Multi-access point coordination (MAPC) is a key technology for enhancing throughput in next-generation Wi-Fi within dense overlapping basic service sets. However, existing MAPC protocols rely on static, protocol-defined rules, which limits their ability to adapt to dynamic network conditions such as varying interference levels and topologies. To address this limitation, we propose a novel Agentic AI Wi-Fi framework where each access point, modeled as an autonomous large language model agent, collaboratively reasons about the network state and negotiates adaptive coordination strategies in real time. This dynamic collaboration is achieved through a cognitive workflow that enables the agents to engage in natural language dialogue, leveraging integrated memory, reflection, and tool use to ground their decisions in past experience and environmental feedback. Comprehensive simulation results demonstrate that our agentic framework successfully learns to adapt to diverse and dynamic network environments, significantly outperforming the state-of-the-art spatial reuse baseline and validating its potential as a robust and intelligent solution for future wireless networks.

网络资源的覆盖范围日益广泛, 算力资源也逐渐成为能够提供泛在计算服务的基础设施. 然而, 在广域网络, 底层网络和计算资源缺乏密切的研究或协同设计, 仍然存在计算服务调度缓慢、 数据分发不灵活、 数据传输效率低等问题. 本文提出算力感知网络(CAN)的系统架构设计, 其核心贡献在于引入感知平面来收集、 管理并综合计算和网络的信息. 这样, 感知平面、控制平面和数据平面组成一个闭环控制系统, 增强了整个系统的感知能力、 决策能力和数据转发功能. 为了使能CAN系统, 本文提出三项关键技术: 算力路由、 弹性广播和广域高吞吐传输. 本文以人工智能(AI)模型训练、 推理和离线参数传输为例, 展示CAN的适用性, 并指出未来的一些研究方向.

We consider the problem of autonomous channel access (AutoCA), where a group of terminals tries to discover a communication strategy with an access point (AP) via a common wireless channel in a distributed fashion. Due to the irregular topology and the limited communication range of terminals, a practical challenge for AutoCA is the hidden terminal problem, which is notorious in wireless networks for deteriorating throughput and delay performances. To meet the challenge, this paper presents a new multi-agent deep reinforcement learning paradigm, dubbed MADRL-HT, tailored for AutoCA in the presence of hidden terminals. MADRL-HT exploits topological insights and transforms the observation space of each terminal into a scalable form independent of the number of terminals. To compensate for the partial observability, we put forth a look-back mechanism such that the terminals can infer behaviors of their hidden terminals from the carrier-sensed channel states as well as feedback from the AP. A window-based global reward function is proposed, whereby the terminals are instructed to maximize the system throughput while balancing the terminals' transmission opportunities over the course of learning. Considering short-packet machine-type communications, extensive numerical experiments verified the superior performance of our solution benchmarked against the legacy carrier-sense multiple access with collision avoidance (CSMA/CA) protocol.

In this paper, we investigate the design of a burst jamming detection method for delay-sensitive Internet-of-Things (IoT) applications. In order to obtain a timely detection of burst jamming, we propose an online principal direction anomaly detection (OPDAD) method. We consider the one-ring scatter channel model, where the base station equipped with a large number of antennas is elevated at a high altitude. In this case, since the angular spread of the legitimate IoT transmitter or the jammer is restricted within a narrow region, there is a distinct difference of the principal direction of the signal space between the jamming attack and the normal state. Most of existing binary hypothesis test based works cannot apply to detect burst jamming, because the attackers’ target time window does not match with the legitimate transmission. Unlike existing statistical features based batching methods, the proposed OPDAD method adopts an online iterative processing mode, which can quickly detect the exact attack time block instance by analyzing the newly coming signal. In addition, our detection method does not rely on the prior knowledge of the attacker, because it only cares the abrupt change in the principal direction of the signal space. Moreover, based on the high spatial resolution and the narrow angular spread, we provide the convergence rate estimate and derive a nearly optimal finite sample error bound for the proposed OPDAD method. Numerical results show the excellent real time capability and detection performance of our proposed method.

