Recently, ride-sharing has become an essential service for daily use. Some popular mobile apps have been utilised to match up between a user and a rider. Mobile apps provide mutual location information and private information for checking security levels, such as their profile photos, the rider’s star level, and feedback records about the user. Despite providing such information, frequent mismatching between a user and a vehicle causes severe safety problems. Therefore, this paper proposes a secure coupling scheme between users and riders via fine-grained cyber-physical interaction to prevent getting into the wrong car. The proposed scheme relies on dynamic Bluetooth low energy (BLE) beaconing to provide precise mutual detection between passengers and riders. Smart mobile devices dynamically generate BLE beacon messages to recognise and authenticate each other, and such beaconing is also used to measure the highly accurate distance between them. For mutual beaconing, the scheme deals with frequent switching of the advertising and scanning modes of the Bluetooth channel and their interval. Since passengers and vehicles might continuously move, accurate distance detection is not trivial. So, this scheme considers edge servers to apply various machine learning techniques and filtering schemes to achieve precise detection. Performance evaluations rely on a proof-of-concept prototype, and various field test data are utilised to prove the proposed concept’s performance.

Along with the advent of COVID-19 epidemic, delivery services have become one of the most common services in our daily life. However, the current services merely assume that the contact for handing over delivery items is denied to avoid facing between carriers and receivers. This extreme concept might lead to the ruin of typical processes for delivery services, such as face-to-face authentication, item checking, and payment in person. So, it may cause insufficiency regarding both sides of the secure item delivery and customers’ personal security. Therefore, this paper comes up with a novel CONTACTFUL delivery service with the digital twin paradigm to resolve contactless problems. The proposed idea establishes two-tier digital twin architecture to arrange digital twins and physical twins of carriers and receivers. Then, inter-digital-twin interaction virtually supports on-site delivery service processes. The inter-twin interaction relies on highly accurate proximity detection for items, carriers, and receivers. So, the proposed scheme considers machine-learning technologies and filtering algorithms to achieve high precision based on mobile-based Bluetooth low energy (BLE) beaconing. This scheme is implemented as a proof-of-concept prototype, and various field data are exploited to build ML-filtering models including their ensemble one. On the steel door, the ensemble model shows an accuracy improvement of 65% compared to SVM, 14% to CNN, and 73% to MLP. On the wood door, the ensemble model shows an accuracy improvement of 21% compared to SVM, 9% to CNN, and 17% to MLP.

This paper proposes a novel scheme to detect and localize the spy cameras based on AI algorithm based raw traffic analytics, named AI-aided Hidden Camera Locator (AHCL). In AHCL, the video streaming data are filtered via the SVM (support vector machine) algorithm to quickly monitor whole raw network traffic from a router to the networks first. Then, gathered traffic data are denoised by the Denoising Autoencoder (DAE) technique to improve the data quality of classification for localization, where a camera transmits video streaming. Based on the proof-of-concept implementation, the proposed scheme can achieve 99.5% positioning accuracy of camera detection with the Ensemble Neural Networks (NNs).

This paper proposes a novel individual monitoring mechanism to support rescuing people in indoor places, named Cyber-Physical Individual monitoring (CP-Im), When firefighters go to the rescue of people existing within an indoor place where is on fire or broken down due to an earthquake or some disaster, the existence and accurate location of victims are most important information. So, this paper tries to solve such practical demand relying on the present life nature with well-deployed mobile network infrastructure and densely-populated smartphone users. CP-Im shows 98.82% of the user recognition precision and 96.5% of the user position accuracy is achieved.

Parking is an everyday ordinary activity in urban lives. In other words, it is the most common and essential requirement of all drivers in a car park to fast search a preferred parking spot. Recently most parking lots provide a sensing-based display system to notify available parking spots. However, such systems are still unable to tell drivers exact parking spots and make any recommendation to improve the traffic conditions and driver experiences. In this paper, a novel machine-learning-aided smart parking system based on Internet of Things, smart mobile devices and edge computing is proposed. This novel parking system aims at providing customized parking experience to users through highly accurate positioning and user activity sensing computed as the edge intelligence. The proposed system mainly focuses on increasing user experience for parking through the concatenated autoencoder design for improving performances of the user positioning mechanism.

