LoRaWAN has become a leading Low Power Wide Area Network (LPWAN) technology for Industrial Internet of Things (IIoT) applications, offering long range communication with low energy consumption. A fundamental challenge lies in selecting the appropriate Spreading Factor (SF) for each device, since this directly influences coverage, packet success ratio (PSR), and airtime. The default Adaptive Data Rate (ADR) mechanism is static and fails to adapt under dynamic network conditions such as mobility. This paper proposes a lightweight linear upper confidence bandit (LinUCB–SF) based reinforcement learning approach for adaptive SF selection. Each end device uses locally observable features to autonomously select its SF, balancing exploration and exploitation. The method is implemented and validated in ns-3 simulations across a range of deployment densities. Results show that our proposed LinUCB-SF algorithm reduces energy consumption by 19.3% in mobile scenarios and 37.8% in static scenarios, while improving PSR by 9.1% and 9.0%, respectively, compared to the EXP3 baseline.

This paper proposes a handheld AI CPR assitive device estimating chest compression depth with location tracking capability. By using a model of Convolutional Neural Networks (CNN) combined with Long Short-Term Memory (LSTM) networks.

This paper introduces a smart device counting and tracking approach for people flow observation, which can be applied to various people flow applications, including cases where vision-based methods cannot be used. The proposed system detects nearby smart devices and counts them by selecting valid devices within the suitable range, that observed by the path loss model for the target service area. This method is more effective than existing counting methods used at people flow applications because it moniotrs devices carried by individuals or crowds passing through the limited specific locations. The performance evaluation showed that the effectiveness of our method was demonstrated with an average improvement of $19.9 \%$ in the correlation coefficient.

The popularity of radio frequency (RF)-based fingerprinting for indoor localization has grown owing to its relatively low cost of equipment deployment and satisfactory accuracy. However, generating a complete radio map by associating unlabeled RF signals with the corresponding location information remains challenging, especially in crowdsourcing-based fingerprinting. In this article, we propose a semi-crowdsourced radio map construction method based on Bluetooth low-energy (BLE) landmarks that harnesses reference points (RPs) in the radio map for coarse localization and facilitates the labeling of location information to WiFi signals. Principally, we acquire RF-received signal strength (RSS) measurements annotating them with location coordinates recorded while a user is walking to provide an efficient method of data collection. Furthermore, we introduce a generative adversarial network (GAN)-based method to increase the amount of training data collected at each RP and reduce human effort by augmenting the fingerprint database. Our proposed method demonstrates promising results, including improved localization accuracy and localization performance comparable to that of traditional site surveys while reducing measurement time and human effort.

LoRaWAN is a low-power wide-area network technology that has become the de-facto for the Internet of Things (IoT) due to its low cost, ultra-low energy consumption, longrange, and support for the massive end devices (EDs). Adaptive data rate (ADR) is the most widely adopted approach for resource assignment with spreading factor (SF) and transmission power (TP) to massive EDs in the LoRaWAN network, recommended for static IoT applications such as metering. However, in a mobile IoT environment, ADR fails to adjust the resources due to dramatic changes in the signal strength owing to the underlying dynamic environment, resulting in massive packet loss and retransmissions. To assign suitable SF and TP parameters to mobile IoT EDs, we propose mobility adaptive data rate (M-ADR) using Kalman Filter. The proposed M-ADR determines the ED status (i.e., either static or mobile) by finding the distance between the previous and current positions of the ED at the NS. When the ED status is determined as mobile, we propose utilizing Kalman Filter to estimate the signal-to-noise ratio (SNR) to accurately determine SF, TP, or both, as these parameters are primarily dependent on SNR. When the Kalman Filter decides the current estimate of the system, the proposed M-ADR further finds the best possible configuration of the SF and TP. Simulation results show that the proposed M-ADR enhanced the packet success ratio by 16.88% compared with the state-of-the-art ADR of LoRaWAN.

