The mobile crowdsensing strategy emerges as a novel and trendy approach that arises from collecting a wide range of physical information in smart cities. In this context, assessing the quality of sensed data becomes essential to ensure that this strategy can be efficiently executed. Typically, popular data assessment strategies have focused on participants’ historical data as well as social contexts, or through online feedback to adjust quality biases. However, when faced with sensing data from more hostile environments, where sufficient supporting information is lacking and data is highly privatized, conventional assessment strategies may appear weak. Therefore, this article introduces an innovative approach by integrating social psychology to address this issue, proposing a data quality assessment strategy termed the “inspector sense model.” In this model, the responsibility for evaluating data quality is assigned to each participant, who serves as both a contributor of sensed data and an evaluator of others’ data quality. Further, this article introduces a probabilistic statistical model to ensure the high reliability of the assessment results. In addition, this article proposes an incentive mechanism called the “All Pay Auction” to ensure that this strategy can be implemented within the cost budget constraints. In particular, the unreliability of submission locations in the uploaded data is cleverly addressed by applying the proposed assessment strategy in reverse, significantly improving the overall data quality assessment capability of the strategy. During the experimental stage, simulation experiments were carried out utilizing temperature observations sourced from the real-world data provided by taxi drivers in Rome, and various performance metrics were analyzed. The results of these experiments demonstrate the capability of the proposed solutions to accurately assess data of unreliable quality, yielding highly precise assessment results.

Deep space internet of things (DS-IoT) aims to establish an intelligent space-based interconnected network covering the Mars and beyond. The deployment of DS-IoT relies critically on high-precision navigation technologies. The pulsar/Doppler integrated navigation can provide DS-IoT satellites with high-accuracy position and velocity information. However, the measured spectral bands are narrow, and Doppler shift can introduce truncation effects that compromise accurate Doppler velocity estimation, thereby degrading navigation accuracy. We analyze spectral distortion from truncation-induced nonlinear phase differences and propose a Fourier domain phase estimation method using ConvMixerCNN-Vision Transformer optimized sparse representation (FPE-CViT-SP). Our approach: (1) constructs a phase shift dictionary via DCT-based spectrum decomposition and shifted reconstruction; (2) formulates a sparse representation objective for optimal spectrum-dictionary matching; (3) employs a ConvMixerCNN-Vision Transformer to efficiently minimize the objective, replacing computationally intensive cross-correlation searches; and (4) applies sparse code-based super-resolution for precise Doppler estimation, effectively addressing nonlinear phase shifts. This method effectively addresses nonlinear phase shift issues through sparse representation characteristics. Furthermore, we develop a pulsar/Doppler integrated navigation system for DS-IoT satellites that significantly enhances collective positioning accuracy through inter-satellite link-based navigation data sharing. Comparative experiments demonstrate that our method achieves 32.17% and 35.29% improvement in position and velocity estimation accuracy respectively over X-ray pulsar navigation.

Visual Mobile Crowd-Sensing (VMCS) collects photos by leveraging camera embedded in mobile users’ phones. There are two important issues in VMCS: determining whether photos collected by mobile users meet the requirements or not and designing an appropriate mechanism to attract mobile users to provide photos that meet the requirements. In this article, we address those two issues when VMCS is applied to emergency monitoring applications. We first model an emergency scene as a disk region and define a coverage angle metric that quantifies the coverage ratio provided by each photo, then formulate a Maximize Coverage Angle with Limited Budget problem. The goal of this work is to recruit mobile users to provide panoramic view coverage for a disk while the total reward paid to participants does not exceed the budget. In our solution, we first propose a Coverage Angle Computation algorithm to calculate the coverage angle of each uploaded photo. Then two incentive mechanisms—the Guidance-based Incentive Mechanism and the Coverage Prediction Incentive Mechanism—are designed to encourage mobile users to upload photos with a coverage angle that are not provided by other mobile users. Finally, we design a mobile app called I-share in the Android system to implement the system. Meanwhile, we recruited students to install I-share and simulated the information interaction between mobile users and the server. We conducted experiments by using I-share without and with an embedded Coverage Angle Computation algorithm to validate the efficiency of the two incentive mechanisms. The experiment results demonstrate that our proposed incentive mechanisms effectively attract mobile users to provide panoramic view coverage of emergency scenes when the budget allows. Additionally, the Coverage Prediction Incentive Mechanism outperforms the Guidance-based Incentive Mechanism, offering a higher coverage ratio with lower rewards.

In the future manned lunar exploration mission, astronauts would work with the lunar robots, which has a high requirement for human–robot interaction (HRI). As the accuracy of gesture recognition interaction does not fulfill the requirement for human–robot joint exploration missions, we propose the DMS-SK/BLSTM-CTC hybrid network to improve the performance of HRI. For gesture recognition, considering VGG-SK has low accuracy and complex architecture, we delete the fourth convolution module, optimize the last global pooling layer, introduce dilated convolution block and multiscale convolution block in VGG-SK, and get the DMS-SK-based gesture recognition sub-network. Compared with the traditional recognition methods, the accuracy and performance of DMS-SK improve. For speech recognition, considering that Bidirectional long–short-term memory unit (BLSTM) has the advantages of processing temporal information, and the Connectionist Temporal Classification (CTC) algorithm can simplify speech data preprocessing, we use BLSTM based on CTC as the speech recognition sub-network. Finally, we combine DMS-SK with BLSTM-CTC, and propose the DMS-SK/BLSTM-CTC hybrid network as the gesture/speech hybrid network. In addition, we use 10 gestures in the American Sign Language (ASL) dataset and 10 speech commands to construct the gesture/speech hybrid dataset. Experimental results show that compared with the pure gesture or pure speech networks, the recognition accuracy of the gesture-speech hybrid network improves by 2% and 12%, respectively, its accuracy reaches 97.38%, which fulfills the requirement of astronauts for HRI.

In this paper we propose a new salient object detection method via structured label prediction. By learning appearance features in rectangular regions, our structural region representation encodes the local saliency distribution with a matrix of binary labels. We show that the linear combination of structured labels can well model the saliency distribution in local regions. Representing region saliency with structured labels has two advantages: 1) it connects the label assignment of all enclosed pixels, which produces a smooth saliency prediction; and 2) regularshaped nature of structured labels enables well definition of traditional cues such as regional properties and center surround contrast, and these cues help to build meaningful and informative saliency measures. To measure the consistency between a structured label and the corresponding saliency distribution, we further propose an adaptive label ranking algorithm using proposals that are generated by a CNN model. Finally, we introduce a K-NN enhanced graph representation for saliency propagation, which is more favorable for our task than the widely-used adjacent-graph-based ones. Experimental results demonstrate the effectiveness of our proposed method on six popular benchmarks compared with state-of-the-art approaches.