Recent advances in vision language models (VLMs) have demonstrated remarkable potential in embodied navigation tasks. However, existing robot-centric datasets primarily focus on traditional 3D tasks such as perception and prediction, lacking adequate support for vision-language tasks. Vision-language-navigation (VLN) is a key capability for achieving human-like and interpretable navigation in complex environments. In this study, we present CoT-VLNBench, the first large-scale benchmark and dataset designed for chain-of-thought (CoT) reasoning in quadruped robot navigation. Our dataset encompasses a diverse range of indoor and outdoor scenes, multi-step navigation trajectories, and rich natural language instructions, all annotated with fine-grained CoT reasoning traces. Specifically, it contains 175K frames, 5.25M 3D bounding boxes, and 875K vision–question–answer (VQA) pairs. This comprehensive resource enables thorough evaluation of embodied agents’ perceptual and step-by-step reasoning abilities. Furthermore, we propose a novel CoT-VLN model, a state-of-the-art 7B VLN model that integrates visual, linguistic, and reasoning modules, to facilitate interpretable and effective navigation. Extensive experiments demonstrate that our approach significantly outperforms existing non-VLMs baselines on the new benchmark, underscoring the importance of CoT-VLN in embodied navigation. We hope that CoT-VLNBench will serve as a valuable resource to advance research at the intersection of robotics, vision, language, and reasoning.

With the advancement of smart terminals and communication technologies, the emergence of heterogeneous Service Subscribers (SSs) and diverse sensing demands has facilitated the development of multi-platform Mobile CrowdSensing (MCS) scenarios. However, unlike traditional single-platform scenarios, Mobile Users' (MUs) bidding privacy is hard to protect in multi platform MCS. Additionally, the privacy disclosure issue of SSs has not been well addressed. To tackle these issues, in this paper, we propose a bilateral, auction-based scheme to preserve bidding privacy in multi-platform MCS, thereby protecting the interests of both SSs and MUs. Specifically, since SSs and MUs strategically choose one another to maximize their utility, we construct the corresponding selection processes for both sides by taking advantage of auction pricing theory. We firstly design a user-oriented forward auction that integrates the 0-1 knapsack problem with the Paillier encryption algorithm to protect the bidding information of both SSs and MUs. Then, we employ the Chinese Remainder Theorem (CRT) to design a reverse auction that hides the bidding behaviors of MUs. Theoretical analysis demonstrates that our scheme can protect the bidding privacy of both parties while ensuring economic robustness. Extensive experiments on a real dataset demonstrate that, compared with existing works, our scheme enables both SSs and MUs to achieve satisfactory utility while maintaining low computational overhead.

The Internet of Vehicles (IoV) generates massive sensitive perception data, typically managed by manufacturer-specific domains. While encryption with domain-specific parameters protects confidentiality, many IoV applications require secure cross-domain data sharing to access complementary information, and expressive keyword search for efficient access. However, existing Attribute-Based Keyword Search (ABKS) schemes are designed for single-domain settings, and thus cannot address heterogeneous key management or provide traceability without a universally trusted authority. To address these issues, we propose TCroS, a traceable cross-domain data sharing scheme that generalizes CP-ABE via proxy re-encryption mechanism, enabling ciphertexts generated in one domain to be securely transformed for authorized requesters in another. To provide traceability, TCroS embeds requester identities into decryption keys using Boneh-Boyen signatures, allowing any party (rather than the universally trusted authority) to trace the source of a leaked key. We further extend TCroS to TCroSS, which incorporates privacy-preserving expressive keyword search supporting Boolean queries, thereby enabling efficient retrieval of authorized data while resisting keyword guessing attacks. Formal security analysis proves that our schemes achieve IND-SCPA and IND-SCKA security. Experimental results demonstrate their practicality, showing that cross-domain sharing can be realized with computation and storage overheads comparable to single-domain setting.

