Visible to the public Biblio

Filters: Author is Li, Yi  [Clear All Filters]
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z 
Z
Zhou, Xingyu, Li, Yi, Barreto, Carlos A., Li, Jiani, Volgyesi, Peter, Neema, Himanshu, Koutsoukos, Xenofon.  2019.  Evaluating Resilience of Grid Load Predictions under Stealthy Adversarial Attacks. 2019 Resilience Week (RWS). 1:206–212.
Recent advances in machine learning enable wider applications of prediction models in cyber-physical systems. Smart grids are increasingly using distributed sensor settings for distributed sensor fusion and information processing. Load forecasting systems use these sensors to predict future loads to incorporate into dynamic pricing of power and grid maintenance. However, these inference predictors are highly complex and thus vulnerable to adversarial attacks. Moreover, the adversarial attacks are synthetic norm-bounded modifications to a limited number of sensors that can greatly affect the accuracy of the overall predictor. It can be much cheaper and effective to incorporate elements of security and resilience at the earliest stages of design. In this paper, we demonstrate how to analyze the security and resilience of learning-based prediction models in power distribution networks by utilizing a domain-specific deep-learning and testing framework. This framework is developed using DeepForge and enables rapid design and analysis of attack scenarios against distributed smart meters in a power distribution network. It runs the attack simulations in the cloud backend. In addition to the predictor model, we have integrated an anomaly detector to detect adversarial attacks targeting the predictor. We formulate the stealthy adversarial attacks as an optimization problem to maximize prediction loss while minimizing the required perturbations. Under the worst-case setting, where the attacker has full knowledge of both the predictor and the detector, an iterative attack method has been developed to solve for the adversarial perturbation. We demonstrate the framework capabilities using a GridLAB-D based power distribution network model and show how stealthy adversarial attacks can affect smart grid prediction systems even with a partial control of network.
Zhang, Conghui, Li, Yi, Sun, Wenwen, Guan, Shaopeng.  2020.  Blockchain Based Big Data Security Protection Scheme. 2020 IEEE 5th Information Technology and Mechatronics Engineering Conference (ITOEC). :574–578.
As the key platform to deal with big data, Hadoop cannot fully protect data security of users by relying on a single Kerberos authentication mechanism. In addition, the single Namenode has disadvantages such as single point failure, performance bottleneck and poor scalability. To solve these problems, a big data security protection scheme is proposed. In this scheme, blockchain technology is adopted to deploy distributed Namenode server cluster to take joint efforts to safeguard the metadata and to allocate access tasks of users. We also improved the heartbeat model to collect user behavior so as to make a faster response to Datanode failure. The smart contract conducts reasonable allocation of user role through the judgment of user tag and risk value. It also establishes a tracking chain of risk value to monitor user behavior in real time. Experiments show that this scheme can better protect data security in Hadoop. It has the advantage of metadata decentralization and the data is hard to be tampered.
W
Wang, Ziyan, Dong, Xinghua, Li, Yi, Fang, Li, Chen, Ping.  2018.  IoT Security Model and Performance Evaluation: A Blockchain Approach. 2018 International Conference on Network Infrastructure and Digital Content (IC-NIDC). :260-264.

It is a research hotspot that using blockchain technology to solve the security problems of the Internet of Things (IoT). Although many related ideas have been proposed, there are very few literatures with theoretical and data support. This paper focuses on the research of model construction and performance evaluation. First, an IoT security model is established based on blockchain and InterPlanetary File System (IPFS). In this model, many security risks of traditional IoT architectures can be avoided, and system performance is significantly improved in distributed large capacity storage, concurrency and query. Secondly, the performance of the proposed model is evaluated through the average latency and throughput, which are meaningful for further research and optimization of this direction. Analysis and test results demonstrate the effectiveness of the blockchain-based security model.

