⚡ electrical engineering

Posterior-Confidence Driven Beamforming for Energy-Efficient Integrated Sensing and Communication

This paper proposes an energy-efficient MIMO beamforming framework for 6G ISAC systems that minimizes transmit power by employing a tracking-aware, skip-enabled sensing policy triggered only when necessary based on posterior confidence and innovation statistics, while maintaining communication quality and robust target tracking through sector-based beampattern constraints.

Original authors: Nusaibah A. Alshorman, Huseyin Arslan

Published 2026-07-16

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Nusaibah A. Alshorman, Huseyin Arslan

Original paper licensed under CC BY 4.0 (http://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

Imagine the invisible web of signals that keeps our modern world humming—phones connecting, cars navigating, and cities sensing their own heartbeat. This is the realm of Integrated Sensing and Communication (ISAC), a technology where the same radio waves that carry your text messages also act like a giant, invisible flashlight, scanning the environment to spot moving objects. Think of it as a lighthouse that not only warns ships of the shore but also sends them a live video feed. However, keeping this lighthouse beam on 24/7, scanning every second of every day, is incredibly energy-hungry. As we look toward the future of wireless networks (the "6G" era), scientists are worried that this constant, aggressive scanning will drain batteries and power grids, contributing to a massive carbon footprint. The big question isn't just "how can we see better?" but "how can we see smartly?" without wasting a single drop of energy.

This is where the story of a new paper by Nusaibah Alshorman and Huseyin Arslan begins. They tackle the problem of "always-on" sensing, which is like a security guard who never blinks, checking every inch of a hallway even when no one has moved for hours. The authors propose a clever, energy-saving strategy called "Posterior-Confidence Driven Beamforming." Instead of blindly shining their radar light at every target all the time, they equip their system with a "smart brain" that decides exactly when to look and when to take a nap.

Here is how their "smart brain" works, using a tool called an Extended Kalman Filter (EKF). Imagine you are tracking a friend running through a park. If you know exactly where they are, how fast they are going, and where they are heading, you don't need to shout their name every single second to find them. You can predict their path. The authors' system does the same thing. It uses two specific "confidence checks" to decide if it needs to send out a radar pulse:

The Confidence Meter: This checks how sure the system is about where the target is. If the system is 99% sure, it feels safe to skip a scan.
The "Surprise" Detector: This checks if the target is doing something unexpected. If the target suddenly swerves or speeds up in a way the prediction didn't catch, the system wakes up immediately to get a fresh look.

If both checks say, "Everything is calm and predictable," the system skips the sensing step. It turns off the expensive radar probing and saves that energy for other things, like making sure your video call stays crystal clear. But, to be safe, they don't turn the light off completely. They leave a tiny "safety floor" of illumination on, just enough to ensure they don't lose the target entirely if it suddenly changes direction. This is like leaving a nightlight on in a hallway; you don't need the bright overhead light, but you need enough glow to avoid tripping.

The paper presents a mathematical framework that optimizes this "on-off" switching. The researchers ran simulations to test their idea against two other methods: one that never stops scanning (the "always-on" approach) and one that scans at fixed, regular intervals (like a metronome). Their results, shown through computer simulations, suggest that their "skip-aware" method is a winner for energy efficiency. In these simulations, the new system used significantly less transmit power than the always-on version, while still keeping the communication quality (the video call) exactly the same and only causing a tiny, acceptable dip in tracking accuracy.

Crucially, the authors argue against the idea that we must constantly probe to get good results. They show that by trusting the predictions of their "smart brain" and only intervening when necessary, we can save a lot of power. The simulations indicate that this approach allows the system to adapt dynamically: if the target is smooth and predictable, the system sleeps; if the target gets chaotic, the system wakes up. The paper concludes that this method offers a realistic path toward "green" 6G networks, proving that reliable tracking doesn't require a constant, energy-draining spotlight, but rather a smart, confident, and occasionally napping one.

Technical Summary: Posterior-Confidence Driven Beamforming for Energy-Efficient ISAC

1. Problem Statement

The integration of sensing and communication (ISAC) in sixth-generation (6G) systems faces a critical energy efficiency (EE) challenge. While communication requirements are often stable, persistent sensing operations consume significant power, creating a trade-off between tracking reliability and energy consumption. Existing approaches typically assume a continuous "always-on" sensing paradigm, where probing occurs at every time epoch regardless of the current state of target estimation. This assumption leads to unnecessary energy expenditure, particularly in scenarios where target states can be predicted with high confidence. Furthermore, current literature often overlooks the temporal dimension of sensing necessity—specifically, when sensing should be activated or skipped based on instantaneous tracking uncertainty. There is a lack of frameworks that jointly optimize transmit beamforming and adaptive sensing activation decisions at the physical layer under energy constraints.

2. Methodology

The paper proposes a Posterior-Confidence Driven Beamforming framework for Multiple-Input Multiple-Output (MIMO) Dual-Functional Radar-Communication (DFRC) systems. The core methodology involves a "skip-aware" sensing policy that dynamically determines whether to perform active probing or enter a "Sensing Sleep" state based on the tracking status.

