and then hoping it survives the real-world hardware, they built a mathematical model where the signal passes through the actual, flawed components of the transmitter and receiver first. They then adjusted the surface and the power distribution of the signal to find the best possible performance given those flaws. This approach, which they named RAHI-OTFS, required them to calculate exactly how the hardware errors would degrade the system's ability to measure distance and speed, and then to find a configuration that minimized that damage.

The results of their simulations, set in a dense urban environment with multiple cars and targets, revealed a stark reality. When the researchers compared their optimized system against a version where the smart mirror was set to random, uncoordinated positions, the improvement was dramatic. The optimized system reduced the error in measuring the distance to a target from nearly five centimeters down to less than one centimeter. In terms of communication, the error rate for the data being sent dropped significantly, moving from a high 28 percent down to 16 percent under the same hardware conditions. These numbers show that carefully tuning the surface and the signal power can recover a substantial amount of performance even when the hardware is imperfect. However, the study also highlighted a hard limit. When they compared their best possible design against a theoretical system with perfect hardware, they found that the real-world imperfections still caused the error rate to be roughly one hundred thousand times higher than it would be in an ideal world.

This finding suggests that while optimizing the smart surface is incredibly valuable, it cannot completely erase the cost of using cheap, imperfect electronics. The researchers demonstrated that the two problems—the design of the smart surface and the reality of the hardware flaws—cannot be solved separately. If engineers design a surface assuming perfect hardware and then install it on a real, imperfect device, the system will perform far worse than predicted. By treating the hardware flaws as an integral part of the design equation, the team showed that it is possible to build a system that is much more robust, bringing the performance of a real-world deployment much closer to the theoretical ideal. The study confirms that for the future of smart cities, where safety depends on precise sensing and reliable communication, we must design our networks with the understanding that the hardware will never be perfect, and our solutions must account for that reality from the start.

Technical Summary: Joint RIS Phase and Power Allocation for a Hardware Impaired OTFS Waveform in Dense Vehicular Integrated Sensing and Communication

Problem Statement
The paper addresses a critical gap in the design of Integrated Sensing and Communication (ISAC) systems for dense vehicular networks. While existing literature extensively covers the joint optimization of Reconfigurable Intelligent Surface (RIS) phase configurations and waveforms (specifically Orthogonal Time Frequency Space, or OTFS) to enhance sensing accuracy and communication rates, these designs typically assume an idealized, noiseless, and linear radio front-end. Conversely, separate studies on hardware impairments (such as oscillator phase noise, finite-resolution Analog-to-Digital Converters (ADCs), and Power Amplifier (PA) nonlinearity) quantify their degradation of link reliability and estimation accuracy but generally ignore the presence of an RIS.

The authors argue that a surface configuration optimized under ideal assumptions may not remain optimal when deployed with the inexpensive, imperfect hardware typical of real-world city-scale deployments. This paper proposes a unified design framework that treats RIS optimization and hardware realism as a single, coupled problem rather than two isolated design tasks.

Methodology
The paper introduces RAHI-OTFS (Reconfigurable Intelligent Surface Aware, Hardware Impaired OTFS), a waveform design that integrates hardware impairments directly into the signal model and optimization loop.