It is known that an ultrawideband (UWB)-based indoor positioning system (IPS) has superior positioning performance and can meet the requirements of location-based services (LBSs) as the Internet of Things (IoT) applications. However, there is a limitation of UWB positioning when it is conducted at the nonline-of-sight (NLOS) channels degrading the UWB ranging accuracy at indoor environments. In this article, we propose an artificial intelligence (AI) applied UWB positioning system that can enhance the positioning performance by classifying channel conditions with channel impulse response (CIR) of the received UWB signal. The proposed system mitigates the positioning degradation caused by the NLOS situations by performing extended Kalman filter (EKF) localization and long short-term memory (LSTM) training of the observed channel status. The main feature of the proposed UWB positioning method is that it can be used even at unknown locations not trained with the LSTM model learning the channel status, because of our training strategy of not the position coordinates, but the UWB ranging error between UWB devices corresponding to CIR of the received UWB signal. This article provides the experimental setup and performance evaluation results of the proposed system. The evaluation results showed that the proposed AI-applied UWB positioning method significantly improved its accuracy performance compared with the existing positioning methods.

Adaptive data rate (ADR) is a widely adopted resource assignment approach in long-range wide-area networks (LoRaWANs) for static Internet of Things (IoT) applications such as smart grids and metering. Blind ADR (BADR) has been recommended for mobile IoT applications such as pet and industrial asset tracking. However, ADR and BADR cannot provide appropriate measures to alleviate the massive packet loss problem caused by the unsuitable spreading factors (SFs) assigned to end devices when they are mobile. This article proposes a novel proactive approach—“artificial intelligence-empowered resource allocation” (AI-ERA)—to address the resource assignment issue in static and mobile IoT applications. The AI-ERA approach consists of two modes, namely offline and online modes. First, a deep neural network (DNN) model is trained with a dataset generated at ns-3 in the offline mode. Second, the proposed AI-ERA approach utilizes the pretrained DNN model in the online mode to proactively assign an efficient SF for the end device before each uplink packet transmission. The proactive behavior of the AI-ERA improved the packet success ratio by an average of 32% and 28% in static and mobility scenarios compared with the typical LoRaWAN ADR, respectively.

LoRaWAN exhibits an essential feature, namely, the adaptive data rate (ADR), which has been recommended for the management of resources (e.g., the spreading factor and transmit power) of static end devices (EDs) based on channel conditions. Blind ADR (BADR) has been introduced for LoRaWAN mobile applications that experience frequent channel attenuation when the ED moves (e.g., pet-tracking). This channel condition leads to massive packet loss and retransmission, which significantly increases energy consumption. In this study, ADR and BADR are investigated in mobility environments, their limitations are highlighted, and a novel ADR “retransmission-assisted resource management (R-ARM)” system is proposed. The proposed R-ARM system operates concurrently on the ED and network server sides. This improves the network performance of the LoRaWAN. When compared to those of typical ADR approaches, the simulation results, in this case, show that R-ARM significantly enhances the packet success ratio and convergence period, and it lowers the energy consumption and packet loss ratio.

Due to the proliferating social and commercial interest on location-based services (LBS), research and development of indoor positioning system (IPS) have been expanded. Most of the efficient IPS methods utilize radio-based solutions to meet the accuracy requirement of indoor LBS. Typically fingerprinting localization is realized, which requires a site survey process where radio signatures of a localization area are annotated with their actual recorded locations. The site survey is time-consuming and labor-intensive that intensifies practical limits and challenges in realizing a reliable and scalable IPS. In this paper, we propose a crowdsourcing-based approach to acquire the training data set for Gaussian process regression (GPR). In particular, we suggest combining access point (AP) proximity information and pedestrian dead reckoning (PDR) to collect labeled data without any human intervention. The crowdsourced training data are fed to model a Gaussian process, which predicts the mean RSS and its corresponding variance across the testbed. To validate the proposed method, we compared the predicted data with the manually measured one and utilized the predicted data for localization using weighted k-nearest neighbor (Wk-NN) and maximum likelihood (ML) based fingerprinting localization. Experimental results obtained by real field deployment show that the average difference between the predicted RSS and manually measured RSS is 3.87 dBm and 80% of the localization estimation error are below 5.5m.

