The impact of urban air pollution on the environments and human health has drawn increasing concerns from researchers, policymakers and citizens. To reduce the negative health impact, it is of great importance to measure the air pollution at high spatial resolution in a timely manner. Traditionally, air pollution is measured using dedicated instruments at fixed monitoring stations, which are placed sparsely in urban areas. With the development of low-cost micro-scale sensing technology in the last decade, portable sensing devices installed on mobile campaigns have been increasingly used for air pollution monitoring, especially for traffic-related pollution monitoring. In the past, some reviews have been done about air pollution exposure models using monitoring data obtained from fixed stations, but no review about mobile sensing for air pollution has been undertaken. This article is a comprehensive review of the recent development in air pollution monitoring, including both the pollution data acquisition and the pollution assessment methods. Unlike the existing reviews on air pollution assessment, this paper not only introduces the models that researchers applied on the data collected from stationary stations, but also presents the efforts of applying these models on the mobile sensing data and discusses the future research of fusing the stationary and mobile sensing data.

We address the problem of compressed sensing (CS) with prior information: reconstruct a target CS signal with the aid of a similar signal that is known beforehand, our prior information. We integrate the additional knowledge of the similar signal into CS via l ₁ -l ₁ and l ₁ -l ₂ minimization. We then establish bounds on the number of measurements required by these problems to successfully reconstruct the original signal. Our bounds and geometrical interpretations reveal that if the prior information has good enough quality, l ₁ -l ₁ minimization improves the performance of CS dramatically. In contrast, l ₁ -l ₂ minimization has a performance very similar to classical CS, and brings no significant benefits. In addition, we use the insight provided by our bounds to design practical schemes to improve prior information. All our findings are illustrated with experimental results.

A new convolution neural network (CNN) architecture for semantic segmentation of high resolution aerial imagery is proposed in this paper. The proposed architecture follows an hourglass-shaped network (HSN) design being structured into encoding and decoding stages. By taking advantage of recent advances in CNN designs, we use the composed inception module to replace common convolutional layers, providing the network with multi-scale receptive areas with rich context. Additionally, in order to reduce spatial ambiguities in the up-sampling stage, skip connections with residual units are also employed to feed forward encoding-stage information directly to the decoder. Moreover, overlap inference is employed to alleviate boundary effects occurring when high resolution images are inferred from small-sized patches. Finally, we also propose a post-processing method based on weighted belief propagation to visually enhance the classification results. Extensive experiments based on the Vaihingen and Potsdam datasets demonstrate that the proposed architectures outperform three reference state-of-the-art network designs both numerically and visually.

Training deep neural networks on large datasets containing high-dimensional data requires a large amount of computation. A solution to this problem is data-parallel distributed training, where a model is replicated into several computational nodes that have access to different chunks of the data. This approach, however, entails high communication rates and latency because of the computed gradients that need to be shared among nodes at every iteration. The problem becomes more pronounced in the case that there is wireless communication between the nodes (i.e., due to the limited network bandwidth). To address this problem, various compression methods have been proposed, including sparsification, quantization, and entropy encoding of the gradients. Existing methods leverage the intra-node information redundancy, that is, they compress gradients at each node independently. In contrast, we advocate that the gradients across the nodes are correlated and propose methods to leverage this inter-node redundancy to improve compression efficiency. Depending on the node communication protocol (parameter server or ring-allreduce), we propose two instances for the gradient compression that we coin Learned Gradient Compression (LGC). Our methods exploit an autoencoder (i.e., trained during the first stages of the distributed training) to capture the common information that exists in the gradients of the distributed nodes. To constrain the nodes' computational complexity, the autoencoder is realized with a lightweight neural network. We have tested our LGC methods on the image classification and semantic segmentation tasks using different convolutional neural networks (CNNs) [ResNet50, ResNet101, and pyramid scene parsing network (PSPNet)] and multiple datasets (ImageNet, Cifar10, and CamVid). The ResNet101 model trained for image classification on Cifar10 achieved significant compression rate reductions with the accuracy of 93.57%, which is lower than the baseline distributed training with uncompressed gradients only by 0.18%. The rate of the model is reduced by 8095x and 8x compared with the baseline and the state-of-the-art deep gradient compression (DGC) method, respectively.

Real-world data processing problems often involve various image modalities associated with a certain scene, including RGB images, infrared images or multi-spectral images. The fact that different image modalities often share certain attributes, such as edges, textures and other structure primitives, represents an opportunity to enhance various image processing tasks. This paper proposes a new approach to construct a high-resolution (HR) version of a low-resolution (LR) image given another HR image modality as guidance, based on joint sparse representations induced by coupled dictionaries. The proposed approach captures complex dependency correlations, including similarities and disparities, between different image modalities in a learned sparse feature domain in lieu of the original image domain. It consists of two phases: coupled dictionary learning phase and coupled super-resolution phase. The learning phase learns a set of dictionaries from the training dataset to couple different image modalities together in the sparse feature domain. In turn, the super-resolution phase leverages such dictionaries to construct a HR version of the LR target image with another related image modality for guidance. In the advanced version of our approach, multi-stage strategy and neighbourhood regression concept are introduced to further improve the model capacity and performance. Extensive guided image super-resolution experiments on real multimodal images demonstrate that the proposed approach admits distinctive advantages with respect to the state-of-the-art approaches, for example, overcoming the texture copying artifacts commonly resulting from inconsistency between the guidance and target images. Of particular relevance, the proposed model demonstrates much better robustness than competing deep models in a range of noisy scenarios.

Time-synchronized channel hopping (TSCH) is currently the most efficient solution for collision-free interference-avoiding communications in ad hoc wireless networks, such as wireless sensor networks, vehicular networks, and networks of robots or drones. However, all variants of TSCH require some form of centralized coordination to maintain the time-frequency slotting mechanism. This leads to slow convergence to steady state and moderate time-frequency slot utilization, particularly under node churn or mobility. We propose decentralized time-synchronized channel swapping (DT-SCS), which is a novel protocol for medium access control (MAC) in ad hoc wireless networks. Under the proposed protocol, nodes first converge to synchronous beacon packet transmissions across all available channels at the physical layer, with a balanced number of nodes in each channel. This is done by the novel coupling of distributed synchronization and desynchronization mechanisms-which are based on the concept of pulse-coupled oscillators-at the MAC layer. Decentralized channel swapping can then take place via peer-to-peer swap requests/acknowledgments made between concurrent transmitters in neighboring channels. We benchmark the convergence and network throughput of DT-SCS, TSCH, and the efficient multichannel MAC protocol (seen as the state of the art in decentralized, interference-avoiding, and multichannel MAC protocols) under simulated packet losses at the MAC layer. Moreover, performance results via a Contiki-based deployment on TelosB motes reveal that DT-SCS comprises an excellent candidate for decentralized multichannel MAC-layer coordination by providing for quick convergence to steady state, high bandwidth utilization under interference and hidden nodes, as well as high connectivity.