The deployment of 5G New Radio in unlicensed spectrum is a promising solution to alleviate the spectrum crunch for cellular networks. With the openness of unlicensed spectrum, 5G New Radio Unlicensed (NR-U) will coexist with the incumbent Wi-Fi networks. It is therefore important to study how to maintain harmonious coexistence with the Wi-Fi network. To address this issue, this paper considers two alternative throughput optimization strategies under the 3GPP fairness by adjusting the access parameter: one is to maximize the total throughput of coexisting scenario, and the other is to maximize the throughput of NR-U network. It is shown that the throughput gain of both optimization strategies are related to the initial backoff window size and the network size of Wi-Fi. Moreover, the first strategy can maximize the total throughput yet it may be unfair to the NR-U network while the second strategy can maximize NR-U throughput yet may be harmful to the total throughput. In practical scenario where the IEEE 802.11 EDCA protocol is adopted in Wi-Fi, the performance of NR-U cannot be guaranteed when optimizing the total throughput, and thus optimizing the throughput of NR-U is suggested for fair coexistence.

Opening the unlicensed bands provides additional spectrum resources for the next generation wireless network, while severe unfairness and performance degradation occur when one coexists with the incumbent users of these bands. Therefore, plenty of efforts have been made towards fair coexistence, mainly focusing on parameter tuning of listen-before-talk (LBT) and duty-cycle (DC) mechanisms. For better utilization of the unlicensed bands, it is of paramount importance to establish an access mechanism that guarantees the fairness objective among feasible mechanisms. Such access mechanism and the corresponding benchmark, nevertheless, remain largely unknown. To address this issue, this paper considers the coexistence between WiFi and the other unlicensed nodes, and aims to maximize the α-fairness between them. A benchmark is first given by solving the optimization problem. Then we propose a deep reinforcement learning (DRL) mechanism to help the unlicensed nodes make access decisions, such that they coexist with WiFi harmoniously. Extensive simulations have been carried out, and the results show that the DRL mechanism can approach the benchmark.

Multi-link operation is considered to be one of the new key features in the next generation WiFi 7, i.e., IEEE 802.11be. This paper studies the maximum network sum rate of a general M-link 802.11be network with two different synchronous multi-link channel access methods being proposed by Task Group BE, i.e., Longest Backoff and Shortest Backoff. By using a Markov renewal process to model the behavior of each Head-of-Line packet, explicit expressions of the maximum network sum rate and the corresponding optimal initial backoff window sizes are derived, and verified by simulation results. The analysis shows that Longest Backoff and Shortest Backoff achieve an identical maximum network sum rate. However, to achieve the performance limit, the initial backoff window sizes need to be adaptively tuned in a different manner under the two access methods. As the number of links grows, the initial backoff window size with Longest Backoff should be monotonically decreased, while that with Shortest Backoff should be enlarged.

The rate adaptation (RA) algorithm, which adaptively selects the rate according to the quality of the wireless environment, is one of the cornerstones of the wireless systems. In Wi-Fi networks, dynamic wireless environments are mainly due to fading channels and collisions caused by random access protocols. However, existing RA solutions mainly focus on the adaptive capability of fading channels, resulting in conservative RA policies and poor overall performance in highly congested networks. To address this problem, we propose a model-free deep reinforcement learning (DRL) based RA algorithm, named as drl RA, in this work, which incorporates the impact of collisions into the reward function design. Numerical results show that the proposed algorithm improves the throughput by 16.5% and 39.5% while reducing the latency by 25% and 19.3% compared to state-of-the-art baselines.

The existing medium access control (MAC) protocol of Wi-Fi networks (i.e., carrier-sense multiple access with collision avoidance (CSMA/CA)) suffers from poor performance in dense deployments due to the increasing number of collisions and long average backoff time in such scenarios. To tackle this issue, we propose an intelligent wireless MAC protocol based on deep learning (DL), referred to as DL-MAC, which significantly improves the spectrum efficiency of Wi-Fi networks. The goal of DL-MAC is to enable not only intelligent channel access but also intelligent rate adaptation. To achieve this goal, we design a deep neural network (DNN) that takes the historical received signal strength indications (RSSIs) as inputs and outputs joint channel access and rate adaptation decision. Notably, the proposed DLMAC takes the constraints of practical applications into account and the DL-MAC is evaluated using the experimental wireless data sampled from the actual environments on the 2. 4GHz frequency band. The experimental results show that our DLMAC can achieve around 86% performance of the global optimal MAC, and about twice the performance of the traditional Wi-Fi MAC in the environments of our lab and the Shenzhen Baoan International Airport departure hall.