S
Sun, Wenwen, Li, Yi, Guan, Shaopeng.  2019.  An Improved Method of DDoS Attack Detection for Controller of SDN. 2019 IEEE 2nd International Conference on Computer and Communication Engineering Technology (CCET). :249–253.
For controllers of Software Defined Network (SDN), Distributed Denial of Service (DDoS) attacks are still the simplest and most effective way to attack. Aiming at this problem, a real-time DDoS detection attack method for SDN controller is proposed. The method first uses the entropy to detect whether the flow is abnormal. After the abnormal warning is issued, the flow entry of the OpenFlow switch is obtained, and the DDoS attack feature in the SDN environment is analyzed to extract important features related to the attack. The BiLSTM-RNN neural network algorithm is used to train the data set, and the BiLSTM model is generated to classify the real-time traffic to realize the DDoS attack detection. Experiments show that, compared with other methods, this method can efficiently implement DDoS attack traffic detection and reduce controller overhead in SDN environment.
M
Min, Congwen, Li, Yi, Fang, Li, Chen, Ping.  2019.  Conditional Generative Adversarial Network on Semi-supervised Learning Task. 2019 IEEE 5th International Conference on Computer and Communications (ICCC). :1448—1452.

Semi-supervised learning has recently gained increasingly attention because it can combine abundant unlabeled data with carefully labeled data to train deep neural networks. However, common semi-supervised methods deeply rely on the quality of pseudo labels. In this paper, we proposed a new semi-supervised learning method based on Generative Adversarial Network (GAN), by using discriminator to learn the feature of both labeled and unlabeled data, instead of generating pseudo labels that cannot all be correct. Our approach, semi-supervised conditional GAN (SCGAN), builds upon the conditional GAN model, extending it to semi-supervised learning by changing the discriminator's output to a classification output and a real or false output. We evaluate our approach with basic semi-supervised model on MNIST dataset. It shows that our approach achieves the classification accuracy with 84.15%, outperforming the basic semi-supervised model with 72.94%, when labeled data are 1/600 of all data.

G
Gilbert, Anna C., Li, Yi, Porat, Ely, Strauss, Martin J..  2017.  For-All Sparse Recovery in Near-Optimal Time. ACM Trans. Algorithms. 13:32:1–32:26.

An approximate sparse recovery system in ℓ1 norm consists of parameters k, ε, N; an m-by-N measurement Φ; and a recovery algorithm R. Given a vector, x, the system approximates x by &xwidehat; = R(Φ x), which must satisfy ‖ &xwidehat;-x‖1 ≤ (1+ε)‖ x - xk‖1. We consider the “for all” model, in which a single matrix Φ, possibly “constructed” non-explicitly using the probabilistic method, is used for all signals x. The best existing sublinear algorithm by Porat and Strauss [2012] uses O(ε−3klog (N/k)) measurements and runs in time O(k1 − αNα) for any constant α textgreater 0. In this article, we improve the number of measurements to O(ε − 2klog (N/k)), matching the best existing upper bound (attained by super-linear algorithms), and the runtime to O(k1+βpoly(log N,1/ε)), with a modest restriction that k ⩽ N1 − α and ε ⩽ (log k/log N)γ for any constants α, β, γ textgreater 0. When k ⩽ log cN for some c textgreater 0, the runtime is reduced to O(kpoly(N,1/ε)). With no restrictions on ε, we have an approximation recovery system with m = O(k/εlog (N/k)((log N/log k)γ + 1/ε)) measurements. The overall architecture of this algorithm is similar to that of Porat and Strauss [2012] in that we repeatedly use a weak recovery system (with varying parameters) to obtain a top-level recovery algorithm. The weak recovery system consists of a two-layer hashing procedure (or with two unbalanced expanders for a deterministic algorithm). The algorithmic innovation is a novel encoding procedure that is reminiscent of network coding and that reflects the structure of the hashing stages. The idea is to encode the signal position index i by associating it with a unique message mi, which will be encoded to a longer message m′i (in contrast to Porat and Strauss [2012] in which the encoding is simply the identity). Portions of the message m′i correspond to repetitions of the hashing, and we use a regular expander graph to encode the linkages among these portions. The decoding or recovery algorithm consists of recovering the portions of the longer messages m′i and then decoding to the original messages mi, all the while ensuring that corruptions can be detected and/or corrected. The recovery algorithm is similar to list recovery introduced in Indyk et al. [2010] and used in Gilbert et al. [2013]. In our algorithm, the messages \mi\ are independent of the hashing, which enables us to obtain a better result.