System Model

Architecture: A base station (BS) with $N_t$ transmit and $N_r$ receive antennas serves $K$ downlink users while tracking $Q$ point targets.
Signal Structure: The transmit signal comprises communication data streams and radar probing streams. The design minimizes total transmit power while satisfying per-user Signal-to-Interference-plus-Noise Ratio (SINR) constraints and target illumination requirements.
Tracking Model: An Extended Kalman Filter (EKF) is employed to estimate target states (position and velocity). The system operates in discrete epochs where beamforming and sensing decisions are optimized.

Key Mechanisms

Dual-Statistic Sensing Activation: The decision to skip sensing for a specific target is governed by two complementary statistics derived from the EKF:
- Posterior Confidence Metric ( $\eta$ ): A scalar metric derived from the trace of the error covariance matrix, representing the long-term reliability of the state estimate.
- Normalized Innovation Squared (NIS) Gating ( $\Lambda$ ): A $\chi^2$ -based test that measures the consistency between predicted and observed measurements.
- Decision Logic: Sensing is activated ( $u=1$ ) only if the confidence falls below a threshold ( $\eta < \eta_{min}$ ) OR if the NIS exceeds a gating threshold ( $\Lambda > \gamma_{\alpha,d}$ ), indicating inconsistency. Otherwise, sensing is skipped ( $u=0$ ).
Safety Illumination Floor: To prevent track divergence during skipped epochs, the framework enforces a non-zero "safety illumination floor." Even when active probing is skipped, a minimum fraction ( $\rho$ ) of the required sensing power is maintained. This ensures continued state observability and prevents the EKF from diverging due to a complete lack of measurements.
Optimization Formulation: The problem is formulated as a convex Semidefinite Program (SDP) that minimizes the trace of the transmit covariance matrix (total power).
- Constraints: Per-user SINR constraints and sector-based beampattern constraints.
- Adaptive Constraint: The sensing constraint is modified based on the skip decision $u$ . If $u=1$ , full illumination $\alpha_q$ is required; if $u=0$ , the constraint is relaxed to $\rho \alpha_q$ .

3. Key Contributions

Energy-Aware Beamforming Framework: The authors introduce a framework that integrates temporal sensing activation decisions directly into the transmit covariance matrix design, moving beyond static or purely spatial optimization.
Posterior-Aware Activation Mechanism: A novel sensing policy is developed where activation is governed by EKF-derived estimation confidence and NIS gating, allowing the system to skip redundant probing when tracking is reliable.
Joint Optimization Problem: A convex optimization problem is formulated to minimize transmit power while ensuring communication QoS and target illumination, explicitly accounting for the skip-aware logic.
Safety Illumination Constraint: The introduction of a safety illumination factor ( $\rho$ ) guarantees minimum target observability during skipped intervals, balancing energy savings with tracking stability and preventing track loss.

4. Numerical Results and Discussion

The proposed Adaptive Skip-aware Probing (ASP) scheme was evaluated against two baselines:

Full-Probing (FP): Continuous sensing at every epoch.
Periodic Low-Power Probing (PLP): Sensing enforced at fixed periodic intervals without adaptive logic.

Findings:

Energy Efficiency: The ASP scheme achieved the lowest normalized transmit power across the entire Signal-to-Noise Ratio (SNR) range, significantly outperforming both FP and PLP. The power savings were most pronounced at higher SNR levels where channel conditions allow for lower transmit power to meet SINR targets.
Communication Performance: All schemes (ASP, FP, PLP) maintained identical average sum-rates. This confirms that the energy-saving "skip" strategy does not degrade communication Quality of Service (QoS), as the optimization strictly enforces per-user SINR constraints.
Tracking Performance:
- ASP exhibited a slight degradation in Root Mean Square Error (RMSE) compared to FP, which is expected due to reduced measurement frequency. However, this degradation was minimal and acceptable.
- ASP significantly outperformed PLP in tracking stability. The fixed periodic nature of PLP led to higher track-loss probabilities, particularly when estimation uncertainty increased during non-probing intervals.
- The safety illumination factor ( $\rho$ ) was found to be critical. A value of $\rho = 0.3$ provided an optimal trade-off, significantly reducing track-loss probability compared to lower values while avoiding the excessive power consumption associated with higher values.
Parameter Sensitivity: The framework demonstrated robustness across a range of NIS thresholds ( $\alpha$ ) and confidence thresholds ( $\eta_{min}$ ), indicating that performance gains are driven by the underlying confidence-aware mechanism rather than fine-tuned parameter selection.

5. Significance and Claims

The paper claims that the proposed framework successfully shifts ISAC operation from a continuous, always-on paradigm to an adaptive, information-driven process. By leveraging the predictive capability of tracking filters, the system can identify temporal redundancies in sensing requirements.

The authors assert that this approach enables substantial energy savings without compromising communication performance or causing excessive tracking degradation. The inclusion of the safety illumination floor is highlighted as a crucial mechanism that prevents the "all-or-nothing" risk of skipping measurements, thereby maintaining system reliability. The work positions itself as a step toward "green" ISAC implementations in future 6G networks, addressing the need to operate under strict energy constraints while meeting the dual demands of sensing and communication. The paper concludes that reliable tracking can be maintained without continuous sensing, provided that observability is preserved through a minimum illumination floor.