Recently, federated learning has received widespread attention, which will promote the implementation of artificial intelligence technology in various fields. Privacy-preserving technologies are applied to users’ local models to protect users’ privacy. Such operations make the server not see the true model parameters of each user, which opens wider door for a malicious user to upload malicious parameters and make the training result converge to an ineffective model. To solve this problem, in this article, we propose a poisoning attack defense framework for horizontal federated learning systems called ADFL. Specifically, we design a proof generation method for users to generate proofs to verify whether it is malicious or not. An aggregation rule is also proposed to make sure the global model has a high accuracy. Several verification experiments were conducted and the results show that our method can detect malicious user effectively and ensure the global model has a high accuracy.

For achieving the goal of two-way communication and power flows, smart grids are integrated with much state-of-the-art hardware and software. However, these newly added components also introduce a lot of vulnerabilities into the power systems, which results in that malicious users can launch various cyber–physical attacks to steal electricity. The existing electricity theft detection techniques suffer from an implicit assumption that malicious users tamper with smart meter readings to values much less than their actual electricity consumptions. These are called large-amount electricity theft (LET) attacks. Nevertheless, in the real world, some malicious users may be cautious enough to deliberately launch small-amount electricity theft (SET) attacks, where smart meter readings are manipulated to numbers slightly lower than the actual values, mainly to escape detection. To address this limitation, we propose a detector that is able to deal with both LET and SET attacks effectively. This detector applies a cumulative sum (CUSUM) control chart and a Shewhart control chart together to analyze users’ reported readings and measurements of a central observer meter. It consists of an electricity theft detection phase, which aims to detect the existence of LET/SET attacks timely and a malicious user identification phase, which aims to identify malicious users exactly. Extensive experiments are conducted to evaluate the proposed detector, and the results show that it has good performance in terms of several metrics.

Spatial keyword queries have attracted much attention over the past decade due to the popularity of location-based services and social networks, which brings great economic benefits. Geo-textual data are encrypted-and-delegated to public clouds for efficient management and utilization while preventing potential data leakage. However, it is still challenging to solve secure ra nge s patial k eyword queries on encrypted data since existing works are either vulnerable or inefficient. In this paper, a secure hybrid index is built to implement efficient filtering, by embedding nodes' paths in a novel symmetrical kd-tree into inverted indexes and employing only lightweight cryptographic techniques. A concrete scheme RASK is constructed on the secure index by utilizing only a little storage and computing resources of clients. Furthermore, RASK+ is proposed based on secure virtual technology by migrating all storage burdens from clients to public clouds. Both schemes are theoretically proved to be indistinguishable under adaptive chosen keyword attacks (IND-CKA2). Through experimental evaluations on three real datasets within consistent environments, both schemes reduce the response time by about 50%-80% compared to state-of-the-art solutions (i.e., SKSE, LSKQ, etc.). The storage overheads for the cloud are also reduced by about 0.5-2 orders of magnitude.

As many services have been subcontracted to the cloud, data in such services (i.e., management ap-plications, signal processing, etc.) are stored on cloud in encrypted form for protecting user private information. In many emerging data and signal processing applications, approximate k-nearest neighbor (ANN) query is a prereq-uisite component for massive high-dimensional data (such as time series, biometric, signal, multimedia data, etc.). Unfortunately, existing secure ANN (SANN) methods en-counter the limitation of dimensionality. To further resolve ANN query over high-dimensionally encrypted data, in this paper, we propose an effective SANN model in Euclidean space. In overview, a secure greedy partition method is carefully designed by applying locality sensitive hashing coding and an optimized linear order. Based on that, a novel partition-based SANN (SANNP) and amulti-division version mSANNP resolve SANN query by sequentially scanning a candidate set, which is produced by matching a cloak query with a map index. Our proposed solutions guarantee security and accuracy simultaneously, and reduce commu-nication cost significantly. Meanwhile, the greedy partition method is proved to be indistinguishable secure under chosen-plaintext attack (IND-CPA secure), which is the foundation of security for the proposed solutions. Through extensive experimental studies on four datasets, the pro-posed mechanisms outperform state-of-the-art approaches and provide effective an tradeoff between result accuracy and communication cost.

Skyline computation is a typical multi-objective optimization problem and has been a hot spot for current research. Most of them have concentrated on improving the efficiency of skyline computing, while the security issues associated with executing skyline computation in the cloud server are rarely taken into account. The cloud data stored in plaintext form is attacked by hackers or malicious administrators, resulting in potential leakage of users’ private information. One of the most effective approaches to solve this problem is to encrypt the private data before storing it in the cloud server. However, how to compute the skyline on the encrypted data stored on a cloud server efficiently is still challenging. In this paper, we present a novel framework called SecSky. Firstly, the improved B+ tree structure and symmetric encryption were combined to construct a secure storage structure. Then, the pruning idea was applied to reduce unnecessary calculation and to achieve efficient skyline query and dynamic update. The whole process was directly completed on the ciphertext data. Finally, both the security and the efficiency of SecSky were analyzed through theoretical proof and simulation experiments. The results showed that our scheme is of sound efficiency on the premise of ensuring private data security.

In the era of the Internet of Things (IoT), the rapid development of cloud computing has been advancing location-based services (LBS). To enjoy the considerable advantages of lower cost and higher performance of cloud computing, it has become the first choice for most IoT enterprises to outsource data and services to public clouds. However, the privacy protection of data and the integrity of query results cannot be effectively guaranteed since public clouds cannot be fully trusted. In this paper, we propose a lightweight verifiable range query scheme (namely VeriRange). First, a pair of mutually perpendicular locality-sensitive hashing (LSH) is adopted to divide the original geographic data into subsets. Then a pivoted k dimensional (PKD) tree and a height-balanced binomial search tree (i.e., AVL tree) are established on the client and the cloud respectively to accelerate the processes of queries. To ensure the verification of query results, a keyed hash function is utilized to generate a verification tag for each subset. Verification tags in the public cloud are encrypted and independent of queries. It ensures fast verification of query results because the verification tags are not repeatedly calculated on each query. Formal security analysis shows VeriRange ensures the privacy of data, queries, and results. Experimental studies were conducted on real and synthetic datasets and demonstrated that the query and verification time in VeriRange is almost 2-3 orders of magnitude faster than that in state-of-art schemes.