In Unmanned Aerial Vehicle (UAV) networks, UAVs usually perform tasks in the form of groups. When tasks change, the Ground Control Station (GCS) will assign the complemental UAV to join the group for notification or reinforcement. Since UAVs communicate over open wireless channels, secure authentication is required for complemental UAV joining the group. However, one-by-one authentication between complemental UAV and the group members leads to high overhead and delays. At the same time, when the UAV group is far away from the coverage of the GCS, the GCS is unable to assist the authentication process in real-time. To solve the above problems, we propose a one-to-many UAV authentication scheme using Identity-Based Broadcast Encryption (IBBE) and batch authentication. This scheme does not require a trusted third party to be online in real time. We also design an algorithm based on reinforcement learning for identifying illegal requests during batch authentication, enhancing efficiency and ensuring successful authentication. Our scheme meets UAV networks’ security requirements, defending against various attacks. Experimental results show that it reduces computational overhead by 55.27% and communication overhead by 23.16% compared to similar schemes. Additionally, the illegal request identification algorithm reduces identification numbers by 15.44% to 25.72% and lowers latency by 14.64% to 25.12% compared to existing methods.

With the popularity of blockchains, low transaction throughput has become a significant bottleneck in applications such as cryptocurrencies. Payment channel networks (PCNs) have received attention as a way to improve throughput. However, due to the difficulty of predicting future transactions for nodes, the transactions are prone to failure when the channel balances do not meet required conditions. It has been shown that increasing buffers (queues) in PCNs can increase the success rate of transactions and throughput. Nevertheless, there is no effective transaction scheduling strategy in buffers when transaction values are flexible and variable. To solve this problem, we first formulate the Scheduling Problem in PCNs (named PSP), and then prove it is NP-hard. We design a neural network solver based on the Sequence to Sequence (Seq2Seq) architecture and train the solver using the reinforcement learning method. With the solver, we first give two scheduling strategies to maximize transaction throughput, and then design a PCN simulator for performance evaluation. Extensive experiments are conducted to show the superiority and various performances of our proposal and illustrate that our proposal can get a significant advantage in terms of the transaction throughput compared to the existing works.

Unmanned Aerial Vehicle (UAV)-assisted IoT networks are receiving a lot of attention in academia and industry. For instance, a UAV can fly and hover over sensors, during which time the sensors simultaneously initiate batch access requests to the UAV. Typically, UAV employs batch authentication to efficiently handle these batch accesses. However, an attacker can initiate illegal requests, causing batch authentication to fail. There are various batch identification algorithms to find illegal requests, enabling legitimate sensors to establish service connections quickly. Existing work wants to choose a suitable one based on the specific attack scenario. However, existing work assumes that the percentage r% of illegal requests is known in advance, which is impractical in real-world scenarios. Besides, existing work only selects a suitable batch identification algorithm based on r%, limiting the performance of batch identification to the capabilities of the alternative algorithms. Drawing inspiration from the Kalman filter, we first propose an adaptive estimation algorithm for the number of illegal requests to address the above problems. Based on the estimated value e%, we design a combinatorial batch identification using reinforcement learning. This approach allows the combination of different algorithms to achieve superior performance. Extensive experiments demonstrate that, for the estimation algorithm, the relative error is less than 20% in 27 out of 40 experiments. Regarding the combinatorial algorithms, the delay can be reduced by approximately 7.15% to 30.86% compared to existing methods.

UAV-assisted mobile edge computing (UAV-MEC) has been proposed to offer computing resources for smart devices and user equipment. UAV cluster aided MEC rather than one UAV-aided MEC as edge pool is the newest edge computing architecture. Unfortunately, the data packet exchange during edge computing within the UAV cluster hasn't received enough attention. UAVs need to collaborate for the wide implementation of MEC, relying on the gossip-based broadcast protocol. However, gossip has the problem of long propagation delay, where the forwarding probability and neighbors are two factors that are difficult to balance. The existing works improve gossip from only one factor, which cannot select suitable forwarding probability and avoid redundant messages. Besides, these schemes do not consider the historical packet reception of new neighbors when UAVs fly around, which decreases forwarding efficiency. To solve these problems, we first propose a data structure called Bitgraph that can record the historical packet reception of UAVs. Then, we formulate gossip broadcasting as a partially observable Markov decision process. Based on Bitgraph, we design the reward function. Finally, we design a multi-agent reinforcement learning algorithm, Branching Deep Graph Network (BDGN), which simultaneously makes decisions on forwarding probability and neighbors. Extensive experiments illustrate that our proposal gets more than 29% advantage in terms of the propagation delay and 20% advantage in terms of the redundant messages compared to the existing works.