In this paper we propose a joint Radar and Communications waveform for next-generation wireless networks (e.g., 6G, next-generation Wi-Fi). The proposed waveform is a multiple carrier signal with each carrier a chirp-like wave instead of a sinusoid in OFDM. The signal is a parameterized waveforms with two adjustable parameters, and could be generated by affine Fourier transform once the parameters are given. We analyze the communications and ranging performances of the waveform by deriving the transmitting-receiving model and the ambiguity function and compare them with an OFDM signal. Further, we investigate the secrecy of the waveform in both communications and ranging and show the sensitivity of the waveform to the mismatch of these parameters. When these parameters are kept secret from the potential adversary, the legitimate signal cannot be demodulated correctly by the eavesdropper and the waveform spoofing attack is also invalided. Simulations demonstrate that the proposed multiple-carrier chirp waveform has better performance than OFDM signal in both communications and ranging, which has lower BER in fading channel and higher resolution in ranging. We also evaluate the performance under eavesdropping and spoofing attack1.

In the next generation wireless networks, more applications will emerge, covering virtual reality movies, augmented reality, holographic three-dimensional telepresence, haptic telemedicine and so on, which require the provisioning of high bandwidth efficiency and low latency services. In order to better support the aforementioned applications and services, novel distributed channel access (DCA) schemes are necessary. Therefore, we propose a new MAC protocol, QMIX-advanced Listen-Before-Talk (QLBT), based on the cutting-edge multi-agent reinforcement learning (MARL) algorithm. It employs a centralized training with decentralized execution (CTDE) framework to exploit the overall information of all agents during training, and ensure that each agent can independently infer the optimal channel access behavior based on its local observation. We enhance QMIX, a well-known MARL algorithm, by introducing an extra individual Q-value for each agent in the mixing network apart from the original total Q-value, which makes QLBT more stable. Moreover, delay to last successful transmission (D2LT) is first introduced in this work as a part of the observations of each QLBT agent, which facilitates agents to reach a cooperative policy that prioritizes the agent with the longest delay. Finally, extensive simulation experiments are provided to show that the proposed QLBT algorithm: 1) outperforms CSMA/CA and even its theoretical performance bound in various scenarios including saturated traffic, unsaturated traffic and delay-sensitive traffic; 2) is robust in dynamic environment; and 3) is able to friendly coexist with “legacy” CSMA/CA stations.

In this paper, we propose a robust secrecy transmission scheme for intelligent reflecting surface (IRS) aided communication systems. Different from all the existing works where IRS has already been deployed at a fixed location, we take the location of IRS as a variable to maximize the secrecy rate (SR) under the outage probability constraint by jointly optimizing the location of IRS, transmit beamformer and IRS phase shifts with imperfect channel state information (CSI) of Eve, where we consider two cases: a) the location of Eve is known; b) only a suspicious area of Eve is available. We show a critical observation that CSI models are different before and after IRS deployment, thus the optimization problem could be decomposed and solved via a two-stage framework. For case a), in the first stage, universal upper bounds of outage probabilities only related to the location of IRS are derived which can be optimized via successive convex approximation (SCA) method. In the second stage, we develop an alternative optimization (AO) algorithm to optimize beamformer and phase shifts iteratively. For case b), we propose a Max-Min SR scheme based on two-stage framework, where the location of IRS is optimized based on the worst location of Eve. Simulation results indicate the importance of the location of IRS optimization.

The Sketch is a compact data structure useful for network measurements. However, to cope with the high speeds of the current data plane, it needs to be held in the small on-chip memory (SRAM). Therefore, the product of the counter size and the number of counters must be below a certain limit. With small counters, some will overflow. With large counters, the total number of counters will be small, but each counter will be shared by more flows, leading to poor accuracy. To address this issue, we propose a generic technique:self-adaptive counters (SEAD Counter). When the value of the counter is small, it works as a standard counter. When the value of the counter is large however, we increment it using a predefined probability, so as to represent this large value. Moreover, in the SEAD Counter, the probability decreases when the value increases. We show that this technique can significantly improve the accuracy of counters. This technique can be adapted to different circumstances. We theoretically analyze the improvements achieved by the SEAD Counter. We further show that our SEAD Counter can be extended to three typical sketches and Bloom filters. We conduct extensive experiments on three real datasets and one synthetic dataset. The experimental results show that, compared with the state-of-the-art, sketches using the SEAD Counter improve the accuracy by up to 13.6 times, while the Bloom filters using SEAD Counter can reduce the false positive rate by more than one order of magnitude.

