Search

In this paper, a metasurface composed of 3-bit coding linear polarization conversion elements and its application to RCS reduction of the patch antenna is intensively studied. At first, 3-bit coding metasurface are constructed by a sequence of eight coded unit cells, which have a similar cross-polarized reflected amplitude response and gradient reflected phase responses covering 0–2π, respectively. Equivalent circuit models of these unit cells are created to describe their electrical behavior for the two linear incident polarizations at the same time. Then, a patch antenna is integrated on the 3-bit metasurface, of which the elements are placed with a 2-dimensional linear coding sequence. The metal square ring is set around the patch antenna to protect it from the disturbance of metasurface. Both the simulation and experiment results demonstrate that the designed metasurface can primarily reduce the antenna RCS at a broadband, while the antenna performances are not degraded significantly.

Purpose The self-rated health (SRH) item is frequently used in health surveys but variations of its form (wording, response options) may hinder comparisons between versions over time or across surveys. The objectives were to determine (a) whether three SRH forms are equivalent, (b) the form with the best construct validity and (c) the best coding scheme to maximize equivalence across forms. Methods We used data from 58,023 respondents of the Swiss Health Survey. Three SRH forms were used. Response options varied across forms and we explored four coding schemes (two considering SRH as continuous, two as dichotomous). Construct validity of the SRH was assessed using 34 health predictors to estimate the explained variance. Results Distributions of response options were similar across SRH forms, except for the “good” and “very good” options (“good” in form 1: 58.6%, form 2: 65.0% and form 3: 44.1%). Explained variances differed across SRH forms, with form 3 providing the best overall explained variance, regardless of coding schemes. The linear coding scheme maximised the equivalence across SRH forms. Conclusion The three SRH forms were not equivalent in terms of construct validity. Studies examining the evolution of SRH over time with surveys using different forms should use the linear coding scheme to maximise equivalence between SRH forms.

Image classification is very important in pattern recognition and computer vision, where, for integrating final representation, feature pooling methods of the max-pooling, sum-pooling and average-pooling have been widely used. In this study, the authors propose a new method called K -strongest responses (KSRs) on the dictionary atoms for integrating the coding coefficients to generate the final representation that is compared with the previous pooling methods, produces better performance for the image classification task. On the basis of the KSR method, to improve classification accuracy and generate more compact and discriminative final representation, a new framework consisting of two-part KSR and bag-of-features is proposed. To evaluate the performance of the proposed method and framework, they apply it to locality-constrained linear coding, linear distance coding and sparse coding by using two datasets from benchmarks of scene classification: 19-class satellite scene and UC Merced Land. The results show that the coding coefficients integrated by their method and framework are more discriminative than other methods.

This paper presents a fighting detection algorithm based on hybrid features. An effective coding scheme called locality-constrained linear coding (LLC) is used for the features by incorporating histograms of quadruples optical flow gradient (HQOFG) into local spatio-temporal features. Our results confirm that the fighting detector only with a linear support vector machine for sequence classification performs well. Compared to the fighting detector using bag-of-words (BOW) framework and local spatio-temporal features, the detector proposed in the paper improves the recognition accuracy.

Millimeter wave (mmWave) multiple-input multiple-output (MIMO) systems have important advantages such as large bandwidths and small wavelengths that enable large antenna arrays. However, the cost and power consumption of the hardware elements is extremely high in mmWave. This limitation makes precoding and combining for multiuser MIMO more constrained, making non-viable the use of fully digital precoders. Hybrid analog-digital transceiver architectures allow to reduce radio frequency (RF) hardware complexity but still, a significant number of RF chains is necessary to accommodate the transmissions from multiple users. In this work, we propose to use a zero-delay analog joint source-channel coding method termed scalar quantizer linear coding (SQLC) to reduce the number of RF chains at the receiver of an uplink multiuser mmWave MIMO system. The use of SQLC allows grouping the users in pairs and halves the number of streams to be managed by the multiuser MIMO receiver. In addition, SQLC performs particularly well when users transmit correlated information.

In order to improve the data integrity encryption ability of cloud storage users, a cloud storage user data integrity encryption technology based on random linear coding is proposed. Firstly, the cloud storage user data integrity encryption key of cloud storage user data integrity object is constructed, then the encryption and decryption coding design are carried out, and the random linear coding is used to optimize the digital encryption key to improve the anti-deciphering level. Finally, the simulation results show that the cloud storage user data integrity encryption technology has high random distribution of encryption, the deciphering rate of encrypted data is effectively controlled, and the performance of the cloud storage user data encryption technology is better than that of the traditional method. It effectively ensures the integrity of cloud storage user data.

Bag of Words (BOW) method with spatio-temporal local features has achieved great performance in human action recognition. However, most of the existing BOW approaches based on vector quantization (VQ) neglect the contextual information of each descriptor, and suffer serious quantization error. There are two main reasons for these: in the first, each local feature is only assigned to one label and second, the information about the spatial layout of the features is disregarded. In this paper, we present a novel and effective coding method called context and locality constrained linear coding (CLLC) to overcome these limitations, in which the relationships among local features and their structural information are preserved. After that, a group-wise sparse representation based classification (GSRC) method is implemented to assign the query sample into one category which yields the smallest reconstruction error. Our method is verified on the challenging databases and achieves comparable performance with state-of-the-art methods. We model a novel and effective human action recognition scheme.We propose context and locality constrained linear coding method to fully describe each local feature."Decisive feature" and "indecisive feature" are aided to collect the contextual information.Group-wise sparse representation based classification method is introduced to category the query samples.Our encoding problem has both analytical solution and approximation solution.

