In open underwater environments, ensuring accurate positions of sensors while protecting private information of localization systems presents a significant challenge. The physical channel differences between terrestrial and underwater networks render most existing privacy protection schemes designed for terrestrial networks inapplicable underwater. Moreover, limited research on underwater privacy protection has led to high implementation complexity and communication expenses. In this paper, to reduce the complexity of privacy protection, a secure mobile localization scheme using autonomous underwater vehicles (AUVs) as anchors is proposed for underwater sensor networks, based on adversarial neural cryptography utilizing acoustic channel features. Depending on whether eavesdroppers show interest in keys, two adversarial cryptography models are proposed to protect transmission of legitimate localization information and to actively counter eavesdroppers with learning capabilities in real time. Furthermore, to obtain effective keys and minimize unnecessary key transmission, random physical layer channel features are dynamically utilized as real-time keys for the cryptography system, and a synchronous channel probing protocol is designed for key generation. Simulation and experimental results demonstrate that, compared to other approaches, the proposed secure localization scheme effectively prevents the leakage of position information and maintains localization accuracy while operating at lower implementation complexity and communication expenses.

A persistent risk in many underwater localization solutions is the leakage of position information, since anchor positions are revealed to the sensors performing position estimation. While cryptography-based secure localization mechanisms for underwater acoustic sensor networks (UASNs) can protect the privacy of anchors, varying levels of privacy preservation impact localization accuracy. Therefore, it is necessary to derive a quantitative relationship between privacy preservation levels and localization accuracy. In this paper, we first model the probability density function (PDF) of anchor decoding errors under privacy preservation constraints when anchors adopt an encrypted system to safeguard position information. The model parameters are optimized using the expectation-maximization (EM) algorithm. Next, based on the ranging PDF and anchor decoding error PDF distribution, we derive the position error bound (PEB) function that links privacy preservation levels with target localization accuracy. Simulation and field experiment results validate the effectiveness of the theoretical model.

In harsh underwater environments, the localization of network nodes faces severe challenges due to open deployment environments. Most existing underwater localization methods suffer from privacy leaks. However, privacy protection schemes applied in terrestrial networks are not viable for underwater acoustic networks due to stratification effects and multipath complexities. In this paper, we introduce a secure localization scheme for underwater wireless sensor networks (UWSNs) utilizing cooperative beamforming among mobile underwater anchor nodes. With this scheme, the underwater sensor communicates and ranges with mobile anchor nodes to perform self-localization via time difference of arrival (TDOA) algorithm. However, the presence of eavesdroppers poses a threat by intercepting information emitted by the anchors. To avoid localization information leakage, then we model the secure localization requirement as a multi-anchors multi-objective dual joint optimization problem to enhance both security and energy performance. The deep reinforcement learning (DRL)-based multi-agent deep deterministic policy gradient (MADDPG) algorithm is applied to solve the optimization problem. Both simulation and field experimental results robustly validate the efficiency and accuracy of the proposed secure localization scheme.

In most existing underwater localization solutions, the risk of position information leakage persists, as anchor positions are revealed to the sensors performing position estimation. Conventional privacy protection methods designed for terrestrial networks often fail underwater due to the unique characteristics of the physical channels. Moreover, solutions proposed for the limited research on underwater privacy protection introduce expensive equipment costs and communication expenses. To tackle these challenges and reduce costs, this paper proposes a novel secure localization scheme tailored for underwater acoustic sensor networks (UASNs) based on adversarial neural cryptography utilizing acoustic channel features. The proposed scheme introduces an adversarial cryptography model to safeguard the transmission of legitimate localization data and dynamically counter eavesdroppers with learning capabilities in real-time. Furthermore, to obtain effective keys and minimize unnecessary key transmission, the scheme strategically employs channel features as cryptographic keys. Simulation results validate the effectiveness of this approach, demonstrating its ability to prevent the leakage of positional data, maintain localization precision, and operate with reduced hardware expenses and communication overhead compared to conventional methods.

