Recent years have seen the rapid development of long-range wide area network (LoRaWAN) operating in region-specific sub-GHz frequency bands (e.g., 868 MHz in Europe and 915 MHz in North America). To achieve global deployment, LoRaWAN has been extended to operate in the globally available 2.4 GHz unlicensed band. However, this shift exposes LoRaWAN to significant interference from coexisting Wi-Fi networks, which share the same band and typically transmit at much higher power levels. To address this problem, this article presents InMAC, an interference-aware medium access control (MAC) protocol designed to improve coexistence between LoRaWAN and Wi-Fi networks. To the best of our knowledge, InMAC is the first MAC protocol specifically tailored to mitigate Wi-Fi interference for 2.4 GHz LoRaWAN. InMAC enhances LoRaWAN communication by probabilistically exploiting the silent time in Wi-Fi traffic, leveraging a Wi-Fi traffic profiling mechanism at LoRaWAN gateways and a packet length adaptation strategy at end devices. In addition to mitigating external interference from Wi-Fi, InMAC also tackles internal interference caused by signal collisions among LoRaWAN end devices. It incorporates a novel channel access mechanism based on channel activity detection (CAD), a carrier-sensing technique adapted specifically for LoRaWAN. Experimental results demonstrate that InMAC reduces both external Wi-Fi interference and internal LoRaWAN collisions, achieving up to a 111% throughput boost over existing approaches.

Low-power wide area networks (LPWANs) have gained significant traction in recent years, with LoRaWAN (long-range WAN) emerging as a prominent representative. Recently, LoRaWAN has expanded into the 2.4 GHz unlicensed band to support global deployments and higher data rates. However, this shift introduces severe cross-technology interference (CTI) from coexisting Wi-Fi networks that share the same spectrum, leading to substantial communication failures. As a physical-layer solution to this problem, this paper presents CoWiL to combat CTI from Wi-Fi to LoRa (the physical layer of LoRaWAN) in the 2.4 GHz band for better coexistence between LoRaWAN and Wi-Fi networks. Existing approaches address this problem by compromising Wi-Fi performance or assuming that Wi-Fi interferes with only a small portion of LoRa signals. Unlike them, CoWiL does not affect normal Wi-Fi communications and remains effective across varying levels of CTI. This is achieved by implementing CoWiL at the LoRa receiver side to directly extract LoRa data out of the CTI from Wi-Fi. Specifically, CoWiL leverages the temporal correlation between the preamble and payload of a LoRa signal, using the demodulated preamble to construct a frequency bin mask that aids in robust payload decoding. Experimental evaluations in diverse real-world settings demonstrate that CoWiL reduces the LoRa packet error rate by up to 90% compared to state-of-the-art methods under Wi-Fi-induced CTI, significantly enhancing coexistence performance.

This paper presents ER-MAC, a novel medium access control protocol for LoRaWAN that enhances both energy fairness and communication reliability. Unlike existing carriersense multiple access (CSMA) schemes, ER-MAC adaptively adjusts the number and interval of Channel Activity Detection (CAD) operations according to the CAD energy consumption of different spreading factors (SFs), ensuring balanced sensing energy among end devices using distinct SFs. ER-MAC also employs a gateway-based announcement mechanism to mitigate the hidden-terminal problem and improve transmission reliability. Experimental results show that ER-MAC achieves more than $\mathbf{2 1 \%}$ improvement in energy fairness and up to $\mathbf{7 0 \%}$ increase in packet reception ratio over state-of-the-art CSMA approaches.