The utilization of blockchain technology as a distributed information sharing system has gained widespread adoption across various domains. However, its application to Flying Ad-Hoc Network (FANET), characterized by severe packet loss, poses significant challenges. The high packet loss rates in FANETs can result in decreased consensus success rates and negatively impact information sharing consistency and efficiency. In this paper, we proposed RoUBC, a novel consensus scheme for Flying Ad-Hoc Networks (FANET), which is based on the Raft protocol and is designed to address the challenges posed by the severe packet loss network in FANET. The proposed scheme consists of two phases: leader election and block consensus. In the leader election phase, we integrate multi-criteria decision-making and link prediction algorithms to design an efficient stable-leader election method. In the block consensus phase, we propose a dynamic block verification algorithm based on historical verification information to achieve efficient block consensus. Our theoretical analysis demonstrates that the proposed consensus protocol is safe and live, effectively ensuring the consistency of message sharing in FANET. Experiment results show that our scheme outperforms traditional Raft schemes, with 35% increase in consensus success rate and 25% improvement in consensus efficiency.

Monocular depth estimation is an ill-posed problem as the same 2D image can be projected from infinite 3D scenes. Although the leading algorithms in this field have reported significant improvement, they are essentially geared to the particular compound of pictorial observations and camera parameters (i.e., intrinsics and extrinsics), strongly limiting their generalizability in real-world scenarios. To cope with this challenge, this paper proposes a novel ground embedding module to decouple camera parameters from pictorial cues, thus promoting the generalization capability. Given camera parameters, the proposed module generates the ground depth, which is stacked with the input image and referenced in the final depth prediction. A ground attention is designed in the module to optimally combine ground depth with residual depth. Our ground embedding is highly flexible and lightweight, leading to a plug-in module that is amenable to be integrated into various depth estimation networks. Experiments reveal that our approach achieves the state-of-the-art results on popular benchmarks, and more importantly, renders significant generalization improvement on a wide range of cross-domain tests.

In a 5G network-sliced environment, mobility management introduces a new form of handover called inter-slice handover among network slices. Users can change their slices as their preferences or requirements vary over time. However, existing handover-authentication mechanisms cannot support inter-slice handover because of the fine-grained demand among network slice services, which could cause challenging issues, such as the compromise of service quality, anonymity, and universality. In this paper, we address these issues by introducing a fast and universal inter-slice (FUIS) handover authentication framework based on blockchain, chameleon hash, and ring signature. To address these issues, we introduce an anonymous service-oriented authentication protocol with a key agreement for inter-slice handover by constructing an anonymous ticket with the trapdoor collision property of chameleon hash functions. In order to reduce the computation overhead of the user side in the process of authentication, a privacy-preserving ticket validation with a ring signature is designed to finish in the consensus phase of the blockchain in advance. Thanks to the edge computing capabilities in 5G, distributed edge nodes help to store the anonymous ticket information, which guarantees that the legal users can finish authentication swiftly during handover. Our scheme's performance is evaluated through simulation experiments to testify the efficiency and feasibility in a 5G network-sliced environment. The results show that compared to other authentication schemes of the same type, the overall inter-slice handover delay has been reduced by 97.94%.