In harsh underwater environments, position accuracy and privacy of sensors are equally important. Most underwater localization schemes suffer from privacy leakage, and privacy protection schemes for terrestrial networks are not applicable to underwater acoustic networks with stratification effects and multipath. Therefore, this paper proposes a secure localization scheme based on autonomous underwater vehicle (AUV) formation cooperative beamforming for underwater wire-less sensor networks (UWSNs) considering acoustic channel char-acteristic. The underwater sensor receives position information from multiple AUV anchors and utilizes time difference of arrival (TDOA) for self-localization. Eavesdroppers exist to overhear the information emitted by the anchors, thereby destroying the entire localization system. We model the secure localization requirement as a multi-AUV multi-objective dual joint optimization problem to optimize the security and energy performance, and adopt the deep reinforcement learning (DRL)-based multi-agent deep deterministic policy gradient (MADDPG) algorithm to solve it. Simulation and sea experimental results corroborate the efficiency and accuracy of the proposed secure localization scheme.

Abstract Using machine learning and deep learning methods to analyze text data from social media can effectively explore hidden emotional tendencies and evaluate the psychological state of social media account owners. However, the label noise caused by mislabeling may significantly influence the training and prediction results of traditional supervised models. To resolve this problem, this paper proposes a psychological evaluation method that incorporates a noisy label correction mechanism and designs an evaluation framework that consists of a primary classification model and a noisy label correction mechanism. Firstly, the social media text data are transformed into heterogeneous text graphs, and a classification model combining a pre-trained model with a graph neural network is constructed to extract semantic features and structural features, respectively. After that, the Gaussian mixture model is used to select the samples that are likely to be mislabeled. Then, soft labels are generated for them to enable noisy label correction without prior knowledge of the noise distribution information. Finally, the corrected and clean samples are composed into a new data set and re-input into the primary model for mental state classification. Results of experiments on three real data sets indicate that the proposed method outperforms current advanced models in classification accuracy and noise robustness under different noise ratio settings, and can efficiently explore the potential sentiment tendencies and users’ psychological states in social media text data.

Emerging autonomous underwater vehicles (AUVs) with small size, low speed, and low noise characteristics have shown promising applications in underwater monitoring. However, this also makes passive localization of underwater non-cooperative targets more challenging. Most localization schemes aim to improve localization accuracy while giving less consideration to additional related costs such as range flexibility, security, and energy consumption. To reduce such costs while ensuring accuracy, a weak-control AUV formation-based cooperative passive localization scheme is proposed, where AUVs do not require strict formation control or rely on anchor nodes to obtain absolute positions. Initially, there is a loose formation of multiple AUVs patrols until a target signal is passively and synchronously detected. Subsequently, the formation calibration is triggered. A rough formation calibration method is proposed based on signal interaction, where stratification effect compensation is considered. A precise formation calibration method adopting an inertial measurement unit and an extended Kalman filter algorithm is proposed to obtain a more accurate formation position by compensating for AUV drift error during rough calibration. Finally, the target position can be localized based on precise AUV formation using time difference of arrival method. Simulation and sea experimental results corroborate the efficiency and accuracy of the proposed localization scheme.

The accurate localization of nodes is one of the most basic tasks in underwater acoustic sensor networks (UASNs). However, in large-scale UASNs, nodes are difficult to be accurately located because GPS signals cannot be received and because underwater nodes cannot establish one-hop communication with beacons due to the limited transmission range. With the development of self-sinking beacon technology, in this article, we propose a mobile-beacon-based iterative localization (MBIL) mechanism to realize node hierarchically positioning of large-scale multihop UASNs with an aim at increasing the percentage of localized nodes and reducing the localization error in the network. The proposed mechanism first uses mobile beacon nodes to localize adjacent sensor nodes. Once the location information is obtained, the sensor node will calculate its confidence value to determine whether it is a qualified reference node. Then, the unknown nodes select three adjacent reference nodes with the highest evaluation index for localization, and the remaining unknown nodes are iteratively localized. Simulation results show that the proposed mechanism can not only achieve a higher proportion of localized nodes in a shorter time, but also effectively reduce the localization error. In addition, MBIL effectively balances the energy consumption of sensor nodes, which can prolong the network's lifetime.

