This systematic review synthesizes peer-reviewed studies published between 2023 and 2026 on communication-efficient networking for distributed agentic AI, multi-agent reinforcement learning and networked autonomous systems concludes that communication efficiency should be treated as a joint optimization problem involving bandwidth, latency, computation, energy and task performance.
Abstract
Distributed agentic artificial intelligence increasingly relies on networked groups of autonomous agents that exchange observations, intentions, plans and task states to achieve collective goals. This systematic review synthesizes peer-reviewed studies published between 2023 and 2026 on communication-efficient networking for distributed agentic AI, multi-agent reinforcement learning and networked autonomous systems. Following PRISMA 2020, studies were identified from major scholarly databases and analyzed thematically across communication protocols, message compression and selection, semantic communication, coordination mechanisms, latency reduction, energy efficiency and deployment challenges. The evidence shows that selective engagement, graph-based compression, delay-aware communication, implicit consensus and value-of-information scheduling can reduce redundant exchanges while maintaining coordination quality. Semantic and edge–cloud approaches further lower payload size and local computational demand, although their benefits depend on channel conditions, resource availability and task placement. Persistent limitations include scalability, protocol interoperability, security and privacy risks, inconsistent energy reporting, dependence on simulated environments and limited standardization of evaluation metrics. The review concludes that communication efficiency should be treated as a joint optimization problem involving bandwidth, latency, computation, energy and task performance. Future research should prioritize interoperable protocols, adaptive communication topologies, secure message exchange, realistic testbeds and standardized reporting frameworks for dependable, scalable and sustainable distributed agentic systems at scale.
The main conclusion is that practical Agentic IoT depends less on placing an entire agent at one tier than on partitioning perception, memory, reasoning, and action under explicit latency, privacy, reliability, and safety constraints.
Artificial Intelligence systems that operate with multiple agents are increasingly being used to address workflows that are complex and distributed but the performance of multi-agent systems is often limited by the fact that the underlying communication overheads can be hidden instead of performance limits. The paper is a systematic exploration of the efficiency of inter-agent communication in three popular orchestration systems: LangGraph, CrewAI, and the OpenAI Agents SDK. A controlled benchmarking environment is crafted with the same task structure to isolate delays associated with coordination, redundancy in messages, and resource usage. The analysis shows that communication patterns have a great impact on the overall system performance, and that graph-based orchestration creates a new coordination latency, whereas sequential delegation models have a quick increase in contextual payloads. On the other hand, lightweight orchestration has lower latency but less flexibility when subject to complex workflows.The main value of the work is the introduction of a communication-focused assessment framework that measures the costs of coordination without depending on the computation of the model. Moreover, the analytically validated optimized strategies, including asynchronous execution, structured message encoding, and adaptive task scheduling are offered. The experimental results reveal that there are quantifiable improvements in the reduction of latency and the efficiency of resources with the use of these strategies. The results offer useful design tips to create scalable, high-performance multi-agent systems and form the basis of future studies on communication-aware AI orchestration.
Shania Rasheed Nalagath, Ankur Gupta, Sandeep Reddy Peddi et al.· International Conference on...· 0 citations
Large swarms of intelligent agents are useful only when coordination still works after the tidy assumptions of a laboratory test disappear. That is the practical problem behind decentralized coordination architectures. The review follows the field from graph-based modeling and consensus theory to the less tidy questions that appear in deployment: delayed messages, changing topology, faulty agents, and limited computation. The discussion treats distributed consensus, event-triggered communication, resilient control, fault-tolerant design, and cognition-inspired adaptation as parts of one architecture problem. The literature from 2020 to 2025 suggests a clear pattern. Stabilizing high-order nonlinear swarms is still difficult; robust performance under intermittent connectivity remains fragile; and adaptive decision-making is not yet fully reliable when the environment keeps changing. A combined route is needed, one in which cognitive intelligence, edge-side computing, and verification tools are developed together. Such a route is especially relevant to intelligent manufacturing, emergency response, autonomous transportation, and other settings where a centralized command chain may be too slow or too vulnerable.
Unknown authors· Applied and Computational En...· 0 citations
The study provides initial evidence of feasibility while identifying the challenges that must be addressed before production deployment and formalize the ADN agent model and workflow and define an operational framework covering communication, lifecycle management, governance, and security.
F. Rossi, Paulo Silas Severo De Souza, Diogo Mainart Monteiro et al.· IEEE Access· 0 citations
In this paper, we propose an adaptive hybrid covert channel (AHCC) framework that enables proactive covert communications, focusing on improving covertness and bit error rate (BER), especially in fluctuating Internet Protocol version 6 (IPv6) network environments. Our AHCC framework is built around three key innovative modules: covert encoding, proactive channel coordination, and agentic AI modulation. Specifically, we first enhance the covertness by introducing an Analog Fountain Codes-based encoding mechanism that without requiring receiver feedback. Next, we propose a proactive covert channel coordination (P3C) algorithm that dynamically adjusts the proportion of timing-based covert channels according to the current network state. Lastly, we design an agentic AI module that coordinates two knowledge-driven agents. A storage knowledge-driven agent that exploits protocol-aware field selection to reduce BER, and a timing knowledge-driven agent that regulates inter-packet delays based on traffic distribution characteristics to enhance covertness. Simulation results show that the BER of our proposed AHCC framework is reduced to less than one-third of that of the baseline schemes under congested network conditions. In addition, the Kullback-Leibler divergence and Kolmogorov-Smirnov statistic decrease by at least 8.85% and 37.5%, respectively, demonstrating the effective covertness of our proposed approach.
Kaikai Huang, Yao Yu, Yanhao Wang et al.· IEEE Transactions on Cogniti...· 0 citations
Timely and dependable information exchange is essential for large-scale unmanned aerial vehicle (UAV) swarms to coordinate under their fast motion, intermittent links, and limited energy on board. However, swarm deployments increasingly must contend with spectrum contention and jamming, as well as a lack of dependable infrastructure, which can reveal the shortcomings of traditional radio-frequency (RF) networking. This paper presents a synthesized overview of communication technologies and networking architectures for UAV swarm operations in FANETs. Representative studies were identified by a structured search and screening process across major venues of scholarly output, which are synthesized using a cross-layer lens including physical links, medium access, routing, information-centric networking, learning-enabled adaptation, and security. In this article, We compare RF/cellular with emerging high-capacity links including millimeter-wave and free-space optical communication, and then show how routing/indirection and content/function-centric paradigms (NDN/NFN) can mitigate fragility imposed by reliance on brittle end-to-end paths. Lastly, we analyse learning-based control (especially multi-agent reinforcement learning) for communication-aware mobility and resources management, as well as security approaches for contested settings. The resulting design perspective highlights recurring trade-offs among reliability, latency, throughput, energy, and integrity, and identifies practical research directions toward more robust and deployable swarm communication systems.
Azzam Almekhlafi, Y. Alqudsi· 2026 6th International Confe...· 0 citations