Search

Location prediction plays an important role in modeling human mobility. Existing studies focused on developing a prediction model which is based solely on the past mobility of only the person of interest (POI), rather than including information on the mobility of her/his companions. In fact, people frequently move in a group, and thus, using mobility data of a person’s companions can enhance accuracy when predicting that person’s future locations. Motivated by this, we propose a two-phase framework for predicting an individual’s future locations that fully benefits from spatio-temporal contexts embedded in that person’s and his/her companions’ mobility. The framework first determines the POI’s companions, then predicts future locations based on mobility information for both the POI and selected companions. Two companion selection methods are proposed in this work. The first method uses spatial closeness (SC) to determine the companions of the POI by measuring the similarity of the individuals’ geographic distributions. The second method builds person ID embedding (PIE) vectors, and cosine similarity is used to select the POI’s companions. To mitigate the curse of dimensionality, the framework also uses a stacked autoencoder in which the encoder compresses a high-dimensional input feature (e.g., location, time, and person ID) into a low-dimensional latent vector. For the second phase of the framework, a bidirectional recurrent neural network (BRNN)-based multi-output model is proposed to predict a person’s future locations in the next several time slots. To train the BRNN model, weighted loss is used, which takes into account the importance of each future time slot to predict the POI’s locations accurately. Experiments are conducted on two large-scale Wi-Fi trace datasets, demonstrating that the proposed model can effectively predict human future locations.

LoRa technology draws attention for its use in industrial monitoring and control systems in which each end device or task is required to send data periodically to a (cloud) server. Despite its provision of a stable link, it suffers from data loss by signal suppression and interference. A real-time LoRa protocol is proposed that uses a slot scheduling to remove collision and device or node grouping based on signal attenuation to deal with signal suppression. Based on the definition of a frame-slot structure, a logical slot indexing algorithm is devised to tag a logical index to each slot. The logical indices enable the easy allocation of slots to nodes such that if each node sends data in the allocated slots, it can satisfy time constraint. To handle external interference caused by other networks, the protocol uses a multiple listen-before-talk (mLBT) mechanism that allows channel detection multiple times within one slot. Our protocol is compared analytically and experimentally with other ones to show its superior throughput and reliability against signal suppression and interference.

In this work, the performance evaluation and the optimization of dual-hop LoRa network are investigated. In particular, the coverage probability (Pcov) of edge end-devices (EDs) is computed in closed-form expressions under various fading channels, i.e., Nakagami-$m$ and Rayleigh fading. The Pcov under Nakagami-$m$ fading is computed in the approximated closed-form expressions; the Pcov under Rayleigh fading, on the other hand, is calculated in the exact closed-form expressions. In addition, we also investigate the impact of different kinds of interference on the performance of the Pcov, i.e., intra-SF interference, inter-SF interference (or capture effect) and both intra- and inter-SF interference. Our findings show that the impact of imperfect orthogonality is not non-negligible, along with the intra-SF interference. Moreover, based on the proposed mathematical framework, we formulate an optimization problem, which finds the optimal location of the relay to maximize the coverage probability. Since it is a mixed integer program with a non-convex objective function, we decompose the original problem with discrete optimization variables into sub-problems with a convex feasible set. After that, each sub-problem is effectively solved by utilizing the gradient descent approach. Monte Carlo simulations are supplied to verify the correctness of our mathematical framework. In addition, the results manifest that our proposed optimization algorithm converges rapidly, and the coverage probability is significantly improved when the location of relay is optimized.

