Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Joint Optimization of Cooperation Efficiency and Communication Covertness for Target Detection with AUVs

Xueyao Zhang, Bo Yang, Zhiwen Yu, Xuelin Cao, Wei Xiang, Bin Guo, Liang Wang, Billy Pik Lik Lau, George C. Alexandropoulos, Jun Luo, Mérouane Debbah, Zhu Han, Chau Yuen

TL;DR

This work tackles covert cooperative target detection with multiple AUVs by formulating a joint trajectory and power optimization problem under a covert communication constraint expressed as $D(\mathcal{H}_{0}||\mathcal{H}_{1})\le 2\varepsilon^{2}$. It introduces a hierarchical framework, HMAPPO, that splits decision-making into macro-level task allocation (MDP with PPO) and micro-level trajectory/power control (POMDP with MAPPO) under a CTDE regime. The approach yields efficient collaboration (high $\eta$ and $\zeta$) while maintaining low detectability (low KL divergence) and energy-aware operation, demonstrated via a high-fidelity 3D underwater simulator showing rapid convergence and favorable scalability. The results highlight a critical trade-off region: as covertness tightens, cooperation can degrade, revealing a threshold beyond which ultra-low power communications impair task execution, but the framework effectively navigates this balance through hierarchical learning and centralized guidance. Overall, the paper provides a practical, scalable method for covert multi-AUV coordination in dynamic underwater environments with clear implications for secure underwater sensing and surveillance tasks.

Abstract

This paper investigates underwater cooperative target detection using autonomous underwater vehicles (AUVs), with a focus on the critical trade-off between cooperation efficiency and communication covertness. To tackle this challenge, we first formulate a joint trajectory and power control optimization problem, and then present an innovative hierarchical action management framework to solve it. According to the hierarchical formulation, at the macro level, the master AUV models the agent selection process as a Markov decision process and deploys the proximal policy optimization algorithm for strategic task allocation. At the micro level, each selected agent's decentralized decision-making is modeled as a partially observable Markov decision process, and a multi-agent proximal policy optimization algorithm is used to dynamically adjust its trajectory and transmission power based on its local observations. Under the centralized training and decentralized execution paradigm, our target detection framework enables adaptive covert cooperation while satisfying both energy and mobility constraints. By comprehensively modeling the considered system, the involved signals and tasks, as well as energy consumption, theoretical insights and practical solutions for the efficient and secure operation of multiple AUVs are provided, offering significant implications for the execution of underwater covert communication tasks.

Joint Optimization of Cooperation Efficiency and Communication Covertness for Target Detection with AUVs

TL;DR

This work tackles covert cooperative target detection with multiple AUVs by formulating a joint trajectory and power optimization problem under a covert communication constraint expressed as . It introduces a hierarchical framework, HMAPPO, that splits decision-making into macro-level task allocation (MDP with PPO) and micro-level trajectory/power control (POMDP with MAPPO) under a CTDE regime. The approach yields efficient collaboration (high and ) while maintaining low detectability (low KL divergence) and energy-aware operation, demonstrated via a high-fidelity 3D underwater simulator showing rapid convergence and favorable scalability. The results highlight a critical trade-off region: as covertness tightens, cooperation can degrade, revealing a threshold beyond which ultra-low power communications impair task execution, but the framework effectively navigates this balance through hierarchical learning and centralized guidance. Overall, the paper provides a practical, scalable method for covert multi-AUV coordination in dynamic underwater environments with clear implications for secure underwater sensing and surveillance tasks.

Abstract

This paper investigates underwater cooperative target detection using autonomous underwater vehicles (AUVs), with a focus on the critical trade-off between cooperation efficiency and communication covertness. To tackle this challenge, we first formulate a joint trajectory and power control optimization problem, and then present an innovative hierarchical action management framework to solve it. According to the hierarchical formulation, at the macro level, the master AUV models the agent selection process as a Markov decision process and deploys the proximal policy optimization algorithm for strategic task allocation. At the micro level, each selected agent's decentralized decision-making is modeled as a partially observable Markov decision process, and a multi-agent proximal policy optimization algorithm is used to dynamically adjust its trajectory and transmission power based on its local observations. Under the centralized training and decentralized execution paradigm, our target detection framework enables adaptive covert cooperation while satisfying both energy and mobility constraints. By comprehensively modeling the considered system, the involved signals and tasks, as well as energy consumption, theoretical insights and practical solutions for the efficient and secure operation of multiple AUVs are provided, offering significant implications for the execution of underwater covert communication tasks.
Paper Structure (39 sections, 2 theorems, 52 equations, 9 figures, 2 tables, 1 algorithm)

This paper contains 39 sections, 2 theorems, 52 equations, 9 figures, 2 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background and Challenges
  3. Motivation and Contributions
  4. Literature Review
  5. Underwater Covert Communications
  6. Multi-Agent Deep Reinforcement Learning Approaches
  7. System Modeling and Problem Formulation
  8. System Model
  9. Underwater Channel Model
  10. Received Signal Model
  11. Covertness Model
  12. Task Model
  13. Task distribution stage
  14. Task execution stage
  15. Result fusion stage
  16. ...and 24 more sections

Key Result

Lemma 1

Considered a received signal $y_d$ that is modeled as a zero-mean real Gaussian random variable under both the null hypothesis $\mathcal{K}_0$ and the alternative hypothesis $\mathcal{K}_1$, with variances $\delta_0^2$ and $\delta_1^2$ respectively, where $\delta_1^2 > \delta_0^2$. The likelihood ra The optimal threshold $\Theta'$ for the energy detector is given by: ${{\Theta }^{'}}=\frac{2\ln \T

Figures (9)

  • Figure 1: The considered underwater covert communication scenario for cooperative target detection is shown in (a), and the structure of the communication time slot is illustrated in (b).
  • Figure 2: The framework of the proposed HMAPPO.
  • Figure 3: Convergence analysis of the proposed HMAPPO. Subplots show the average episodic trends for: the central AUV's macro-level reward, the AUVs' micro-level reward, the task coverage rate $\zeta$, the task efficiency $\eta$, the task execution time $T_{task}$, and the average KL divergence.
  • Figure 4: Individual AUV trajectories during the final training episode. The black stars mark the exact position where the AUVs first reached their sub-target areas.
  • Figure 5: The central AUV critic and actor loss.
  • ...and 4 more figures

Theorems & Definitions (4)

  • Lemma 1
  • proof
  • Lemma 2
  • proof