Search

A plethora of information is now readily available for traffic prediction, making an effective use of them enables better traffic planning. With data coming from multiple sources, and their features spanning spatial and temporal dimensions, there is an increasing demand to exploit them for accurate traffic prediction. Existing methods, however, do not provide a solution for this, as they tend to require expertise feature engineering. In this paper, we propose a general architecture for SpatioTemporal Data Fusion (STDF) with parameter efficiency. To make heterogeneous multi-source data fusion effectiveness, we separate all data into traffic directly related data and traffic indirectly related data. With traffic indirectly related data as the input to Spatial Embedding by Temporal convolutiON (SETON) that simultaneously encodes each feature in both space and time dimensions and traffic directly related data as the input to the graph convolutional network(GCN), we designed a fine-grained feature transformer to match the ones generated by GCN. This is then followed by a fusion module to combine all features to make final prediction. Compared to using GCNs training with only traffic directly related data, experimental results show that our model can achieve a 6.1% improvement in prediction accuracy measured by Root Mean Squared Error.

Limited by the bandwidth of IEEE 802.11p protocol, a large-size data block commonly cannot be transmitted within one communication link in a vehicular ad hoc network (VANET). When transmitting large-size data block in VANETs, the traditional routing-layer-based solutions usually employed a large number of routing paths to extend the data link duration through continuously changing routing path, which commonly leads to frequent communication link disconnection and re-connection. Another shortcoming of the routing-layer solutions in transmitting large-size data block is that dense relay node is required to obtain sufficient stable routing paths for maintaining the long data transmission duration. In this paper, we present a new scheme for efficiently transmitting large-size data block in VANETs, which dedicate to improve the data transmission efficiency through shortening the total duration of the data transmission process at the network architecture and physical connection levels. Based on the assistance of the dual-communication mode and edge computing technique, a heterogeneous network architecture is proposed to introduce multiband orthogonal frequency division multiplexing (MB-OFDM) based ultra-wideband (UWB) technique into the VANETs for transmitting large-size data block. The key challenges of the scheme and alternative solutions for the challenges are analyzed, in which several mechanisms for quick parameter pre-configuration, the quick MB-OFDM UWB link establishment and highly efficient multi-user access to the OFDM-UWB network are presented. The key obstacles that limit the effectiveness of the scheme are also evaluated, of which the results show the presented scheme is feasible.

With the exploding growth of smart devices and application of 5G communications, there are more and more computation-intensive tasks which are delay-sensitive. Mobile edge computing (MEC) is a new paradigm that provides rich computing resources in close proximity to mobile users. However, how to make the decision of whether offloading or not as well as control the transmission power jointly remains a challenge. In this paper, we first study the joint multiuser offloading and transmission power control optimization problem in a multi-channel wireless interference scenario. We formulate the joint optimization problem into a system-wide computation overhead minimization problem as a mixed integer non-linear program. We then propose an efficient semi-distributed algorithm consisting of two subalgorithms of offloading scheme and transmission power control. The numerical simulation results demonstrate that the proposed algorithm can effectively reduce system-wide computation overhead and increase the number of beneficial edge computing users, compared with all local or edge computing and fixed transmission power schemes.

The user operates the smart home devices year in year out, have produced mass operation data, but these data do not be utilized well in past. Nowadays, these data can be used to predict user’s behavior custom with the development of big data and machine learning technologies, and then the prediction results can be employed to enhance the intelligence of a smart home system. In view of this, this paper proposes a novel unsupervised user behavior prediction (UUBP) algorithm, which employs an artificial neural network and proposes a forgetting factor to overcome the shortcomings of the previous prediction algorithm. This algorithm has a high-level of autonomous and self-organizing learning ability while does not require too much human intervention. Furthermore, the algorithm can better avoid the influence of user’s infrequent and out-of-date operation records, because of the forgetting factor. Finally, the use of real end user’s operation records to demonstrate that UUBP algorithm has a better level of performance than other algorithms from effectiveness.

Vehicular Edge Computing (VEC) has been studied as an important application of mobile edge computing in vehicular networks. Usually, the generalization of VEC involves large-scale deployment of dedicated servers, which will cause tremendous economic expense. We also observe that the Parked Vehicles (PVs) in addition to mobile vehicles have rich and underutilized resources for task execution in vehicular networks. Thus, we consider scheduling PVs as available edge computing nodes to execute tasks, and this leads to a new computing paradigm, called by parked vehicle edge computing (PVEC). In this paper, we investigate PVEC and explore opportunistic resources from PVs to run distributed mobile applications. PVs coordinate with VEC servers for collective task execution. First, a system architecture with primary network entities is proposed for enabling PVEC. We also elaborately design an interactive protocol to support mutual communications among them with security guarantee. Moreover, we measure the availability of opportunistic resources and formulate a resource scheduling optimization problem by using Stackelberg game approach. A subgradient-based iterative algorithm is presented to determine workload allocation among PVs and minimize the overall cost of users. Numerical results indicate that compared with existing schemes, PVEC serves more vehicles and reduces service fee for users. We also demonstrate that the Stackelberg game approach is effective and efficient.

The high implementation complexity of multiband orthogonal frequency-division multiplexing (MB-OFDM) ultra-wideband (UWB) technology is the major hurdle for applying it to vehicular communications. This paper presents a comprehensive synchronizer with low complexity and high performance for the MB-OFDM UWB. A low-complex overall architecture is proposed, in which the sub-functions are divided into amplitude-detection-based functions and phase-detection-based functions. All of the amplitude-detection-based functions are implemented based on a simplified cross correlation-based matched filter, and a serial structure-based auto-correlation block is designed exclusively for carrier frequency offset estimation. Several effective methods for the sub-functions are proposed based on the proposed overall architecture. Evaluation results show that the proposed synchronizer has high performance with low complexity.