Under varying conditions, the stage evolution of tool wear exhibits disparate characteristics, with pronounced differences in wear rate and transition points across stages. This variability renders real-time, high-precision measurement and unified monitoring of tool wear exceptionally challenging. To address this issue, a data-distribution-based feature screening and three-stage physics-informed meta-learning are proposed in this paper for multi-condition tool wear unified monitoring. First, high-dimensional data acquired during the cutting process are exploited to construct a time–frequency feature matrix. Information criteria are employed to identify the distribution of tool wear increments. And features whose exhibit maximal multidimensional similarity to both the wear state and its distribution are selected via FSI. Then, a novel three-stage wear physical model is embedded into the monitoring model. These physics-based constraints, together with the measured data, jointly restrict the solution space. An improved meta-optimizer is subsequently adopted to distill the domain-invariant representations shared across multi-conditions. Finally, the monitoring model is rapidly adapted to a new condition with only a handful of samples, enabling real-time and accurate tool wear monitoring. A multi-condition wear experiment conducted on indexable CNC milling inserts demonstrated that, compared with state-of-the-art methods, the proposed method exhibits markedly higher precision, stability, and adaptability in new conditions.

Recent studies show that graph neural networks (GNNs) are vulnerable to backdoor attacks. Existing backdoor attacks against GNNs use fixed-pattern triggers and lack reasonable trigger constraints, overlooking individual graph characteristics and rendering insufficient evasiveness. To tackle the above issues, we propose ABARC, the firstAdaptiveBackdoorAttack withReasonableConstraints, applying to both graph-level and node-level tasks in GNNs. For graph-level tasks, we propose a subgraph backdoor attack independent of the graph's topology. It dynamically selects trigger nodes for each target graph and modifies node features with constraints based on graph similarity, feature range, and feature type. For node-level tasks, our attack begins with an analysis of node features, followed by selecting and modifying trigger features, which are then constrained by node similarity, feature range, and feature type. Furthermore, an adaptive edge-pruning mechanism is designed to reduce the impact of neighbors on target nodes, ensuring a high attack success rate (ASR). Experimental results show that even with reasonable constraints for attack evasiveness, our attack achieves a high ASR while incurring a marginal clean accuracy drop (CAD). When combined with the state-of-the-art defense randomized smoothing (RS) method, our attack maintains an ASR over 94%, surpassing existing attacks by more than 7%.

Crowdsensing has become a popular method of sensing data collection while facing the problem of protecting participants' location privacy. Existing location-privacy crowdsensing mechanisms focus on static tasks and participants without considering sensing tasks' time requirements and participants' mobility, which cannot achieve satisfactory collected data quality and task completion in crowdsensing with dynamic tasks and participants. Inspired by this, we proposed a location-preservation crowdsensing mechanism, FedSense, considering dynamic tasks and participants based on federated learning (FL) and reinforcement learning (RL). In FedSense, through RL's outstanding decision-making ability, participants select sensing tasks to perform by well-trained RL models without uploading location information to servers for task allocation. We propose an independent tasks selection environment that defines actions, states, and rewards of RL to enable FedSense to achieve satisfactory task completion and data quality while preserving location privacy. Besides, FedSense applies an asynchronous FL aggregation algorithm that reduces participants' network stabilization and device computing ability requirements. Analysis proves that participants' location information does not leave the local device during the model training and task selection process, effectively avoiding privacy leakage. Simulation shows that compared with existing location-preservation crowdsensing mechanisms, FedSense achieves the highest task completion and sensing accuracy for dynamic tasks and participants.