Long-range wide area network (LoRaWAN) has established itself as a key enabler of long-range, low-power wireless communication for the Internet of Things (IoT). Despite its advantages, maintaining reliable connectivity in dead zones, i.e., areas where signals experience severely degraded signal-to-noise ratios (SNRs) due to obstacles, reflections, and attenuation in urban or indoor environments, remains a fundamental challenge. Existing approaches typically rely on hardware augmentation, deep learning-based signal recovery, or repeated retransmissions, which compromise cost-efficiency, energy consumption, and system scalability. In this paper, we introduce LoDez, an innovative solution designed to enhance LoRa communication reliability in dead-zone environments without requiring additional hardware, computationally intensive models, or redundant transmissions. LoDez leverages chirp redundancy, a novel physical-layer technique that improves the detectability and robustness of LoRa signals by embedding additional structured redundancy into the chirp waveform. This redundancy improves resilience to noise and fading, enabling successful demodulation even under extremely low SNR conditions. LoDez is implemented through a cross-layer design that spans the physical and medium access control layers of the LoRaWAN protocol stack, ensuring compatibility with standard LoRa deployments. We validate the effectiveness and robustness of LoDez through comprehensive field experiments conducted across diverse environments. Experimental results show that LoDez effectively mitigates dead zone communication failures, achieving up to a 70% reduction in packet error rate and a 66% reduction in energy consumption compared with state-of-the-art solutions.

Natural disasters often cause severe damage to terrestrial communication infrastructure, resulting in disrupted communication within and outside disaster zones and hindering rescue coordination efforts. Most existing studies focus on NTN performance or single emergency scenarios but lack strategies for multi-scenario resource allocation for post-disaster needs. This study explores NTNs’ technical features, proposing a multi-scenario resource allocation framework to optimize dynamic resource scheduling and services during rescue and recovery phases.

Long-range wide area network (LoRaWAN) serves as the building blocks for a variety of cutting-edge Internet-of-Things applications because of its low-power and long-range wireless networking features. Despite the increasing interest, concurrent LoRa signal transmissions frequently happen in nonorthogonal logical channels that are built over the same physical channel with the same spreading factor. This inevitably results in LoRa signal collisions at LoRa receivers, leading to network capacity degradation. To address this problem, existing proposals introduce considerable computational overhead mainly caused by redundant execution of complex signal processing operations such as discrete Fourier transform and signal sample recovery. In this article, we aim to reduce the computational overhead by presenting the design of LiLoRa, a much more lightweight signal collision resolution approach to achieve concurrent LoRa transmissions in nonorthogonal logical channels. Given a LoRa signal collision, LiLoRa includes a simple one-shot demodulation procedure regarding all the colliding chirps of the involved LoRa signals. More importantly, LiLoRa merely compares the demodulation results of colliding chirps from the perspectives of magnitude and frequency without any complex arithmetic. We conduct both theoretical analysis and practical experiments to assess the performance of LiLoRa. Evaluation results show that compared with the existing solutions, LiLoRa reduces the computational overhead by 50%-99% while maintaining the performance of signal collision resolution at a similar level.

Long-range wide area network (LoRaWAN) has become a key enabling technology for large-scale Internet of Things (IoT) deployments because of its long communication range and low power consumption. However, practical deployments often suffer from dead zones, where LoRa transmissions experience extremely low signal-to-noise ratios (SNRs), leading to severe reliability degradation. Recent studies have shown that increasing chirp redundancy can enable reliable LoRa communication in such challenging environments, but using a fixed redundancy level is inefficient and impractical under dynamic channel conditions. In this paper, we present CRA, a data link-layer chirp redundancy adaptation mechanism that jointly improves reliability and data rate for dead-zone LoRa communication. CRA leverages physical-layer signal quality measurements to dynamically adjust chirp redundancy for each end device. This proactive adaptation allows CRA to maintain network connectivity while avoiding unnecessary redundancy overhead. Experimental results across diverse environments show that CRA achieves comparable communication reliability while improving data rate by up to 148% compared with state-of-the-art approaches, thereby validating its effectiveness and robustness for enabling reliable and efficient dead-zone LoRa communication in practice.