In wireless sensor networks, sink mobility support is one of the essential functionalities in many applications. With continuous advancement, future applications will require not only sink mobility support but also high-performance data delivery service. Multipath routing is one of the promising technologies for improving data delivery performance by collaboratively using alternative or redundant multiple routing paths. However, existing multipath routing protocols had not dealt with sink mobility. As a result, they lead to bad performance in terms of energy efficiency due to the end-to-end path reconstruction. Consequently, a novel multipath management scheme is required thereby supporting sink mobility without performance degradation. In this paper, we propose a multipath management scheme for supporting sink mobility. The proposed scheme dynamically constructs multipath along the moving path of a sink. In addition, the proposed scheme provides the path shortening schemes according to the sink’s movement for reducing energy consumption. Our simulation results show that the proposed scheme is superior to existing path management schemes in terms of reliability and energy efficiency.

Parking at an affordable place is the precedent task for all activities of the everyday life in urban environments such as shopping, working, exercising, etc. So, it is the most common and essential requirement of all users in a car park to fast search a preferred parking spot closely associated their current intent. Although modern parking lots have installed the sensing and display systems to inform drivers on the availability of parking areas, such systems are unable to tell drivers exact parking spots and make any recommendation to improve the traffic conditions and driver experiences. In this paper, a novel analytic- based smart parking system clustering Internet of Things, smart mobile devices and edge computing is proposed. This novel parking system aims at providing customized parking experience to users through highly accurate positioning and user status detection which are achieved by joint mobile sensing-machine learning based analytics as the edge intelligence. Based on the proof-of-concept implementation, the proposed scheme can achieve 99.1% positioning accuracy of a parking spot; in terms of user status sensing, especially getting in a car and out of a car, detection accuracy shows 96%; finally, it shows much shorter service consumption time of 15.6 times than the legacy approach.

Indoor positioning of individuals is one of the most important technologies in smart home applications for user-customized support. The indoor positioning is typically fulfilled through radio signal strength indicators (RSSIs) of referred devices with specific media such as Wi-Fi access points (APs) and Wi-Fi station devices (STAs). So, the capability of typical positioning schemes is highly close to the signal acquisition environments of such media. In other words, since STAs frequently fall into a sleep mode to save the battery power and some data acquisition technologies are based on advertising intervals, the amount of RSSIs of referring devices could not be gathered enough to detect correct positions of individuals. In this paper, a novel individual positioning mechanism, called PIPing (Proxy Individual Positioning), is come up with. The PIPing mechanism carries out proxy signal acquisition via IEEE 802.11 monitor mode devices to overcome such restrictions. In addition, PIPing includes machine learning based signal data analytics to provide high reliable results for positioning. Based on the proof-of-concept prototype, PIPing can acquire much higher amount of RSSI data than existing manners, about 330% increment; the reliability of positioning for a home with seven rooms shows 96.4% via the support vector machine (SVM) and 96.5% by the multilayer perceptron (MLP) with autoencoder denoising to tune up signals.

Nowadays, mobile applications are engaged with and depended on the physical world more and more. Interaction with a certain proximity and thing is bringing smarter services into our daily lives through cyber-physical world collaboration. Push notification of meaningful content associated with the proximity or thing is one of key technologies of the cyber-physical services. For example, empty parking spot recommendation in a parking lot or promotion coupon notification of a shop at a shopping cent er should be delivered accordingly. This paper explores reliable push notification of content for novel interactive mobile applications over Bluetooth Low Energy (BLE) beacon infrastructure. Since mobile users move around with different speeds, this paper deals with such difference to relay right content in a reliable manner, called Reliable Push Notification (RPN). RPN dynamically clusters contents and links them to a beacon to achieve reliability, and then adaptively pre-allocate contents for transmission.

In wireless sensor networks, the multipath has been attracted for reliable data transmission due to the limited feature of sensors. The disjoint feature of multipath is one of important issues because it maintains how to separate each path in order to prevent the node energy depletion or node failure. In this issue, void areas should be considered since it disturbs the disjoint path construction. Existing multipath studies proactively detect the voids in the entire network by void modeling and thus construct paths by detouring the voids. However, void modeling causes large overhead to energy-limited nodes and even it could lead to the energy exhaustion of nodes or the node failure. To deal with this, we propose an energy- efficient and disjointed multipath routing for void handling based on geographic void handling approach. The proposed scheme is able to construct disjoint multipath avoiding the overspend of node energy and detouring the void area without additional method like modeling. Simulation results show the proposed scheme is better than the previous works in terms of energy consumption, delivery delay time, and path length around void area.

Geographic greedy routing in Wireless Sensor Networks (WSNs) suffers from the local minimum problem, which cannot select the next forwarding node such as the void. Especially, the void, which encountered during data transmission, have a negative effect on the time-sensitive application by the time delay in order to detour the void. Further, it is caused the void diffusion problem to exploiting the same node repeatedly in order to detour the void. To solve these problems, we proposed a dynamic anchor based void avoidance scheme for real-time service in WSNs. The proposed scheme broadcast the information of the void in network initialization phase in order to reduce a disadvantage caused the shape and location of the void. Furthermore, the source node calculates the dynamic anchor points using the information in order to satisfy the requirement of the real-time application and transmits the data to the sink node. Simulation result shows the proposed scheme is suitable for time-sensitive application and alleviate the void diffusion problem.