Most Driver Status Monitoring (DSM) systems consist of spatial features extractions and temporal status recognition in sequence. Extracting the spatial features, which include driver facial behavior such as eye closing and mouth opening, generally requires considerable computation. It causes DSM systems to lose the valuable instantly-occurring driver facial information during real-time processing. Loss of facial information affects the accuracy of the system, and its impact is more severe on restricted computing resources. To solve this problem, this paper proposes an Adaptive Batch-Image (ABI) based DSM (ABI-DSM) system. The ABI enables the DSM system to use images captured in real-time while the DSM process previous input images. For real-time operation on a lightweight GPU-equipped Single-Board Computer (SBC), the ABI-DSM system is designed as follows. First, the system uses the driver's facial behavior to reduce the dimension of the time-series data for recognizing the status of the driver. The second, detection and tracking of driver's faces are not used for facial behavior recognition. Also, the system works with PydMobileNet, which has lower parameters and FLOPs than MobileNetV2, for facial behavior recognition. Experiments show that the ABI-DSM systems based on MobileNetV2 and PydMobileNet perform better than others in terms of both FPS and Precision. In particular, the PydMobileNet-based ABI-DSM system outperforms competitors when the size of Batch-Image is over six images.

The vehicular ad hoc network is an emerging area of technology that provides intelligent transportation systems with vast advantages and applications. Frequent disconnections between the vehicular nodes due to high-velocity vehicles impact network performance. This can be addressed by efficient clustering techniques. Several recent studies have attempted to develop optimal clustering algorithms to improve network performance metrics using soft computing techniques. Although sufficient work on soft computing techniques has been carried out, it seems less commonplace to find an analysis of various algorithms’ network parameters together. This paper provides a systematic analysis of the clustering-based routing protocols used in vehicular networks that are aware of soft computing techniques. The categorization is performed according to various soft computing techniques: particle swarm optimization, k-means, neural networks, artificial bee colony, genetic algorithm, firefly algorithm, and fuzzy logic. A comparative study of soft computing strategies is also provided in the survey with a focus on their objectives, along with their strengths and limitations. This survey makes it easier for researchers to pick the required soft computing technique used in vehicular networks in order to improve metrics such as packet delivery ratio, end-to-end delay, throughput, cluster lifetime, and message overhead.

In recent years, intelligent transport system (ITS) applications with unmanned ground vehicles (UGV) have increased dramatically to improve the safety, efficiency, and unique services in low- to medium-range communication deployment. Orthogonal frequency division multiplexing (OFDM) is adopted as a modulation scheme for this communication system to combat fading. Unfortunately, because of possible high UGV speed, the orthogonality between subcarriers in OFDM signals is prone to intercarrier interference (ICI) caused by the Doppler spread. This paper discusses ICI suppression analysis using Universal Software Radio Peripheral (USRP) N210 under several Harbor transmission scenarios. Such communication links may contain line of sight (LoS) components in transmission. The channel is thus modeled as the well-known fast time varying path loss model. Furthermore, this paper derives the signal to interference ratio (SIR) and evaluates this parameter for several types of fading and Doppler spread shapes in UGV channels. Additionally, the paper presents a practical method for calculating the spectral efficiency of OFDM systems, with an example for actual implementation in UGV systems. Eventually, the testbed results indicate that the SIR will decrease if the maximum Doppler spread expands. In the Rician channel, it can be seen that the SIR depends on the maximum Doppler speed, the angle of arrival (AoA), and even the Rician element of the LoS component.

This paper describes a nonorthogonal multiple access (NOMA)-based device-to-device (D2D) communication system underlaying in a cellular network sharing time and frequency resource. To optimize the total transmit power for all users in the network, we investigate a comprehensive resource allocation solution consisting of joint subchannel assignments, user pairings, and power control. The optimization problem is a mixed integer linear program, which is solved in three low-complexity steps. We first group D2D users to minimize mutual interference from cellular users (CU). Second, we demonstrate a heuristic solution for coupling D2D users in each group. Finally, we minimize the total transmit power of all devices in the network while maintaining the SINR requirements for both CU and D2D users. The optimal solution is derived off-line using the Rayleigh fading channels for any given QoS level of CU and NOMA-D2D users. We recognize that both subchannel assignment and user pairings play the important roles in reducing the total transmit power of the network, in which both cellular and D2D users coexist. However, optimization of the subchannel assignment contributes more toward minimizing the total network transmit power than the proposed model of optimizing user pairings.