Reconstruction attacks against federated learning (FL) aim to reconstruct users’ samples through users’ uploaded gradients. Local differential privacy (LDP) is regarded as an effective defense against various attacks, including sample reconstruction in FL, where gradients are clipped and perturbed. Existing attacks are ineffective in FL with LDP since clipped and perturbed gradients obliterate most sample information for reconstruction. Besides, existing attacks embed additional sample information into gradients to improve the attack effect and cause gradient expansion, leading to a more severe gradient clipping in FL with LDP. In this paper, we propose a sample reconstruction attack against LDP-based FL with any target models to reconstruct victims’ sensitive samples to illustrate that FL with LDP is not flawless. Considering gradient expansion in reconstruction attacks and noise in LDP, the core of the proposed attack is gradient compression and reconstructed sample denoising. For gradient compression, an inference structure based on sample characteristics is presented to reduce redundant gradients against LDP. For reconstructed sample denoising, we artificially introduce zero gradients to observe noise distribution and scale confidence interval to filter the noise. Theoretical proof guarantees the effectiveness of the proposed attack. Evaluations show that the proposed attack is the only attack that reconstructs victims’ training samples in LDP-based FL and has little impact on the target model’s accuracy. We conclude that LDP-based FL needs further improvements to defend against sample reconstruction attacks effectively.

To ensure the machining quality of micro/nano scale structural units for meter scale workpieces, relay turning with multiple single-point diamond tools has been broadly required. However, the existing tool setting methods have the problems of long tool setting time and low tool setting accuracy. To address the above issues, a novel normalized variational auto-encoder model with an adaptive activation function (NVAE-AAF) is proposed in this article. The batch normalization and the adaptive activation function are introduced into the variational auto-encoder model to learn robust features of force signals at the tool idle move state. Then, the reconstruction error threshold is constructed according to the kernel density estimation method to realize the nanoscale tool setting. In the ultraprecision tool setting experiments based on piezoelectric ceramic force sensing, the reconstruction error of the force signals at the tool idle move state is less than 0.07%, and the contact detection accuracy reached 92%. Compared to the traditional trial cutting for tool setting method, the proposed method significantly improves tool setting accuracy by 75%–85%, reaching a level of 75 nm.

Users in dynamic spectrum access (DSA) with federated reinforcement learning (FRL) autonomously access channels, avoiding centralized coordination and protecting users’ privacy. However, existing FRL-based DSA mechanisms are limited to ideal network states, i.e., assuming that channel states and users’ interference relationships are unchanged. Besides, users should upload intermediate results simultaneously for federated aggregation. The above conditions are impractical for mobile users since their network states and locations are unstable. Meanwhile, newly connected users have to train their models through local data with numerous computing resources since global models are unsuitable for them. We propose FRDSA, an FRL-based secure and lightweight channel selection mechanism in DSA for mobile users under dynamic network states. An independent channel selection environment with a virtual group strategy is presented to avoid interference between users under unstable channel states. Furthermore, an asynchronous parameter aggregation method in FRDSA dynamically adjusts the aggregation factors without users simultaneously uploading intermediate results. Simulations based on real trajectory data show that FRDSA significantly reduces approximately 60% interference between mobile users under unstable network states. Newly connected users can directly apply the well-trained global model to access channels autonomously instead of retraining a model, effectively reducing mobile users’ computing resource requirements.

By separating the forwarding and control planes, Software-Defined Networking (SDN) facilitates flexible traffic routing and network management for a service network. Because of the impact of controller deployment on message transmission distances and network latency, controller placement problems have drawn many researchers’ attention. However, assumptions in most existing research that all control packets are either transmitted in the service network (i.e., in-band control) or through predetermined control-service connection (i.e., out-of-band control) are not reasonable due to bandwidth resources occupation on the service network or high construction costs. In this paper, we are the first to jointly discuss the controller placement and control-service connection problem for latency minimization in the hybrid-band control mode, which is essentially a bi-level programming optimization problem. Specifically, we introduce auxiliary variables to simplify the above NP-hard problem. Next, Generalized Benders decomposition is used to obtain an optimal solution in theory. In addition, we propose a time-efficient fireworks algorithm with a little latency increment for large-scale networks. Extensive evaluations show that the two proposed algorithms accomplish the desired objectives and respectively achieve up to 35% and 25% latency decrement than greedy algorithms.