Indoor positioning technology has received much attention in a field of the internet of things (IoT), and various technology such as Bluetooth low energy (BLE), WiFi fingerprinting, and ultra-wideband (UWB) have been studied for positioning. Among these technology, the UWB provides features of a low power consumption and high positioning accuracy caused by good immunity to multi-path fading. However, the performance of positioning feature is deteriorated in non-line of sight (NLOS) channel. Therefore, the identification of non-line of sight (NLOS) in UWB channel is required to verify and enhance the effects of UWB positioning. In this paper, we identify NLOS by measuring signal strengths on the first path and multi-path and estimate the distance from UWB anchor to a tag in through-the-wall and hard-NLOS channels. The results show that the proposed algorithm efficiently identified NLOS in UWB channel environments.

Network lifetime and power consumption are crucial for any energy constrained ubiquitous networks, such as wireless sensor network (WSN) and wireless body area network (WBAN). A special challenge in WSN and WBAN is a transmission of a large volume of data, such as medical or non-medical images and video, which are becoming more and more demanding in various applications. Media transmission on a wireless network is very much prone to communication errors. Thus, for the efficient design, the large data messages are broken down into smaller fragments, and those smaller fragments are transmitted sequentially. But, this approach introduces the burden of exchanging redundant control packets increasing energy consumption as well as transmission delay. In this paper, we propose a data fragmentation scheme using a block acknowledgment mechanism to minimize the number of the control packets and transmission delay caused by fragmentation. We implemented the proposed scheme in various existing medium access control protocols and compared it with the original protocols through NS-2 simulations. The simulation results verify that our scheme can decrease energy consumption as well as end-to-end delay. The proposed scheme can also be easily adapted to other wireless networks, such as in medical and non-medical WBAN and security monitoring system.

Real-time health monitoring with wearable sensors is an active area of research. In this domain, observing the physical condition of elderly people or patients in personal environments such as home, office, and restroom has special significance because they might be unassisted in these locations. The elderly people have limited physical abilities and are more vulnerable to serious physical damages even with small accidents, e.g. fall. The falls are unpredictable and unavoidable. In case of a fall, early detection and prompt notification to emergency services is essential for quick recovery. However, the existing fall detection devices are bulky and uncomfortable to wear. Also, detection system using the devices requires the higher computation overhead to detect falls from activities of daily living (ADL). In this paper, we propose a new fall detection system using one sensor node which can be worn as a necklace to provide both the comfortable wearing and low computation overhead. The proposed necklace-shaped sensor node includes tri-axial accelerometer and gyroscope sensors to classify the behaviour and posture of the detection subject. The simulated experimental results performed 5 fall scenarios 50 times by 5 persons show that our proposed detection approach can successfully distinguish between ADL and fall, with sensitivities greater than 80% and specificities of 100%.

Energy conservation has been the key focus of research in low-rate wireless personal area networks (WPANs). Several algorithms have been proposed to dynamically adjust the duty cycle of nodes in beacon-enabled WPANs. Increasing beacon interval for lowering the duty cycle of end devices decreases the probability of fast node association, when passive scan is used. Moreover, the requirement of performing coordinator discovery increases with mobility of end devices because of frequent cell changes. In this paper, we propose a scheme for facilitating prompt passive discovery of coordinator(s) by mobile end devices. Simulation analysis has shown that proposed scheme can swiftly discover the coordinator(s) irrespective of beacon interval length.

In this paper, a joint uplink (UL) and downlink (DL) framework for wireless mobile camera networks over an OFDMA-based infrastructure is proposed. On the UL, this system collects unicast real-time video streams from mobile wireless camera stations (CSs) and forwards them to a control center. On the DL, the aggregated video streams are multicast to multiple mobile stations (MSs) to facilitate monitoring of the same video scenes. A target bit rate is set for each UL video stream, based on an objective function that includes both the quality of experience (QoE) and popularity of video contents as key criteria. A QoE-driven scheduling and resource allocation policy at the MAC/PHY layers, and a real-time scalable video layer adaptation algorithm in the APP layer are proposed. On the DL, a previously published opportunistic layered multicasting scheduling algorithm is applied. Simulation results demonstrate that this proposed novel architecture can significantly enhance both the spectral efficiency and the QoE of users in both UL and DL directions.