The Deep learning of optical flow has been an active area for its empirical success. For the difficulty of obtaining accurate dense correspondence labels, unsupervised learning of optical flow has drawn more and more attention, while the accuracy is still far from satisfaction. By holding the philosophy that better estimation models can be trained with betterapproximated labels, which in turn can be obtained from better estimation models, we propose a self-taught learning framework to continually improve the accuracy using self-generated pseudo labels. The estimated optical flow is first filtered by bidirectional flow consistency validation and occlusion-aware dense labels are then generated by edge-aware interpolation from selected sparse matches. Moreover, by combining reconstruction loss with regression loss on the generated pseudo labels, the performance is further improved. The experimental results demonstrate that our models achieve state-of-the-art results among unsupervised methods on the public KITTI, MPI-Sintel and Flying Chairs datasets.

Recent work has shown that optical flow estimation can be formulated as a supervised learning problem. Moreover, convolutional networks have been successfully applied to this task. However, supervised flow learning is obfuscated by the shortage of labeled training data. As a consequence, existing methods have to turn to large synthetic datasets for easily computer generated ground truth. In this work, we explore if a deep network for flow estimation can be trained without supervision. Using image warping by the estimated flow, we devise a simple yet effective unsupervised method for learning optical flow, by directly minimizing photometric consistency. We demonstrate that a flow network can be trained from end-to-end using our unsupervised scheme. In some cases, our results come tantalizingly close to the performance of methods trained with full supervision.

Multi-shot person Re-IDentification (Re-ID) has recently received more research attention as its problem setting is more realistic compared to single-shot Re-ID in terms of application. While many large-scale single-shot Re-ID human image datasets have been released, most existing multishot Re-ID video sequence datasets containonly a few (i.e., several hundreds) human instances, which hinders further improvement of multi-shot Re-ID performance. To this end, we propose a deep cross-modality alignment network, which jointly explores both human sequence pairs and image pairs to facilitate training better multi-shot human Re-ID models, i.e., via transferring knowledge from image data to sequence data. To mitigate modality-to-modality mismatch issue, the proposed network is equipped with an image-to-sequence adaption module called cross-modality alignment sub-network, which successfully maps each human image into a pseudo human sequence to facilitate knowledge transferring and joint training. Extensive experimental results on several multi-shot person Re-ID benchmarks demonstrate great performance gain brought up by the proposed network.

We consider the problem of finding feature correspondences among a collection of feature sets, by using point-wise unary features. This is fundamental in computer vision and pattern recognition, which also relates to areas e.g. operational research. Different from two-set matching which can be transformed to a quadratic assignment programming task that is known NP-hard, inclusion of merely unary attributes leads to a linear assignment problem. This problem has been well studied and there are effective polynomial global optimum solvers such as the Hungarian method. However, it becomes ill-posed when the unary attributes are (heavily) corrupted. The global optimal correspondence concerning the best score defined by the attribute affinity/cost between the two sets can be distinct to the ground truth correspondence since the score function is biased by noises. To combat this issue, we devise a method for matching a collection of feature sets by synergetically exploring the information across the sets. In general, our method can be perceived from a (constrained) clustering perspective: in each iteration, it assigns the features of one set to the clusters formed by the rest of feature sets, and updates the cluster centers in turn. Results on both synthetic data and real images suggest the efficacy of our method against state-of-the-arts.

Distributed power control schemes have been intensively studied in the literature for uplink transmissions in cellular networks as well as in ad hoc networks. In the schemes with active link protection, the signal to interference plus noise (SINR) requirements of the new users are gradually approached without violating the existing links. In this paper, we consider a downlink cellular scenario, in which new nomadic cells seek admission to the network. A distributed power control algorithm with active cell protection is presented, where a cell is said to be active if it has sufficient resources to support the connected users and, otherwise, it is said to be inactive. With the proposed algorithm, inactive cells lower their loads by gradually performing power ramping, while active cells scale their transmission power accordingly, to avoid being overloaded. We prove the active cell protection property and compare the convergence of the algorithm under different interference assumptions. Further, we present an algorithm for adapting the power ramping factor in power limited scenarios for further performance enhancements.

