🔢 mathematics

Power consumption Reduction in ELAA-Assisted ISAC Systems

This paper proposes an energy-efficient subarray activation framework for extremely large antenna array-assisted integrated sensing and communication systems that minimizes total power consumption while satisfying quality-of-service constraints for both sensing and communication through a successive convex approximation-based optimization algorithm.

Original authors: Xiaomin Cao, Mohammadali Mohammadi, Hien Quoc Ngo, Michail Matthaiou

Published 2026-01-30

📖 3 min read🧠 Deep dive

CC BY 4.0

Original authors: Xiaomin Cao, Mohammadali Mohammadi, Hien Quoc Ngo, Michail Matthaiou

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

The Big Picture: The "Super-Flashlight" Problem

Imagine a next-generation cell tower (the Base Station) that acts like a giant, super-powered flashlight. This flashlight has two jobs:

Talk to people: It sends messages to your phone (Communication).
See things: It shines a beam to detect cars, drones, or obstacles nearby (Sensing).

To do this perfectly, especially for things that are close by, the tower needs a massive array of tiny light bulbs (antennas) packed tightly together. This is called an Extremely Large Antenna Array (ELAA).

The Problem: If you turn on every single light bulb at full power, the tower becomes incredibly bright and precise, but it also becomes a power-hungry monster. It would consume so much electricity that it would be expensive to run and generate too much heat. It's like trying to light up a whole stadium just to read a book on a bench.

The Solution: The "Smart Switchboard"

The authors of this paper propose a clever way to save energy without losing the flashlight's ability to see or talk. Instead of turning on every light bulb, they suggest using a smart switchboard that only turns on the specific group of bulbs needed for the job at hand.

Think of the massive array of antennas not as one giant block, but as a grid of smaller neighborhoods (called "subarrays").

The Old Way: Turn on the entire city grid to send one text message.
The New Way: The system looks at who needs a message and where the object to be sensed is. It then figures out exactly which "neighborhoods" of lights are needed to cover those specific spots and turns off the rest.

How They Did It (The Recipe)

The researchers created a mathematical "recipe" (an algorithm) to solve a very tricky puzzle:

The Goal: Use the least amount of electricity possible.
The Rules:
- The phone users must still get a clear signal (no dropped calls).
- The sensing beam must be strong enough to detect targets.
- You can only turn whole "neighborhoods" of lights on or off, not individual bulbs.

Because figuring out the perfect combination of neighborhoods is like trying to find the best route through a maze with billions of paths, the authors invented a step-by-step guessing game (called a "Successive Convex Approximation" algorithm).

The Metaphor: Imagine you are trying to fit a suitcase into a car trunk. You don't just throw everything in; you try a few items, see if it fits, adjust, and try again. The algorithm does this mathematically, making small adjustments to its "guess" of which subarrays to turn on until it finds the most energy-efficient combination that still fits the "trunk" (meets the performance requirements).

What They Found (The Results)

The team ran computer simulations to test their idea. Here is what happened:

Energy Savings: Their method saved a massive amount of power—up to 50% less than turning on the entire array. That's like cutting your electricity bill in half.
Speed: The "guessing game" algorithm was very fast. It figured out the best solution in just a few seconds (about 5 to 7 rounds of calculation).
Flexibility: The system worked well whether there were many users far away or a few users very close to the tower. It also found that sometimes, the same group of lights could handle both talking to phones and sensing targets,

Technical Summary: Power Consumption Reduction in ELAA-Assisted ISAC Systems

Problem Statement
Extremely Large Antenna Arrays (ELAAs) are pivotal for Integrated Sensing and Communication (ISAC) in sixth-generation (6G) networks, particularly for enabling high-resolution near-field sensing through spherical wavefronts. However, deploying ELAAs with conventional fully-digital architectures presents significant challenges. Activating all antenna elements simultaneously requires a dedicated radio frequency (RF) chain for each element, leading to prohibitive hardware complexity and excessive power consumption. While hybrid beamforming and antenna selection techniques have been proposed to mitigate hardware costs, they often struggle with near-field effects, such as frequency-varying phase shifts and significant channel gain variations across the array. Furthermore, existing methods may not fully exploit the spatial non-stationarity of ELAAs, where users typically interact only with a specific "visibility region" (VR) of the array.

Methodology
To address these challenges, the authors propose an energy-efficient subarray activation framework for a monostatic ELAA-assisted ISAC system. The system simultaneously serves near-field users (NFUEs), far-field users (FFUEs), and illuminates a target for sensing in the downlink.

The core of the methodology involves formulating a joint optimization problem (P1) aimed at minimizing the total power consumption of the ELAA. The decision variables are binary indicators representing whether specific subarrays are activated for serving NFUEs, FFUEs, or sensing. The optimization is subject to:

Quality-of-Service (QoS) Constraints: Minimum Signal-to-Interference-plus-Noise Ratio (SINR) requirements for both NFUEs and FFUEs, and a minimum beampattern gain threshold for the sensing target.
Power Constraints: Limits on total transmit power and per-subarray transmit power.
Hardware Constraints: Binary activation logic ensuring that a subarray used for communication is also available for sensing if needed.

The problem is inherently non-convex and combinatorial due to the binary variables and the non-convex nature of the SINR and beampattern constraints. To solve this, the authors develop a successive convex approximation (SCA)-based iterative algorithm (Algorithm 1). The approach involves:

Relaxation: Relaxing binary constraints to continuous variables within $[0, 1]$ and adding penalty terms to the objective function to encourage binary solutions.
Linearization: Approximating non-convex terms (such as interference and beampattern gain expressions) using first-order Taylor expansions to create convex lower bounds.
Iterative Solving: Solving the resulting convex sub-problem (P2) at each iteration using standard solvers (e.g., CVX) and updating the penalty parameters until convergence.

Key Contributions

Optimization Framework: The paper formulates a novel optimization problem that selects an optimal subset of subarrays to minimize total power consumption while strictly adhering to dual-function (sensing and communication) QoS requirements.
Algorithmic Solution: An efficient iterative algorithm based on SCA is developed to tackle the non-convex and combinatorial nature of the problem. The proposed method yields a high-quality suboptimal solution with polynomial computational complexity, offering a significant advantage over exhaustive search methods which possess exponential complexity.
System Modeling: The work incorporates a realistic system model that accounts for near-field channel characteristics (spherical wavefronts), spatial non-stationarity (visibility regions), and the specific constraints of hybrid subarray architectures.

Numerical Results
Simulation results validate the effectiveness of the proposed scheme:

Convergence: The algorithm demonstrates rapid convergence, stabilizing

The Big Picture: The "Super-Flashlight" Problem

The Solution: The "Smart Switchboard"

How They Did It (The Recipe)

What They Found (The Results)

More like this