Long-range wide area network (LoRaWAN) has emerged as a leading wireless communication technology for the Internet of Things (IoT) because of its long-range and low-power capabilities. However, dead-zone LoRa communication remains a major challenge, where end devices experience extremely low signal-to-noise ratios at gateways due to environmental factors such as signal blockage, reflection, and severe attenuation in urban and indoor environments. Existing solutions primarily rely on additional hardware, deep learning-based techniques, or repeated signal retransmissions. However, these approaches suffer from increased deployment costs, scalability issues, and energy inefficiencies. In this paper, we propose DeLoRa, a novel physical-layer solution that enables reliable LoRa communication in dead zones without requiring extra hardware, deep learning models, or excessive signal retransmissions. DeLoRa introduces chirp redundancy, which enhances demodulation robustness by improving the signal's distinguishability from the noise floor. We validate DeLoRa through extensive real-world experiments in different environments. Our results show that DeLoRa significantly enhances LoRaWAN coverage by at least 48 %, making it a practical solution for large-scale IoT deployments.

With the rapid growth of Large Language Models (LLMs), deep learning models trained on massive datasets have demonstrated impressive capabilities in understanding human language and generating diverse responses. However, current LLMs are mainly limited to static question-answering tasks and cannot interact dynamically with external information systems. This study proposes creating an intelligent chatbot system that integrates LLMs with the Model Context Protocol (MCP). This system offers features such as real-time generation of personalized travel itineraries, assistance with booking services like hotels and restaurants, cross-lingual interactions, and intelligent recommendations. By utilizing MCP’s modular context tracking and task-oriented workflow architecture, the system is designed to handle complex user scenarios and deliver personalized travel services.

In long-range wide area network (LoRaWAN), end devices located in urban environments (e.g., deep inside buildings or basements) frequently suffer from poor network connection due to signal attenuation and blockages, causing dramatic reduction of LoRaWAN coverage. To solve this problem, existing solutions resort to additional efforts from multiple LoRaWangateways, multiple antennas at a gateway, or a third-party transmitter. However, these types of network architecture incur increased hardware costs and are hard to be implemented in all LoRaWAN use cases. In this paper, we propose iLoc to improve LoRaWAN coverage in a zero-effort manner by only leveraging a single-antenna gateway. At the heart of iLoc is a novel technique that exploits the redundancy of chirps in a LoRa signal to enhance the demodulation of signals transmitted over weak links. Experimental results show that iLoc can introduce more than 40% improvement of LoRaWAN coverage in practice.

Channel estimation has attracted extensive attention for millimeter wave (mmWave) massive multiple input multiple output systems. The particle swarm optimization (PSO)-based channel estimation algorithm suffers from low convergence accuracy and falls into local extremality. This paper proposes the improved PSO (IPSO)-based channel estimation algorithm to solve these problems in the PSO-based algorithm. Firstly, a group of particles containing velocity and location information is randomly generated. Then, the proposed IPSO-based algorithm combines the hybridization technique and the extreme perturbation mechanism. Specifically, the optimization measures of updating the velocity and position information of the particle swarm through the hybridization technology can improve the convergence accuracy of the algorithm, and the introduction of the extreme perturbation mechanism expands the search area to avoid the problem of falling into the local extremum. The velocity information and the position information of the particles are updated by comparing the fitness function values. Finally, the obtained global optimal value is the required channel matrix. Theoretical studies and simulation results show that the proposed algorithm is superior to other algorithms, which is a practical channel estimation algorithm. The convergence accuracy of the proposed IPSO-based algorithm is two orders of magnitude better than that of the PSO-based algorithm, which indicates that the proposed IPSO-based algorithm has strong global search ability compared to the PSO-based algorithm. At -5, 0, 5, 10, 15, and 20 dB, the NMSE estimation accuracy of the proposed IPSO-based algorithm is 60.3

LoRa (long range) networks offer a promising solution for low-power, wide-area wireless connectivity in the Internet of Things (IoT). However, maintaining reliable LoRa signal reception under ultra-low SNR (signal-to-noise ratio) conditions remains challenging in obstructed environments. Existing solutions often rely on hardware enhancements or repeated retransmissions, compromising cost, energy efficiency, and scalability. To address these limitations, this paper presents ReLoS, a cross-layer design that embeds structured chirp redundancy at the physical layer to improve signal detectability, while dynamically adapting chirp redundancy levels at the medium access control layer based on real-time signal quality. Field experiments demonstrate that ReLoS significantly improves decoding reliability and energy efficiency under ultra-low SNR conditions, achieving up to a 335% increase in chirp recovery ratio and a 56% reduction in energy consumption compared with state-of-the-art solutions.

Millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) systems have become a research hotspot because of its advantages. Particularly, the corresponding channel estimation problem has attracted a lot of attention. However, the problem of low accuracy and high complexity cannot be ignored. Aiming at the problem of high computational complexity, a channel estimation method based on l_1/2 -FRSVT (fast randomized singular value thresholding) principle is proposed. Firstly, the channel estimation problem is transformed into the optimization problem of the objective function related to angle parameters. Then, channel estimation is performed based on the quantum particle swarm optimization (QPSO) algorithm. By updating the process equation of particles, it keeps getting closer to the true angle values. Meanwhile, in order to reduce the computational complexity, a preprocessing scheme based on the FRSVT principle is proposed to avoid calculating the singular value decomposition (SVD) directly. The key of the proposed method is to extract the approximate basis of the matrix range from the compressed matrix, and calculate some singular values of the original matrix from a small factorization matrix. In addition, a basic approximation can be avoided in each iteration by adopting the range propagation technique. Simulation results show that the proposed l_1/2 -FRSVT-based channel estimation method has good estimation accuracy, and reduces computational complexity simultaneously.

Long-range wide area network (LoRaWAN) has been newly designed to exploit the 2.4 GHz unlicensed band instead of the traditional sub-GHz bands. However, this makes LoRaWAN frequently suffer from wireless communication failures caused by the cross-technology interference (CTI) from coexisting Wi-Fi networks using the same 2.4 GHz band. As a physical-layer solution to this problem, this paper presents CoWiL to combat the CTI from Wi-Fi to LoRa (the physical layer of LoRaWAN) in the 2.4 GHz band for better coexistence between LoRaWAN and Wi-Fi networks. Existing approaches address this problem by sacrificing Wi-Fi transmission performance or assuming that Wi-Fi interferes with only a small portion of LoRa signals. Unlike them, CoWiL does not affect normal Wi-Fi communications and is workable regardless of the degree of the CTI. This is achieved by implementing CoWiL at a LoRa receiver to directly extract LoRa data out of the CTI from Wi-Fi. Specifically, CoWiL exploits the correlation of the signal demodulation results between the preamble and the payload of a LoRa signal. A novel frequency bin mask is generated based on the demodulated preamble and then applied to the following payload for data decoding. Experimental results in various real-world environments show that in comparison with the existing solutions, CoWiL can reduce the packet error rate of LoRa transmissions by up to 96% under the CTI from Wi-Fi.

As a compelling wireless networking technology for the Internet of Things, long-range wide area network (Lo-RaWAN) has been newly designed to exploit the 2.4 GHz unlicensed band instead of the traditional sub-GHz bands. However, this makes LoRaWAN frequently suffer from communication failures caused by the signal interference from coexisting Wi-Fi networks using the same 2.4 GHz band. To mitigate this problem, this paper presents a physical-layer solution to directly extract LoRaWAN data out of Wi-Fi interference. This is achieved by exploiting the correlation of the signal demodulation results between the preamble and the payload of a LoRa (the physical layer of LoRaWAN) signal. Experimental results show that in comparison with existing approaches, our proposed technique can achieve up to 83% reduction of LoRa packet error rate.

This study develops a deep learning-based Intrusion Detection System (IDS) for the Open Fronthaul (OFH) interface in Open Radio Access Network (O-RAN) architecture, focusing on detecting and mitigating Distributed Denial of Service (DDoS) attacks. It aims to enhance O-RAN network security, particularly in the CUS-Plane of the OFH interface, by employing advanced deep learning techniques for threat prediction and prevention. The research contributes practical insights and solutions to improve the security resilience of O-RAN's open architecture.

