Abstract
Integrated Sensing and Communication (ISAC) is a pivotal technology for 6G networks, enabling simultaneous data transmission and environmental radar sensing. While downlink ISAC has been extensively studied, the uplink scenario presents unique challenges due to mutual interference between communication signals and target reflections. This paper presents a unified framework for uplink MIMO-OFDM ISAC systems using Mutual Information (MI) as a performance metric. By deriving analytical bounds for both communication and sensing MI, the authors develop a waveform optimization strategy that balances the inherent trade-offs between spectral efficiency and sensing accuracy.
Introduction
The evolution toward 6G wireless networks necessitates a paradigm shift from pure communication to a dual-functional architecture known as Integrated Sensing and Communication (ISAC). By sharing hardware and spectral resources, ISAC promises to enhance spectrum efficiency and enable new applications like autonomous driving and industrial automation. Most research to date has focused on the downlink, where a Base Station (BS) transmits to a User Equipment (UE). However, the uplink—where the UE transmits to the BS while the signal also reflects off environmental targets—remains less explored.
In the uplink ISAC context, the signal received at the BS is a superposition of the direct communication signal and the echoes from sensing targets. This creates a complex interference environment: the reflected signals act as interference for data demodulation, while the communication data itself acts as clutter for sensing. This paper addresses the need for a unified mathematical framework to optimize these competing objectives.
System Model and Metric Derivation
The study considers a MIMO-OFDM system where a UE equipped with multiple antennas transmits to a BS. The signal serves a dual purpose: delivering information to the BS and detecting targets in the environment.
Mutual Information as a Unified Metric
The authors propose using Mutual Information (MI) as the primary performance metric. MI is uniquely suited for ISAC because it quantifies the reduction in uncertainty for both tasks—recovering a transmitted bitstream (communication) and estimating the parameters of a target (sensing).
Analytical Bounds
A significant challenge in uplink ISAC is that the sensing channel is often unknown or partially known. To address this, the authors derive:
- Communication MI Bounds: By treating the unknown sensing reflections as structured interference, they establish a lower bound for spectral efficiency.
- Sensing MI Bounds: They incorporate potential demodulation errors from the communication receiver as a noise source for the sensing receiver, providing a realistic estimate of sensing performance.
Waveform Optimization Strategy
To balance the two functions, the authors formulate an optimization problem aimed at maximizing the weighted sum of the communication and sensing MIs. The objective function is defined as a weighted sum where a weighting factor ((\omega_r)) allows network operators to prioritize tasks based on application requirements (e.g., high-speed data vs. high-resolution sensing).
Algorithmic Solution
The resulting optimization problem is non-convex and complex. The authors employ a multi-step approach:
- Decomposition: Using Singular Value Decomposition (SVD) to simplify the MIMO channel matrices into parallel subchannels.
- Power Allocation: Applying Karush-Kuhn-Tucker (KKT) conditions to determine the optimal power distribution across subcarriers and antennas.
Performance Evaluation
The proposed framework was validated using Monte Carlo simulations (4000 iterations) under a frequency-selective MIMO channel model. The simulation utilized the Kronecker model to account for spatial correlation with a maximum coefficient of 0.5.
Key Findings
- Pareto Optimality: The results demonstrate a clear trade-off curve (Pareto frontier). As the weighting factor shifts, the system moves smoothly between an Optimal Communication (OPC) state and an Optimal Sensing (OPS) state.
- Superiority over Baselines: The proposed ISAC waveform significantly outperforms traditional methods such as Equal Allocation (EA) and Random Allocation (RA) in both spectral efficiency and sensing rate.
- SNR Robustness: The derived MI bounds remain tight across a wide range of Signal-to-Noise Ratios (SNR), confirming the robustness of the mathematical model under varying operating conditions.
Conclusion and Future Directions
This research provides a rigorous foundation for uplink ISAC optimization. By utilizing Mutual Information, the authors successfully unified two disparate functions into a single optimization framework. However, several challenges remain for real-world implementation. Future work should address the impact of synchronization errors (timing and frequency offsets), the development of "blind" optimization techniques that require less prior channel knowledge, and the integration of hardware constraints such as Peak-to-Average Power Ratio (PAPR) limitations common in OFDM systems.
Reference: Piao, J., Wei, Z., Yuan, X., et al. "Mutual Information Metrics for Uplink MIMO-OFDM Integrated Sensing and Communication System." arXiv:2310.06382v1, 2023.
Summary updated