Sina Weibo, the most popular Chinese social platform with hundreds and millions of user-contributed images and texts, is growing rapidly. However, the noise between the image and text, as well as their incomplete correspondence, makes accurate image retrieval and ranking difficult. In this paper, we propose a deep learning framework using visual features, text content and popularity of Weibo to calculate the similarity between the image and the text based on training the model to maximize the likelihood of the target description sentence given the training image. In addition, the retrieval results are reranked using the popularity of the image. The comparison experiment of the large-scale Sina Weibo dataset proves the validity of the proposed method.

With the rapid development of public service of wireless network, the quantities of free Access Points (APs) in open system authentication increase explosively. However, the security problems also raise prominent, especially, personal privacy disclosure brought by fake APs. We analyze the attack behaviors of the intruders. It is proved that open free AP can be attacked successfully by fake AP, compromised AP and spider AP. One type of fake AP attack (a software of airhack is developed) has been implemented to attack auto-reconnect method of legal AP and launched in real environment effectively. In order to ensure secure and trusted access, a new access method for AP in open system authentication is analyzed and improved to defense spider AP. This method is verified by the real typical instance of airhack.

In the underwater environment exist marine mammals and underwater wireless sensor networks (UWSNs), both of which use acoustic signals for communication. However, most of existing researches only focus on the performance, such as energy efficiency, packet error rates and propagation delay, of acoustic communication in UWSNs, which seems not friendly to marine mammals. With the purpose of sharing the scarce acoustic spectrum with marine mammals in a friendly way, an environment-friendly spectrum decision strategy is proposed in this paper. The strategy features a localization scheme for marine mammals and an environment-friendly channel allocation scheme. Based on the estimated locations of marine mammals, the transmission power and available channels are jointly allocated to the contending users. In this way, the proposed strategy can maximize the network capacity while avoiding interference with marine mammals. Simulation results show that the localization scheme has a success rate of localization around 90% and a relatively small average localization error at about 10m; the environment-friendly channel allocation scheme can work effectively as the locations of marine mammals change.

With the development of mobile internet, wireless communications via mobile devices, which is typically in the form of Delay Tolerant Networks (DTNs), becomes a hot research topic. One critical issue in the development of DTNs is the routing. Although there are a lot research works addressing routing issues in DTNs, they handle routing problem from only one or two aspects, which cannot produce an advanced solution to this comprehensive challenge. In view of these defects in the existing works, we propose a novel solution to address the routing issue of one type of DTNs in which mobile nodes can be divided into different clusters. A simple routing protocol can be used for the intra-cluster communication, while, for the inter-cluster one, messages will be forwarded to a relay node, which is selected by a new selection policy, supported by the network coding technique. The simulation results show that our proposed scheme can significantly improve the performance of the routing in DTNs.

In this paper, a novel face recognition algorithm named elastic block set reconstruction (EBSR) is proposed. In our method, the EBSR face is used to represent a set of training faces and to simulate different factors in a query image. An EBSR face is constructed by using the blocks from the training face images which best match to the blocks of the query image at the corresponding locations. The elastic local reconstruction (ELR) error is then used to evaluate how well a block pair matches, and the query image is classified based on the accumulated reconstruction error. The proposed method can effectively explore local information in the training set and deal with various conditions well. Also, the reconstruction error can be considered as a kind of dissimilarity measure, which gives a new approach to designing the training set so as to maximize robustness of recognition. Experiments show that consistent and promising results are obtained.

This paper proposes an innovative algorithm named Gabor Boost Linear Discriminant Analysis (GBLDA) for face recognition. In our method, we want to estimate the distribution of high dimensional Gabor wavelet (GW) features in a low dimensional LDA subspace without computing the GW feature of an input image. The computational complexity can be reduced significantly. Hence, GBLDA is suitable for real-time applications. Experimental results show that our proposed method not only possesses the advantages of linear subspace analysis approaches such as low computational complexity, but also has the advantage of a high recognition performance in the Gabor based methods.