Node movement feature of Vehicular ad hoc network (VANET) closely resembles with that of mobile ad hoc network (MANET) but its high speed mobility and unpredictable movement characteristics are the key contrasting feature from that of MANET. The similarity nature suggests that the prevailing routing protocol of MANET is very much applicable to VANET. However, on the same line, the dissimilarity characteristics result in frequent loss of connectivity. This necessitates upgradation of the existing routing protocols to adapt itself into VANET scenario. The key parameter that needs to be fed into these protocols is a realistic mobility model which contains criterion linked to speed, road intersections, traffic light effect etc. In this paper, we compare performances of reactive routing protocols named Dynamic Source Routing (DSR), Ad hoc On Demand Distance Vector (AODV) and Ad hoc On Demand Multipath Distance Vector (AOMDV) in VANET using different Mobility Models provided in VanetMobiSim framework. The performances are evaluated by varying mobility, number of sources and node speed while packet delivery fraction, end to end delay and normalized routing load are used as performance metrics. The simulations have shown that AOMDV performs comparatively better than DSR and AODV in different mobility models in terms of end to end delay as performance metric.

In this paper, we present the new energy efficient MAC protocol for WSN, named as medium reservation preamble based MAC (MRPM). Inspired by S-MAC, the proposed MRPM is also a synchronized duty cycle MAC protocol. However, unlike S-MAC, the proposed protocol doesn't have separate time frame for Sync and data traffics. Both traffics are integrated in a short listen period. The listen period is further shortened by excluding the channel contention from listen period and transferring it to new period called contention period. The contention period precedes the listen period, and only transmitters wake up in this contention period and contend for medium reservation, avoiding the receivers from idle listening. Now, with these features, our MRPM is much energy efficient than the conventional synchronized duty cycle MAC protocols. We compared MRPM with S-MAC and TEEM protocols through ns-2 simulation. The simulation results show that MRPM is 55% more energy efficient than S-MAC and 35 % more than TEEM.

Several MAC protocols such as S-MAC, T-MAC, DSMAC, and TEEM have exploited scheduled sleep/listen cycles to conserve energy in sensor networks. These protocols use periodic SYNC packet in their SYNC period to follow the same schedule with their neighbors. We have found that these protocols use around 40% of their listen period for SYNC period. In an average, unused SYNC periods consume more than 20% of the total energy consumption. In this paper, we analyze the periodic nature of SYNC packet and develop a new algorithm, named as intelligent network synchronization (INS), which exploits the periodic nature of SYNC packet to reduce energy consumption. The proposed INS makes nodes bypass their own SYNC period by monitoring sleep/listen cycles of their each neighbor. We evaluate INS through both mathematical analysis and simulation. These results show the achievement of up to 25% of energy saving.

Error concealment is a well-known technique to improve reproduced picture quality in cases which parts of the coded picture are not available at a decoder for reconstruction. The conventional error concealment methods generally make use of the correlation between a damaged marcoblock and its adjacent macroblocks in the same frame and/or the previous frame. However, performance results of concealment methods are different each other and dependent on how much damaged frames and macroblocks are correlated to surrounding ones temporally and spatially. Moreover, most of them can not conceal effectively the consecutively damaged frames or macro blocks caused by wireless bursty packet loss features, since motion vectors and correlation information between the damaged images are already destroyed. The proposed error concealment aware error resilient video coding is a new method which has an intelligent error concealment selector by using pre-estimation technique of the possibly damaged motion vectors at the encoder. The method recommends the best error concealment per macroblock by transmitting additionally its error concealment selection codes to the decoder. The chosen error concealment helps to conceal the damaged MB at the decoder even when losses of frames are bursty. The experimental results show the substantial effectiveness to improve video quality at the cost of a little bit overhead at the transmitted bit streams.

Mobile ad hoc networks (MANETs) are collections of autonomous mobile nodes with links that are made or broken in an arbitrary way. Due to frequent node and link failures, multipath MANET is preferred than single-path MANET in many applications. Multipath routing schemes are used for achieving various goals such as robustness, reliability, and load balancing. However, before they can be successfully deployed several security threats must be addressed. Due to lack of fixed infrastructure, security in ad-hoc routing is challenging task, especially in multipath MANET. In this paper, we propose a robust multipath routing scheme for MANET and also security mechanism for such a routing scheme. We discuss security analysis for our scheme. And we conduct simulation to evaluate the cost of the proposed secure multipath routing scheme and present some preliminary results.