In this work, we propose an artificial intelligence (AI) based channel quality assessment algorithm for Bluetooth adaptive frequency hopping (AFH) to avoid interference between heterogeneous systems coexist in 2.4GHz Industrial, Scientific, and Medical (ISM) band. The proposed network takes the received signal strength indicator (RSSI) of all Bluetooth channels as input and outputs the estimated channel quality, which is further used to update AFH channel map. A gated recurrent unit (GRU) is adopted to extract the temporal information of interference on each channel. A novel loss function combining classification loss and ranking loss is designed to improve the performance of the neural network. The complexity analysis shows that the network considered is lightweight and resource-friendly. Moreover, simulation experiments under different interference models show that the proposed method outperforms several existing channel selection schemes.

Sketch-based measurement has emerged as a promising alternative to the traditional sampling-based network measurement approaches due to its high accuracy and resource efficiency. While there have been various designs around sketches, they focus on measuring one particular flow key, and it is infeasible to support many keys based on these sketches. In this work, we take a significant step towards supporting arbitrary partial key queries, where we only need to specify a full range of possible flow keys that are of interest before measurement starts, and in query time, we can extract the information of any key in that range. We design CocoSketch, which casts arbitrary partial key queries to the subset sum estimation problem and makes the theoretical tools for subset sum estimation practical. To realize desirable resource-accuracy tradeoffs in software and hardware platforms, we propose two techniques: (1) stochastic variance minimization to significantly reduce per-packet update delay, and (2) removing circular dependencies in the per-packet update logic to make the implementation hardware-friendly. We implement CocoSketch on four popular platforms (CPU, Open vSwitch, P4, and FPGA) and show that compared to baselines that use traditional single-key sketches, CocoSketch improves average packet processing throughput by 27.2x and accuracy by 10.4x when measuring six flow keys.

Passive radio-frequency identification (RFID) communication raises new transmission secrecy protection challenges, since passive tags stored information lack effective information protection mechanisms. Due to constraints of passive tags, such as limited computation and storage capabilities, security solutions based on the physical-layer security (PLS) are promising candidates compared to those based on conventional lightweight cryptography. Unlike existing endeavors on PLS of RFID wireless communication, we consider an RFID system in the presence of a special proactive eavesdropper, which is able to both enhance the information wiretap and interfere with the information detection at the RFID reader simultaneously by broadcasting its own continuous-wave (CW) signal. To defend against proactive eavesdropping attacks, we propose a wiretap-channel-conscious artificial-noise (AN)-aided secure transmission scheme for the RFID reader, which first estimates both legitimate and wiretap channels and then superimposes an AN signal on the CW signal to confuse the proactive eavesdropper. The transmit power and power allocation between the AN signal and the CW signal are optimized to maximize the secrecy rate. Furthermore, we model the attack and defense process between the proactive eavesdropper and the RFID reader as a hierarchical security game and prove it can achieve the equilibrium. The simulation results show the superiority of our proposed scheme in terms of the secrecy rate and the interactions between the RFID reader and the proactive eavesdropper.

Sketch is a compact data structure for network measurements. To achieve fast speed, it needs to be held in the on-chip memory (SRAM), which is very small. To enable the sketch fit into the on-chip memory, the product of counter size and number of counters must be below a certain limit. If we use small counters, e.g., 8 bits, some counters will overflow. If we use large counters, e.g., 16 bits per counter, the total number of counters will be small, each counter will be shared by more flows, leading to poor accuracy. To address this issue, we propose a generic technique: self-adaptive counters (SA Counter). When the value of the counter is small, it works as a normal counter. When the value of the counter is large, we increment it using a predefined probability, so as to represent a large value. Moreover, in SA Counter, the probability decreases when the value increases. This technique can significantly improve the accuracy of sketches. To verify the effectiveness of SA Counter, we apply SA Counter to three typical sketches, and conduct extensive experiments on one real dataset and one synthetic dataset. Experimental results show that, compared with the state-of-the-art, sketches using SA Counter improve the accuracy by up to 13.6 times.