Communication networks are traditionally assumed to be connected. However, in delay and disruption tolerant networks(DTN), there are many unconventional difficulties, such as intermittent connectivity, large delay and may never have an end-to-end contemporaneous path, a node has to store-and-carry messages for a long time, until a new forwarding opportunity arises. Because every node needs to store message, content storage becomes the core service of the DTNs, we can implement a content-based forwarding. This paper proposes a new data dissemination method, which classifies the forwarding messages by their content, every node only requests the message that it is interested in. So the messages transmitted in the network can be cut down largely. Of course, in order to improve delivery rate, nodes also request and store messages that requested by other nodes they have contacted with. Meanwhile, the paper adopts a buffer management scheme based on content popularity, a node manages its buffer depending on the times that messages are requested, when the buffer has no adequate capacity, firstly deletes the message that is requested the least. Simulation experiments illustrate that this method can provide low overhead while maintaining high delivery rate and low delivery latencies compared to epidemic routing.

In disruption tolerant networks, aggressive packet forwarding scheme like flooding has a major drawback in terms of network congestion. In this paper, we proposed a new routing algorithm, called adaptive randomized epidemic routing (ARER). ARER dynamically adjusts the forwarding probability for each message according to a new metric, replications density. Meanwhile, ARER arranges the forwarding sequence and the dropping priority based on their assigned weight. The weight is determined by the replication density, the delivery predictability, and TTL. An extensive simulation of ARER using various scenarios was carried out and its performance was compared to well known DTN routing protocols: epidemic routing, randomized routing and spray-and-wait routing. Our results show that ARER outperforms them in all scenarios in terms of packet delay and delivery.

Delay tolerant networks (DTNs) use a store-carry-forward communication model relying on the mobility of nodes because a persistent end-to-end path from source to destination generally does not exist. Epidemic routing is a typical protocol in which data can be replicated along multiple opportunistic paths. With the advent of network coding, it is intuitive that data can not only be replicated, but also coded, when the transmission opportunity arises. In this paper we have implemented the epidemic routing with network coding based on the qualnet platform to validate this theory. What's more, to obtain better performance we introduced adaptive scheduling mechanism. We analytically show that with network coding, the performances of the network including the delivery delay, throughput and the delivery rate have been improved, reflecting many realistic scenarios. And through the simulation result, the correctness of our analysis is confirmed. To sum up, our research will generate a significant impact on DTN, as one focus of future network.

Multi-dimensional signal processing usually requires the high-speed performance obtained from instruction-level parallelism. The Sobel edge detector algorithm can be classified as a typical multi-dimensional signal processing application since it is applicable to images represented in a two-dimensional space. The kernel of such algorithm consists of a nested loop with embedded conditional branches, which are used to determine the existence of an edge. The hardware implementation of a Sobel algorithm may require either extra execution time due to pipeline stalls or resource redundancy in order to handle the branch. This paper shows a new architecture design, which establishes, in advance, the outcome of the conditional branches and allows the execution of the loop with an apparent one hundred percent prediction accuracy.

The FARIMA (p,d,q) model is a good traffic model capable of capturing both the long-range and short-range behavior of a network traffic stream in time. In this paper, we propose a prediction-based admission control algorithm for an integrated service packet network. We suggest a method to simplify the FARIMA model fitting procedure and hence to reduce the time of traffic modeling and prediction. Our feasibility-study experiments showed that FARIMA models which have number of parameters could be used to model and predict actual traffic on quite a large time scale.

Previous traffic measurements have found the coexistence of both long-range and short-range dependence in network traffic. Therefore, models are required to predict traffic that has both long-range and short-range dependence. This paper provides a procedure to model and predict traffic using FARIMA (p,d,q) models. Our experiments illustrate that the FARIMA model is a good model and is capable of capturing the property of actual traffic. We provide guidelines to simplify the FARIMA model fitting procedure and thus to reduce the time of traffic modeling and prediction.