Generalized mutual information (GMI) is used to compute achievable rates for fading channels with various types of channel state information at the transmitter (CSIT) and receiver (CSIR). The GMI is based on variations of auxiliary channel models with additive white Gaussian noise (AWGN) and circularly-symmetric complex Gaussian inputs. One variation uses reverse channel models with minimum mean square error (MMSE) estimates that give the largest rates but are challenging to optimize. A second variation uses forward channel models with linear MMSE estimates that are easier to optimize. Both model classes are applied to channels where the receiver is unaware of the CSIT and for which adaptive codewords achieve capacity. The forward model inputs are chosen as linear functions of the adaptive codeword's entries to simplify the analysis. For scalar channels, the maximum GMI is then achieved by a conventional codebook, where the amplitude and phase of each channel symbol are modified based on the CSIT. The GMI increases by partitioning the channel output alphabet and using a different auxiliary model for each partition subset. The partitioning also helps to determine the capacity scaling at high and low signal-to-noise ratios. A class of power control policies is described for partial CSIR, including a MMSE policy for full CSIT. Several examples of fading channels with AWGN illustrate the theory, focusing on on-off fading and Rayleigh fading. The capacity results generalize to block fading channels with in-block feedback, including capacity expressions in terms of mutual and directed information.
相關內容
In this paper, we present a method to encrypt dynamic controllers that can be implemented through most homomorphic encryption schemes, including somewhat, leveled fully, and fully homomorphic encryption. To this end, we represent the output of the given controller as a linear combination of a fixed number of previous inputs and outputs. As a result, the encrypted controller involves only a limited number of homomorphic multiplications on every encrypted data, assuming that the output is re-encrypted and transmitted back from the actuator. A guidance for parameter choice is also provided, ensuring that the encrypted controller achieves predefined performance for an infinite time horizon. Furthermore, we propose a customization of the method for Ring-Learning With Errors (Ring-LWE) based cryptosystems, where a vector of messages can be encrypted into a single ciphertext and operated simultaneously, thus reducing computation and communication loads. Unlike previous results, the proposed customization does not require extra algorithms such as rotation, other than basic addition and multiplication. Simulation results demonstrate the effectiveness of the proposed method.
We consider massive multiple-input multiple-output (MIMO) systems in the presence of Cauchy noise. First, we focus on the channel estimation problem. In the standard massive MIMO setup, the users transmit orthonormal pilots during the training phase and the received signal at the base station is projected onto each pilot. This processing is optimum when the noise is Gaussian. We show that this processing is not optimal when the noise is Cauchy and as a remedy propose a channel estimation technique that operates on the raw received signal. Second, we derive uplink-downlink achievable rates in the presence of Cauchy noise for perfect and imperfect channel state information. Finally, we derive log-likelihood ratio expressions for soft bit detection for both uplink and downlink, and simulate coded bit-error-rate curves. In addition to this, we derive and compare the symbol detectors in the presence of both Gaussian and Cauchy noises. An important observation is that the detector constructed for Cauchy noise performs well with both Gaussian and Cauchy noises; on the other hand, the detector for Gaussian noise works poorly in the presence of Cauchy noise. That is, the Cauchy detector is robust against heavy-tailed noise, whereas the Gaussian detector is not.
Fuzzing has emerged as a powerful technique for finding security bugs in complicated real-world applications. American fuzzy lop (AFL), a leading fuzzing tool, has demonstrated its powerful bug finding ability through a vast number of reported CVEs. However, its random mutation strategy is unable to generate test inputs that satisfy complicated branching conditions (e.g., magic-byte comparisons, checksum tests, and nested if-statements), which are commonly used in image decoders/encoders, XML parsers, and checksum tools. Existing approaches (such as Steelix and Neuzz) on addressing this problem assume unrealistic assumptions such as we can satisfy the branch condition byte-to-byte or we can identify and focus on the important bytes in the input (called hot-bytes) once and for all. In this work, we propose an approach called \tool~which is designed based on the following principles. First, there is a complicated relation between inputs and branching conditions and thus we need not only an expressive model to capture such relationship but also an informative measure so that we can learn such relationship effectively. Second, different branching conditions demand different hot-bytes and we must adjust our fuzzing strategy adaptively depending on which branches are the current bottleneck. We implement our approach as an open source project and compare its efficiency with other state-of-the-art fuzzers. Our evaluation results on 10 real-world programs and LAVA-M dataset show that \tool~achieves sustained increases in branch coverage and discovers more bugs than other fuzzers.