When network is undergoing problems such as congestion, scan attack, DDoS attack, etc, measurements are much more important than usual. In this case, traffic characteristics including available bandwidth, packet rate, and flow size distribution vary drastically, significantly degrading the performance of measurements. To address this issue, we propose the Elastic sketch. It is adaptive to currently traffic characteristics. Besides, it is generic to measurement tasks and platforms. We implement the Elastic sketch on six platforms: P4, FPGA, GPU, CPU, multi-core CPU, and OVS, to process six typical measurement tasks. Experimental results and theoretical analysis show that the Elastic sketch can adapt well to traffic characteristics. Compared to the state-of-the-art, the Elastic sketch achieves 44.6 ~ 45.2 times faster speed and 2.0 ~ 273.7 smaller error rate.

The problem of multi-set membership query plays a significant role in many network applications, including routers and firewalls. Answering multi-set membership query means telling whether an element belongs to the multi-set, and if yes, which particular set it belongs to. Most traditional solutions for multi-set membership query are based on Bloom filters. However, these solutions cannot achieve high accuracy and high speed at the same time when the memory is tight. To address this issue, this paper presents the ID Bloom Filter (IBF) and ID Bloom Filter with ones' Complement (IBFC). The key technique in IBF is mapping each element to k positions in a filter and directly recording its set ID at these positions. It has a small memory usage as well as a high processing speed. To achieve higher accuracy, we propose IBFC that records the set ID and its ones' complement together. The experimental results show that our IBF and IBFC are faster than the state-of-the-art while achieving a high accuracy.

When network is undergoing problems such as congestion, scan attack, DDoS attack, etc., measurements are much more important than usual. In this case, traffic characteristics including available bandwidth, packet rate, and flow size distribution vary drastically, significantly degrading the performance of measurements. To address this issue, we propose the Elastic sketch. It is adaptive to currently traffic characteristics. Besides, it is generic to measurement tasks and platforms. We implement the Elastic sketch on six platforms: P4, FPGA, GPU, CPU, multi-core CPU, and OVS, to process six typical measurement tasks. Experimental results and theoretical analysis show that the Elastic sketch can adapt well to traffic characteristics. Compared to the state-of-the-art, the Elastic sketch achieves 44.6 ∼ 45.2 times faster speed and 2.0 ∼ 273.7 smaller error rate.

Sketch is a probabilistic data structure widely used for per-flow measurement in the real network. The key metrics of sketches for per-flow measurement are their memory usage, accuracy, and speed. There are a variety of sketches, but they cannot achieve both high accuracy and high speed at the same time given a fixed memory size. To address this issue, we propose a new sketch, namely the OM (One Memory) sketch. It achieves much higher accuracy than the state-of-the-art, and achieves close to one memory access and one hash computation for each insertion or query. The key methodology of our OM sketch is to leverage word constraint and fingerprint techniques based on a hierarchical structure. Extensive experiments based on real IP traces show that the accuracy is improved up to 10.64 times while the speed is improved up to 2.50 times, compared with the well-known CM sketch [1]. All the related source code has been released at GitHub [2].

Application Driven Network(ADN) is a new paradigm that provides on-demand differentiated services for applications. A physical network in ADN is sliced into various logically isolated sub-networks. Each network slice can have its own network architecture and protocol to serve one application exclusively. ADN enhances the user experience while keeping the resource efficiency by further imposing multiplexing among these logically isolated sub-networks.

We consider the joint channel and power allocation problem for the OFDMA system. The problem is to find a joint channel and power allocation strategy to minimize the total transmission power subject to quality of service constraints and the OFDMA constraint (i.e, at most one user is allowed to access each channel). Since the problem is generally NP-hard, the idea of the existing algorithms is to heuristically allocate the channel and power resources separately. In this paper, we propose a novel iterative reweighted minimization framework based on an effective relaxation, which is beneficial by reformulating the combinatorial OFDMA constraint as an equivalent continuous optimization problem. The proposed framework simultaneously allocates the channel and power resources, and thus is sharply different from the existing ones. Simulation results show the proposed iterative reweighted minimization methods significantly outperform the existing algorithms.