The wireless sensor networks (WSNs) are special network which has purpose of gathering information in certain area. The reliability is an important factor in the WSNs for transferring the information to the destination. However, the reliability is decreased while multi-hop transmission of the information because a node which composes the WSNs has limited communication range. Also, the transmission process may suffer from the various factors including the faulty nodes and network congestion. To solve this, the multipath routing protocols, which transfer the information through a multipath to the destination at the same time, have been proposed for improving the reliability. In the existing multipath routing protocols, however, the network scale should be sufficient for constructing multipath to guarantee of the disjoint path and the energy is consumed for multipath construction/maintenance consistently. The energy consumption leads to the reduction of networks lifetime and the restricted network scale may not even ensure the enough reliability. Thus, we proposed Low-Cost Multipath Routing Protocol by Adapting Opportunistic Routing which can reduce consistent energy consumption and less affected by network scale. The proposed protocol exploits the opportunistic routing that improves the single-hop reliability in order to improve the reliability in each path. It reduces the number of paths, which constructed for ensuring the enough reliability. Also, a fewer path reduces the energy consumption and alleviates the restriction of network scale. Simulation results show that the proposed protocol provides the more reliable performance than the existing protocols in the network, which restricted scale, and consumes the less energy.

In wireless sensor networks, sink mobility support is one of the essential functionality in many applications. With continuous advancement, the future applications will require not only sink mobility but also high performance. Multipath routing is one of the promising technologies for improving network performance. However, existing multipath routing protocols have not dealt with the sink mobility. As a result, they lead to bad performance in terms of energy efficiency due to the end-to-end path reconstruction. Consequently, a novel path management scheme is required thereby supporting sink mobility without performance degradation. In this paper we propose a multipath management scheme for supporting sink mobility. The proposed scheme dynamically constructs multipath along the moving path of a sink. In addition, the proposed scheme provides the path shortening mechanism according to the sink's movement for reducing energy consumption. Our simulation results show that the proposed scheme is superior to existing path management schemes in terms of reliability and energy efficiency.

In wireless sensor networks, one of the most important application is tracking diffusive and large-scale phenomena, called continuous object. A number of sources from the large size of a continuous object bring huge communication overhead that former studies have to deal with. Meanwhile, recent application scenarios, such as forest fire suppression by firefighters, require sink mobility support. Most sink mobility support schemes have proposed for individual object tracking. They provide one-to-one connection between a mobile sink and a source. That is, too many connections resulting excess energy consumption are built if continuous object tracking directly adopts them. In this paper, we propose a solution to the problem with building a one-to-many connection between a sink and sources detecting a continuous object. The proposed scheme gradually extends the initial connection between a sink and the first source according to the spread of the continuous object. Our simulation results show that the proposed scheme is superior to existing schemes in terms of energy consumption.

Opportunistic routing (OR) has been emerged as a promising routing strategy to improve transmission reliability for wireless networks. By taking broadcast nature of wireless medium, a set of forwarding candidate nodes receive a data packet simultaneously, so one of them can relay it unless they all fail to receive. This strategy can dramatically improve one-hop reliability even in unreliable wireless links, however, it cannot be directly applied into large-scale wireless multi-hop networks since multi-hop communication significantly decreases end-to-end reliability. In this letter, we propose a novel opportunistic routing scheme for large-scale wireless multi-hop networks, which adopts multipath based routing in order to improve end-to-end reliability. While taking advantages of opportunistic routing, each intermediate node can determine whether it branches into multiple paths or not without any network topology information, so that a desired delivery ratio from a source to a destination given by application demands can be achieved in a fully localized manner.

Real-time applications in wireless sensor networks require that packets should reach to a sink within the time deadline. To satisfy the requirements, existing studies exploit nodes whose speed is higher than the desired delivery speed decided by the spatiotemporal approach. However, they fail real-time transmission in the area where there is no neighbor nodes which meet the requirements toward a sink since the desired delivery speed is fixed. This paper proposes a scheme to pass the area using time gain which occurs by the nature of the existing real-time transmission scheme. Thus, in the area, a sending node could make full use of nodes which do not satisfy the requirements. In the end, the packet passes across the area with a lower speed than the desired delivery speed. The simulation results show that the proposed scheme is superior in terms of real-time transmission success ratio to the existing schemes.