The vision transformer is a model that breaks down each image into a sequence of tokens with a fixed length and processes them similarly to words in natural language processing. Although increasing the number of tokens typically results in better performance, it also leads to a considerable increase in computational cost. Motivated by the saying "A picture is worth a thousand words," we propose an innovative approach to accelerate the ViT model by shortening long images. Specifically, we introduce a method for adaptively assigning token length for each image at test time to accelerate inference speed. First, we train a Resizable-ViT (ReViT) model capable of processing input with diverse token lengths. Next, we extract token-length labels from ReViT that indicate the minimum number of tokens required to achieve accurate predictions. We then use these labels to train a lightweight Token-Length Assigner (TLA) that allocates the optimal token length for each image during inference. The TLA enables ReViT to process images with the minimum sufficient number of tokens, reducing token numbers in the ViT model and improving inference speed. Our approach is general and compatible with modern vision transformer architectures, significantly reducing computational costs. We verified the effectiveness of our methods on multiple representative ViT models on image classification and action recognition.
Shannon's channel coding theorem characterizes the maximal rate of information that can be reliably transmitted over a communication channel when optimal encoding and decoding strategies are used. In many scenarios, however, practical considerations such as channel uncertainty and implementation constraints rule out the use of an optimal decoder. The mismatched decoding problem addresses such scenarios by considering the case that the decoder cannot be optimized, but is instead fixed as part of the problem statement. This problem is not only of direct interest in its own right, but also has close connections with other long-standing theoretical problems in information theory. In this monograph, we survey both classical literature and recent developments on the mismatched decoding problem, with an emphasis on achievable random-coding rates for memoryless channels. We present two widely-considered achievable rates known as the generalized mutual information (GMI) and the LM rate, and overview their derivations and properties. In addition, we survey several improved rates via multi-user coding techniques, as well as recent developments and challenges in establishing upper bounds on the mismatch capacity, and an analogous mismatched encoding problem in rate-distortion theory. Throughout the monograph, we highlight a variety of applications and connections with other prominent information theory problems.
A major security threat to an integrated circuit (IC) design is the Hardware Trojan attack which is a malicious modification of the design. Previously several papers have investigated into side-channel analysis to detect the presence of Hardware Trojans. The side channel analysis were prescribed in these papers as an alternative to the conventional logic testing for detecting malicious modification in the design. It has been found that these conventional logic testing are ineffective when it comes to detecting small Trojans due to decrease in the sensitivity due to process variations encountered in the manufacturing techniques. The main paper under consideration in this survey report focuses on proposing a new technique to detect Trojans by using multiple-parameter side-channel analysis. The novel idea will be explained thoroughly in this survey report. We also look into several other papers, which talk about single parameter analysis and how they are implemented. We analyzed the short comings of those single parameter analysis techniques and we then show how this multi-parameter analysis technique is better. Finally we will talk about the combined side-channel analysis and logic testing approach in which there is higher detection coverage for hardware Trojan circuits of different types and sizes.
A binary code of blocklength $n$ and codebook size $M$ is called an $(n,M)$ code, which is studied for memoryless binary symmetric channels (BSCs) with the maximum likelihood (ML) decoding. For any $n \geq 2$, some optimal codes among the linear $(n,4)$ codes have been explicitly characterized in the previous study, but whether the optimal codes among the linear codes are better than all the nonlinear codes or not is unknown. In this paper, we first show that for any $n\geq 2$, there exists an optimal code (among all the $(n,4)$ codes) that is either linear or in a subset of nonlinear codes, called Class-I codes. We identified all the optimal codes among the linear $(n,4)$ codes for each blocklength $n\geq 2$, and found ones that were not given in literature. For any $n$ from $2$ to $300$, all the optimal $(n,4)$ codes are identified, where except for $n=3$, all the optimal $(n,4)$ codes are equivalent to linear codes. There exist optimal $(3,4)$ codes that are not equivalent to linear codes. Furthermore, we derive a subset of nonlinear codes called Class-II codes and justify that for any $n >300$, the set composed of linear, Class-I and Class-II codes and their equivalent codes contains all the optimal $(n,4)$ codes. Both Class-I and Class-II codes are close to linear codes in the sense that they involve only one type of columns that are not included in linear codes. Our results are obtained using a new technique to compare the ML decoding performance of two codes, featured by a partition of the entire range of the channel output.