This study concerns joint channel and power allocation scheme for multi‐user orthogonal frequency division multiple access system. The author's highlight is margin adaptive (MA) resource allocation problem namely minimising the total transmit power of users with rate requirement constraints. MA is generally provable NP‐hard; the typical methods are either to relax and round, or to fix the transmission mode of users (e.g. modulation and coding). Differently, they reorganise MA problem with only power variables left and design a novel relaxation scheme to enable the convexity. The polynomial‐time algorithm‐interior‐point method‐is employed to solve the relaxation problem and the theoretical complexity is further presented. Simulation results demonstrate that the author's scheme can provide high energy efficiency compared with the existing methods, 100% relative error bounds with respect to the optimum in most cases, and low computational complexity.

We propose a new scheme to mitigate the uplink co-tier interference in dense femtocell networks, assuming that advanced receivers are employed by the femtocell base stations (FBSs). The conventional solution where interfering cells are assigned orthogonal spectrum resources leads to the inefficient spectrum usage. We exploit resource reuse by taking advantage of the successive group decoder (SGD) such that users can opportunistically access the entitled resources of nearby cells. The SGD decodes some interference signals in order to best decode the useful signal. Multi-cell uplink resource allocation with SGDs is formulated as a joint channel, rate and decoding group (CRG) allocation problem to maximize the weighted sum rates of the variable bit rate (VBR) users while meeting the rate requirements of the guaranteed bit rate (GBR) users. Due to the provable NP-hardness of the problem, we propose a greedy algorithm where the idea is to let GBR users opportunistically transmit on the channels of nearby cells and release more interference-free channels for high-rate transmission of VBR users. A semi-analytical framework is developed to provide a rough estimate of the potential throughput gain by the proposed technique. Simulation results show that the throughput gain over the conventional orthogonal resource allocation ranges from 10%–140% with different traffic proportion, number of users and rate requirement.

Typical interference avoidance methods in HetNets are arranging orthogonal resources between macrocell and small cells such as in frequency domain in inter-cell interference coordination (ICIC) or time domain in enhanced ICIC (eICIC). However, wireless channels experience time-frequency fading appealing for a two dimensional resource allocation scheme. By optimizing the resource pattern based on variable channel conditions, plus the introduction of group lasso term, our scheme exploits channel variations in both frequency and time domains and as well avoid interference. By grouping the resources in our optimization model, our scheme is flexible and robust against heavy bursty traffic. It is also a lightweight (in terms of coordination overhead) and completely distributed approach, making it suitable for practical implementation. Simulation results show the effectiveness of our proposed method compared with sole frequency domain and time domain approaches.

In this study, the authors study the downlink resource allocation optimisation in femtocell networks, to maximise the number of guaranteed users whose data rate requirements are fully met. The spectral access of femtocell networks is based on orthogonal frequency division multiple access. In their work, two challenges are solved. The first is the intractable inter‐cell interference coordination brought about by the transmission delay in backhaul connections, and the second is the incorporation of physical interference model into problem formulations. To solve these problems, the authors propose a novel two‐level resource allocation scheme, implemented in both radio resource management controller and femtocell base stations, based on the maximisation of the number of guaranteed users. Notably, their proposed scheme is efficient and requires low overhead. Simulation results show that the proposed scheme offers significant performance improvement in both the percentage of guaranteed users and spectrum spatial reuse over existing methods proposed in the literature.

Femtocells have emerged as an effective solution to enhance indoor coverage and improve system performance in cellular networks. However, the inter-cell interference (ICI) caused by the unplanned nature of femtocells considerably leads to the degradation in throughput of users with guaranteed performance (GP). Meanwhile, the existing resource allocation algorithms bring about high complexity and large overhead. By tracking channel variation as well as arrival and departure of users, we propose a hybrid centralized/distributed resource allocation algorithm to maximize the number of GP users, with lower complexity and smaller overhead. Simulation results show that the proposed algorithm can significantly improve the system performance compared to the existing algorithms such as Q-FCRA.