Channel state information (CSI) estimation is one of the most fundamental problems in wireless communication systems. Various methods, so far, have been developed to conduct CSI estimation, which usually requires high computational complexity. However, these methods are not suitable for 5G wireless communications due to many techniques (e.g., massive MIMO, OFDM, and millimeter-Wave (mmWave)) to be employed in 5G wireless communication systems. In this paper, we propose an efficient online CSI prediction scheme, called OCEAN, for predicting CSI from historical data in 5G wireless communication systems. Specifically, we first identify several important features affecting the CSI of a radio link and a data sample consists of the information of these features and the CSI. We then design a learning framework that is a combination of a CNN (convolutional neural network) and an LSTM (long short term with memory) network. We further develop an offline-online two-step training mechanism, enabling the prediction results to be more stable when applied in practical 5G wireless communication systems. To validate OCEAN's efficacy, we conduct extensive experiments by considering four typical case studies, i.e., two outdoor and two indoor scenarios. The experiment results show that OCEAN not only obtains the predicted CSI values very quickly but also achieves highly accurate CSI prediction with up to 2.650% - 3.457% average difference ratio (ADR) between the predicted and measured CSI.

Computation offloading is indispensable for mobile edge computing (MEC). It uses edge resources to enable intensive computations and save energy for resource-constrained devices. Existing works generally impose strong assumptions on radio channels and network queue sizes. However, practical MEC systems are subject to various uncertainties rendering these assumptions impractical. In this paper, we investigate the energy-efficient computation offloading problem by relaxing those common assumptions and considering intrinsic uncertainties in the network. Specifically, we minimize the worst-case expected energy consumption of a local device when executing a time-critical application modeled as a directed acyclic graph. We employ the extreme value theory to bound the occurrence probability of uncertain events. To solve the formulated problem, we develop an ϵ-bounded approximation algorithm based on column generation. The proposed algorithm can efficiently identify a feasible solution that is less than (1+ϵ) of the optimal one. We implement the proposed scheme on an Android smartphone and conduct extensive experiments using a real-world application. Experiment results corroborate that it will lead to lower energy consumption for the client device by considering the intrinsic uncertainties during computation offloading. The proposed computation offloading scheme also significantly outperforms other schemes in terms of energy saving.

Solving large-scale sparse linear systems of equations (SLSEs) is one of the most common and fundamental problems in big data, but it is very challenging for resource-limited users. Cloud computing has been proposed as a timely, efficient, and cost-effective way of solving such expensive computing tasks. Nevertheless, one critical concern in cloud computing is data privacy. Specifically, clients’ SLSEs usually contain private information that should remain hidden from the cloud for ethical, legal, or security reasons. Many previous works on secure outsourcing of linear systems of equations (LSEs) have high computational complexity, and do not exploit the sparsity in the LSEs. More importantly, they share a common serious problem, i.e., a huge number of memory I/O operations. This problem has been largely neglected in the past, but in fact is of particular importance and may eventually render those outsourcing schemes impractical. In this paper, we develop an efficient and practical secure outsourcing algorithm for solving large-scale SLSEs, which has low computational and memory I/O complexities and can protect clients’ privacy well. We implement our algorithm on Amazon Elastic Compute Cloud, and find that the proposed algorithm offers significant time savings for the client (up to 74 percent) compared to previous algorithms.

Data privacy is one of the highly discussed issues in recent years as we encounter data breaches and privacy scandals often. This raises a lot of concerns about the ways the data is acquired and the potential information leaks. There are opportunities where the privacy of the data could be violated when used in Artificial Intelligent (AI) models. A considerable portion of user-contributed data is in natural language, and in the past few years, many researchers have proposed NLP-based methods to address these data privacy challenges. To the best of our knowledge, this is the first interdisciplinary review discussing privacy preservation in the context of NLP. In this paper, we present a comprehensive review of previous research conducted to gather techniques and challenges of building and testing privacy-preserving systems in the context of Natural Language Processing (NLP). We group the different works under four categories: 1) Data privacy in the medical domain, 2) Privacy preservation in the technology domain, 3) Analysis of privacy policies, and 4) Privacy leaks detection in the text representation. This review compares the contributions and pitfalls of the various privacy violation detection and prevention works done using NLP techniques to help guide a path ahead.

The popularity of mobile devices has driven the occurrence of ultra-dense wireless networks (UDNs) to accommodate high volumes of data therein generated by mobile users (MUs) running various mobile applications, such as virtual reality/augmented reality (VR/AR) and online gaming. Since many mobile applications are computation-intensive, mobile edge computing (MEC) is considered as a promising computing paradigm that allows MUs to offload computation tasks to EgSvrs in their proximity. Previous works have developed schemes to schedule tasks to EgSvrs for computation offloading. However, these developed schemes are centralized, which are vulnerable to congested coordination operations, and have no trustworthiness guarantee. In this paper, we propose to design a decentralized coordination scheme to orchestrate MUs and EgSvrs by taking advantage of blockchain technology. Specifically, we develop an efficient task-VM matching algorithm that jointly considers task execution time and energy consumption. Particularly, we prove the stability of the developed matching algorithm. Besides, we further implement the developed task-VM matching algorithm on the blockchain by developing a smart matching contract to perform matching on the blockchain without trusted third parties. Extensive simulation results demonstrate that our decentralized coordination scheme can improve the performance significantly and converge to a stable state very quickly.

Well-trained neural network models are deployed on edge servers to provide valuable inference services for clients. To protect the client data privacy, a promising way is to exploit various types of secret sharing to implement privacy-preserving neural network inference. However, the existing schemes are hardly practical due to their high communication rounds and overhead. In this paper, we leverage the lightweight arithmetic secret sharing to develop a new communication-efficient privacy-preserving neural network inference model, which is called Cenia. Specifically, we exploit arithmetic multiplication operations to develop low-interaction secure comparison protocols, that can be used to realize secure activation layers (e.g., ReLU) and secure pooling layers (e.g., max pooling) without expensive garbled circuit and oblivious transfer primitives. Besides, we also design secure exponent and division protocols to realize secure normalization layers (e.g., Sigmoid). Theoretical analysis demonstrates the security and low complexity of Cenia. Extensive experiments have also been conducted on benchmark datasets and classical models, and experimental results show that Cenia achieves privacy-preserving, accurate, and efficient neural network inference. The communication rounds and communication overhead of Cenia are only 37.5% and 60.76% of Sonic (i.e., the state-of-the-art), respectively.