Sum-rank-metric codes have wide applications in universal error correction, multishot network coding, space-time coding and the construction of partial-MDS codes for repair in distributed storage. Fundamental properties of sum-rank-metric codes have been studied and some explicit or probabilistic constructions of good sum-rank-metric codes have been proposed. In this paper we give three simple constructions of explicit linear sum-rank-metric codes. In finite length regime, numerous larger linear sum-rank-metric codes with the same minimum sum-rank distances as the previous constructed codes can be derived from our constructions. For example several better linear sum-rank-metric codes over ${\bf F}_q$ with small block sizes and the matrix size $2 \times 2$ are constructed for $q=2, 3, 4$ by applying our construction to the presently known best linear codes. Asymptotically our constructed sum-rank-metric codes are close to the Gilbert-Varshamov-like bound on sum-rank-metric codes for some parameters. Finally we construct a linear MSRD code over an arbitrary finite field ${\bf F}_q$ with various square matrix sizes $n_1, n_2, \ldots, n_t$ satisfying $n_i \geq n_{i+1}^2+\cdots+n_t^2$ , $i=1, 2, \ldots, t-1$, for any given minimum sum-rank distance. There is no restriction on the block lengths $t$ and parameters $N=n_1+\cdots+n_t$ of these linear MSRD codes from the sizes of the fields ${\bf F}_q$. \end{abstract}
Graph Convolutional Networks (GCNs) have been widely applied in various fields due to their significant power on processing graph-structured data. Typical GCN and its variants work under a homophily assumption (i.e., nodes with same class are prone to connect to each other), while ignoring the heterophily which exists in many real-world networks (i.e., nodes with different classes tend to form edges). Existing methods deal with heterophily by mainly aggregating higher-order neighborhoods or combing the immediate representations, which leads to noise and irrelevant information in the result. But these methods did not change the propagation mechanism which works under homophily assumption (that is a fundamental part of GCNs). This makes it difficult to distinguish the representation of nodes from different classes. To address this problem, in this paper we design a novel propagation mechanism, which can automatically change the propagation and aggregation process according to homophily or heterophily between node pairs. To adaptively learn the propagation process, we introduce two measurements of homophily degree between node pairs, which is learned based on topological and attribute information, respectively. Then we incorporate the learnable homophily degree into the graph convolution framework, which is trained in an end-to-end schema, enabling it to go beyond the assumption of homophily. More importantly, we theoretically prove that our model can constrain the similarity of representations between nodes according to their homophily degree. Experiments on seven real-world datasets demonstrate that this new approach outperforms the state-of-the-art methods under heterophily or low homophily, and gains competitive performance under homophily.
Pre-trained deep neural network language models such as ELMo, GPT, BERT and XLNet have recently achieved state-of-the-art performance on a variety of language understanding tasks. However, their size makes them impractical for a number of scenarios, especially on mobile and edge devices. In particular, the input word embedding matrix accounts for a significant proportion of the model's memory footprint, due to the large input vocabulary and embedding dimensions. Knowledge distillation techniques have had success at compressing large neural network models, but they are ineffective at yielding student models with vocabularies different from the original teacher models. We introduce a novel knowledge distillation technique for training a student model with a significantly smaller vocabulary as well as lower embedding and hidden state dimensions. Specifically, we employ a dual-training mechanism that trains the teacher and student models simultaneously to obtain optimal word embeddings for the student vocabulary. We combine this approach with learning shared projection matrices that transfer layer-wise knowledge from the teacher model to the student model. Our method is able to compress the BERT_BASE model by more than 60x, with only a minor drop in downstream task metrics, resulting in a language model with a footprint of under 7MB. Experimental results also demonstrate higher compression efficiency and accuracy when compared with other state-of-the-art compression techniques.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191