This paper presents an optimization framework for energy savings in nomadic relay networks, where interference and load are changing over time due to varying assignments. We prove the existence of an explicit load function taking assignments as arguments and we show the function is continuously differentiable. Based on the properties of the load function, we design iterative relay and user association algorithm for energy savings, where the non-convex load constraints are linearly approximated such that each sub-problem is a linear program and hence can be solved efficiently. Simulation results confirm that our proposed algorithmic approach leads to a significant reduction of the energy consumption when compared with algorithms considering the worst-case interference.

We present a statistical approach for estimating the interference coupling coefficients in an LTE network based on a set of various measurements available at the network and terminal level. The proposed approach combines the measurements with prior information (spatial correlation among interference links) and takes into account measurement uncertainty. The result is a simple closed-form estimator that allows for fast realtime interference estimation.

This paper presents an optimization framework for energy-aware relay selection and user association in cellular networks aided by nomadic relays. The framework of sparse optimization is used to minimize network energy consumption for desired service provisioning of the terminals. We show that some constraints in the underlying optimization are of quadratic form due to the assumption of relays with wireless backhaul links. Hence, previously proposed algorithms for activation of network elements with wired backhaul links are not applicable. In this paper, therefore, novel algorithms based on different relaxations of the quadratic constraints are proposed and evaluated for energy savings. Simulation results confirm that the proposed algorithms may significantly reduce the overall energy consumption of cellular networks compared with conventional cell selection schemes.

Next generation mobile and wireless communication systems beyond 2020, aka Fifth Generation (5G) systems, aim at providing ubiquitous user experience with the utmost in quality. One of the promising technologies targeted for 5G systems is the flexible network deployment based on nomadic nodes (NNs). An NN is a low-power movable access node that provides coverage extension and capacity improvement on demand. Yet, NNs require flexible backhaul. One possible cost-efficient realization for flexible backhaul is in-band relaying. In this context, the capacity of the wireless backhaul link between an NN and its serving base station (BS) has a crucial role in the achievable end-to-end performance. The flexible backhaul can be exploited by dynamic NN selection to overcome the limitations of the backhaul link and, thus, to enhance the system performance. To this end, dynamic NN selection is carried out via selecting the serving NN from a set of available candidates considering the signal-to-interference-plus-noise ratio (SINR) on the backhaul link. In this regard, coarse NN selection takes into account only shadowing. Nevertheless, as NNs are stationary or slowly moving during operation, the wireless channels pertaining to NNs are usually subject to simultaneous impairments by both shadowing and multi-path fading, i.e., composite fading/shadowing. In this paper, we present the performance of coarse NN selection in composite fading/shadowing environments with co-channel interference. Further, we evaluate the performance in terms of backhaul link SINR, link rates, and end-to-end rate. Results show that coarse NN selection can yield high performance improvements.

Modern vehicles will have strong requirements with regard to seamless mobility support in future cellular systems, in order to enable advanced cooperative driver assistance and infotainment systems that guarantee traffic safety and efficiency. In this work, we introduce street-specific handover parameters for vehicular terminals. In particular, we propose an adaptive optimization algorithm that exploits vehicle context information in order to tune the handover parameters. Simulation results confirm that the proposed concept has the potential to improve handover performance significantly.

Deploying relay nodes is foreseen a cost-efficient solution to combat the severe propagation loss at cell edge. However, relay cell coverage is limited by the low transmit power, limited antenna capabilities and wireless backhaul link bottleneck which may lead to load imbalances and hence low resource utilization efficiency. Further challenges in relay deployments are attributed to increased interference levels in the network compared with macrocell-only deployments, causing degradation of the user throughput. In this context, relay cell coverage expansion and interference coordination techniques are expected to improve the performance of relay deployments. In this study, we analyze the impact of the additional interference due to the relay node transmissions. Jointly with our previous study on cell expansion, we show how the interference coordination mitigates the interference from relays and hence optimizes the system performance. Comprehensive system level simulations are carried out within the LTE-Advanced framework. Results confirm that the interference imposed by relay nodes can be reduced. Furthermore, both cell edge throughput gains and cell average throughput gains are observed.