Achieving higher spectrum utilization, auction-based mechanisms has been regarded as a popular tool in dynamic spectrum access (DSA). Recently, Sybil attacks in auction-based DSA mechanisms have been investigated, where a cheating bidder can manipulate an auction by submitting bids under multiple fake identities. Existing Sybil-proof mechanisms in DSA are limited to prevent Sybil attacks from primary users (PUs) or secondary users (SUs). However, both of PUs and SUs may perform Sybil attacks in DSA, i.e., double Sybil attacks. The challenge of solving the double Sybil attacks is that fictitious identities and fake bids can directly affect allocation results, but the malicious bidders cannot be straightforwardly distinguished from all bidders. To resist the double Sybil attacks, we propose STEAM, the first double Sybil-proof and two-dimensional Truthful spEctrum Auction Mechanism for DSA. Specifically, STEAM merges suspicious buyers based on geographic characteristics and sorts sellers by a bid-independent sorting method to minimize the impact of untruthful bids and Sybil attacks on the allocation results. Theoretical analysis and extensive evaluations prove that STEAM is double Sybil-proof, two-dimensional truthful, individual rational and budget-balanced, while the performance loss in various metrics within 8% compared to the existing auction-based mechanisms.

The location information for tasks may expose sensitive information, which impedes the practical use of mobile crowdsensing in the industrial Internet. In this article, to our knowledge, we are the first to discuss the privacy protection of task locations and propose a codebook-based task allocation mechanism to protect it. Considering the cost of system utility caused by privacy protection technology, the tradeoff between local privacy and system utility is formalized a multiobjective optimization problem. The optimal solution is theoretically derived, and the optimal task allocation scheme is obtained. In addition, the selected allocation codebook (SAC) method is introduced to solve the problem of high computational resource consumption in the task allocation process and protect the task location privacy to some extent. The experimental results show that the SAC method sacrifices system utility but improves the privacy protection for task locations by 60% on average.

Due to the pervasive adoption of sensor-embedded mobile devices yet increasing demand on data and computing resources, mobile crowdsensing is a promising paradigm with rapid growth. One of the most challenging issues is how to maximize the utilities of crowdsensing platforms under inaccurate distributed sensing. The nature of such inaccuracy is due to the fact that energy-based sensing can be greatly impacted by thermal and environmental noise, which significantly affects task allocation strategies of crowdsensing platforms. Because of the allocation efficiency and fairness concerns, auction-based mechanisms have been extensively used in crowdsensing systems. However, the existing auction-based mechanisms for crowdsensing do not take sensing inaccuracy into consideration, while guaranteeing that each participator obtains her maximal utility by bidding with her true cost for tasks. To tackle this issue, in this article, we propose OSIER, an optimal mobile crowdsensing incentive under sensing inaccuracy. Specifically, a quantitative analytical framework on characterizing the impact of sensing inaccuracy on a crowdsensing platform is presented, and an optimization problem involving sensing inaccuracy is solved to achieve a maximum utility of the platform. Furthermore, depending on whether a user needs to perform all tasks simultaneously or not, indivisible tasks and divisible tasks are discussed, and OSIER-I and OSIER-D are presented for these two kinds of tasks. Simulation results verify the truthfulness of OSIER, and given a sample set with 5%-20% noise in spectrum sensing, OSIER can achieve 10% higher utilities than the existing crowdsensing mechanisms on average.

Dynamic spectrum access (DSA) is a promising platform to solve the problem of spectrum shortage for which the most challenging issue is spectrum allocation under uncertain availability information, which is referred as a nondeterministic spectrum allocation problem. The nature of such a problem is due to inaccurate spectrum sensing results, which are induced by that power or energy based sensing can be greatly impacted by thermal and environmental noise. For spectrum allocation, auction-based mechanisms have been extensively studied because of channel allocation efficiency, and its potential to achieve bidding truthfulness for secondary uses (SUs). However, most existing spectrum auction mechanisms focus on realizing the truthfulness under certain spectrum availability information. In this paper, we propose FORTUNE, the first truthful online auction mechanism for nondeterministic spectrum allocation by considering uncertain spectrum availability and dynamic spectrum requests. Specifically, we take limited information to compute expected income and losses when interference between primary users (PUs) and SUs occurs, and present a virtual request method for changing of spectrum's actual state. Thorough theoretical analysis proves the truthfulness of FORTUNE. Furthermore, given a sample set with 5%-30% noise in spectrum sensing, FORTUNE achieves not only truthfulness, but also up to 50% higher channel utilization than existing spectrum auction mechanisms.