System Model:
- Geometry: A roadside unit (RSU) communicates with multiple vehicles and senses targets via a cascaded channel through a passive RIS. Direct paths are assumed blocked, making the RIS the dominant propagation path.
- Waveform: OTFS is used to map information symbols onto a delay-Doppler grid, offering robustness against high mobility.
- Hardware Impairments: Three specific non-idealities are modeled directly in the transmit and receive chains:
  - PA Nonlinearity: Modeled using the Rapp characteristic to account for saturation and compression.
  - ADC Quantization: Modeled via the Additive Quantization Noise Model (AQNM) to represent finite resolution effects.
  - Oscillator Phase Noise: Modeled as a discrete-time Wiener process.
- Effective Noise: These impairments are combined to define an "effective noise power" ( $N_{0,eﬀ}$ ), which includes thermal noise plus hardware-induced noise floors.
Optimization Problem:
- Objective: A joint optimization of the RIS phase vector ( $\theta$ ) and the delay-Doppler power allocation matrix ( $W$ ). The objective function maximizes a weighted sum of the communication sum-rate and the inverse of the joint Cramér-Rao Bound (CRB) for target delay and Doppler estimation.
- Constraints: Total power budget, non-negative power allocation, unit modulus for RIS phases, and Peak-to-Average Power Ratio (PAPR) limits enforced via PA input back-off.
- Fisher Information: A full derivation of the Fisher Information Matrix (FIM) is provided for the impaired link. The authors demonstrate that under a centered grid, the FIM is diagonal, allowing for a closed-form CRB calculation that explicitly includes the hardware noise floor.
Algorithm:
- An Alternating Optimization (AO) algorithm is proposed. It iterates between:
  - Updating the power allocation ( $W$ ) using a closed-form solution that splits power between a uniform communication floor and a sensing boost.
  - Updating the RIS phase vector ( $\theta$ ) using Simultaneous Perturbation Stochastic Approximation (SPSA) to handle the non-convex unit modulus constraint.
- Discrete phase quantization is applied only after the continuous phase optimization converges to avoid destabilizing the gradient ascent.

Key Contributions

Unified Design Framework: The paper formulates RAHI-OTFS, optimizing RIS phases and power allocation simultaneously against a hardware-aware CRB and communication rate, rather than optimizing for one metric under ideal assumptions.
Direct Hardware Integration: Unlike prior works that treat impairments as post-analysis, this work embeds PA nonlinearity, ADC quantization, and phase noise directly into the transmit/receive signal models and the objective function.
Theoretical Derivation: A complete derivation of the Fisher Information and CRB for the delay-Doppler pair under combined RIS gain and hardware noise floors is presented.
Low-Complexity Algorithm: A practical alternating optimization algorithm is developed to solve the non-convex joint problem, validated for dense roadside scenarios.

Results
Simulations were conducted for a representative dense roadside scenario (28 GHz carrier, 64-element RIS, 4 communicating vehicles, 3 sensing targets) comparing three configurations: the proposed RAHI-OTFS, a random RIS phase baseline (with same hardware impairments), and an idealized hardware baseline (optimized RIS, no impairments).

Performance Gains: Compared to the random RIS baseline, the optimized RAHI-OTFS configuration reduced the range Root Mean Square Error (RMSE) from 4.9 cm to 0.9 cm and improved the Bit Error Rate (BER) from 0.28 to 0.16 under identical hardware impairments.
Hardware Cost: Even with the optimized surface, the hardware-impaired system exhibited a BER roughly five orders of magnitude higher than the idealized hardware baseline (10⁻⁶ vs. 10⁻¹), quantifying the significant cost of real-world component limitations.
Trade-offs: The study showed that surface optimization improves both communication sum-rate and sensing accuracy simultaneously, rather than requiring a trade-off between the two.
Sensitivity: Results indicated that performance is not highly sensitive to the specific power split between sensing and communication or the exact weighting of the objective function within reasonable ranges.
Hardware Limits: Increasing ADC resolution beyond 6–7 bits yielded diminishing returns, as oscillator phase noise became the dominant impairment floor.

Significance and Claims
The paper claims that the results demonstrate that surface optimization and hardware realism must be treated as a single design problem. The authors assert that a configuration optimal for an ideal transceiver is not necessarily sensible for a real-world deployment using inexpensive hardware.

The study concludes that while RIS optimization provides substantial gains in dense vehicular environments, the residual performance gap caused by hardware impairments is severe. Therefore, future ISAC designs cannot safely ignore the interaction between the physical layer waveform, the RIS configuration, and the specific non-idealities of the radio front-end. The paper explicitly notes limitations, including simplified channel models (free-space path loss without multipath), the absence of phase tracking/compensation algorithms, and the use of simplified impairment models (e.g., Gaussian AQNM), positioning the work as a foundational step toward more realistic, hardware-aware ISAC design.

Technical Summary: Joint RIS Phase and Power Allocation for a Hardware Impaired OTFS Waveform in Dense Vehicular Integrated Sensing and Communication

More like this