Long-range wide area network (LoRaWAN) has been newly designed to exploit the 2.4 GHz unlicensed band instead of the traditional sub-GHz bands. However, this makes LoRaWAN frequently suffer from wireless communication failures caused by the cross-technology interference (CTI) from coexisting Wi-Fi networks using the same 2.4 GHz band. As a physical-layer solution to this problem, this paper presents CoWiL to combat the CTI from Wi-Fi to LoRa (the physical layer of LoRaWAN) in the 2.4 GHz band for better coexistence between LoRaWAN and Wi-Fi networks. Existing approaches address this problem by sacrificing Wi-Fi transmission performance or assuming that Wi-Fi interferes with only a small portion of LoRa signals. Unlike them, CoWiL does not affect normal Wi-Fi communications and is workable regardless of the degree of the CTI. This is achieved by implementing CoWiL at a LoRa receiver to directly extract LoRa data out of the CTI from Wi-Fi. Specifically, CoWiL exploits the correlation of the signal demodulation results between the preamble and the payload of a LoRa signal. A novel frequency bin mask is generated based on the demodulated preamble and then applied to the following payload for data decoding. Experimental results in various real-world environments show that in comparison with the existing solutions, CoWiL can reduce the packet error rate of LoRa transmissions by up to 96% under the CTI from Wi-Fi.

This paper proposes a heterogeneous carrier-sense multiple access (CSMA) protocol named LoHEC as the first research attempt to improve energy fairness when applying CSMA to long-range wide area network (LoRaWAN). LoHEC is enabled by Channel Activity Detection (CAD), a recently introduced carrier-sensing technique to detect LoRaWAN signals even below the noise floor. The design of LoHEC is inspired by the fact that existing CAD-based CSMA proposals are in a homogeneous manner. In other words, they require LoRaWAN end devices to perform identical CAD regardless of the differences of their used network parameter - spreading factor (SF). This causes energy consumption imbalance among end devices since the consumed energy during CAD is significantly affected by SF. By considering the heterogeneity of LoRaWAN in terms of SF, LoHEC requires end devices to perform different numbers of CAD operations with different CAD intervals during channel access. Particularly, the number of needed CADs and CAD interval are determined based on the CAD energy consumption under different SFs. We conduct extensive experiments regarding LoHEC with a practical LoRaWAN testbed including 60 commercial off-the-shelf end devices. Experimental results show that in comparison with the existing solutions, LoHEC can achieve up to 0.85 x improvement of the energy fairness on average.

Enabling energy-efficient and long-distance wireless networking is expected by various Internet-of-Things systems and consumer electronics applications. Low-power wide area networks can commendably satisfy this requirement because of their combination of low-power and long-range features. Thereinto, open-standard Long-Range Wide Area Network (LoRaWAN) is the representative player with hundreds of deployment cases worldwide. Unfortunately, the current multiple access (MA) strategy for LoRaWAN has been proved to incur severe wireless signal interference under large-scale deployment. This is the inherent hindrance for the current LoRaWAN to support large-scale wireless networking. In this context, this article aims to provide a detailed tutorial of MA issues in large-scale LoRaWAN for the first time. We start with several practical use cases of large-scale LoRaWAN. We then explain the detailed challenges of large-scale LoRaWAN networking. Representative MA solutions for better LoRaWAN networking are then described and compared. Future research directions for enhanced MA protocol design are also discussed.

Recognizing if two objects are in close physical contact (CPC) is the basis of various Internet-of-Things services such as vehicle proximity alert and radiation exposure reduction. This is achieved traditionally through tailor-made proximity sensors that proactively transmit wireless signals and analyze the reflection from an object. Despite its feasibility, the past few years have witnessed the prosperity of reactive CPC detection techniques that do not need spontaneous signal transmission and merely exploit received wireless signals from a target. Unlike existing approaches entailing additional effort of multiple antennas, dedicated signal emitters, human intervention, or a back-end server, this article presents TONARI, an effortless CPC detection framework that performs in a reactive manner. TONARI is developed for the first time with LoRa, the representative of unlicensed low-power wide area network (LPWAN) technologies, as the wireless signal for CPC detection. At the heart of TONARI lies a novel feature arbitrator that decides whether two devices are in CPC or not by distinguishing different types of LoRa chirp-based additive sample magnitude sequences. Software-defined radio-based experiments are conducted to show that the achievable CPC detection accuracy via TONARI can reach 100% in most practical cases.