To utilize the geometry structure information and similarity information within the neighborhood surrounding a spatio-temporal interest point for human action recognition task, we employ the axes of a regular polyhedron as a reference locating system, and build a novel local feature named polyhedron neighborhood feature (PNF). Then, to reduce quantization error in the coding stage, locality-constrained linear coding method is used to encode the obtained PNFs. Next, multi-temporal-scale PNFs (MPNFs) are created for handling the problem of various action speeds. In classification, support vector machine (SVM) based on linear kernel is used as classifier taking time consumption into account. The experiments on the KTH and UCF sports datasets show that the recognition system based on PNFs achieves better performance than the competing local spatio-temporal feature-based human action recognition methods.

The convolutional neural network(CNN) has received an excellent performance in the recent tasks of image retrieval and image classification. Besides, some local coding methods have been focused on because of their outstanding local description. In this paper, we present a novel method for palmprint recognition, which combines CNN and local coding techniques. We conduct feature extraction on palmprint images using the pre-trained networks and obtain the local convolutional features. Then three different local coding methods are adopted to code the features from CNN, including Bag of Visual Word(BoVW), Locality-constrained Linear Coding(LLC) and Vector of Locally Aggregated Descriptors (VLAD). The features extracted from CNN have better representation. Meanwhile, the local coding makes local features more important. Hence the combination of them can get the anticipated better results. The proposed method is extensively evaluated on the PolyU palmprint database and PolyU multispectral palmprint database, illuminating the remarkable performance on palmprint recognition.

We study a two-phase caching network consisting of one server with N files connected to K users through an error-free shared link. Each user has a cache memory which can store M files in the placement phase. In the delivery phase, each user requests L files, and the server transmits the messages accordingly. Using the message sent by the server combined with the cache memory, each user reconstructs the L files they requested. In this work, we focus on the case L > 1, i.e., the case of multiple file requests. We adopt the symmetric batch caching scheme and propose a general delivery scheme. To prove the optimality of the proposed general delivery scheme, we apply two converse techniques. The first converse technique is for general coding schemes and is obtained through virtual user construction. The second converse technique is for vector linear coding schemes and is obtained using an interference alignment point of view. With these two converse techniques, we characterize either the unconstrained optimal coding rate, or optimum linear coding rate, with symmetric batch caching for certain cases.

In aging research, morphological age of tissue helps to characterize the effects of aging on different individuals. While currently manual evaluations are used to estimate morphological ages under microscopy, such operation is difficult and subjective due to the complex visual characteristics of tissue images. In this paper, we propose an automated method to quantify morphological ages of tissues from microscopy images. We design a new sparse representation method, namely dual discriminative local coding (DDLC), that classifies the tissue images into different chronological ages. DDLC in- corporates discriminative distance learning and dual-level local coding into the basis model of locality-constrained linear coding thus achieves higher discriminative capability. The morphological age is then computed based on the classification scores. We conducted our study using the publicly avail- able terminal bulb aging database that has been commonly used in existing microscopy imaging research. To represent these images, we also design a highly descriptive descriptor that combines several complementary texture features extracted at two scales. Experimental results show that our method achieves significant improvement in age classification when compared to the existing approaches and other popular classifiers. We also present promising results in quantification of morphological ages.

Author(s): Maleki, H; Cadambe, VR; Jafar, SA | Abstract: The index coding problem is studied from an interference alignment perspective providing new results as well as new insights into, and generalizations of, previously known results. An equivalence is established between the capacity of multiple unicast index coding (where each message is desired by exactly one receiver), and groupcast index coding (where a message can be desired by multiple receivers), which settles the heretofore open question of insufficiency of linear codes for the multiple unicast index coding problem by equivalence with groupcast settings, where this question has previously been answered. Necessary and sufficient conditions for the achievability of rate half per message in the index coding problem are shown to be a natural consequence of interference alignment constraints, and generalizations to feasibility of rate 1/(L+1)per message when each destination desires at least L messages, are similarly obtained. Finally, capacity optimal solutions are presented to a series of symmetric index coding problems inspired by the local connectivity and local interference characteristics of wireless networks. The solutions are based on vector linear coding. © 1963-2012 IEEE.

This paper discusses random linear network coding with and without the use of a Vandermonde matrix to obtain the coding coefficients. Performance comparisons of such random linear network coded networks with networks employing traditional store and forward technique are also provided. It is shown that random linear network coding using a Vandermonde matrix can improve the network utilization factor by reducing the overhead compared to random linear coding that does not use a Vandermonde matrix. Our numerical results show that random linear network coding with a Vandermonde matrix provides a considerable improvement in throughput and delay when compared to a network employing a traditional store and forward strategy. An inherent feature of random linear network coding which makes it possible to employ simple encryption techniques is also discussed.

In Delay Tolerant Networks (DTNs) the core challenge is to cope with lack of persistent connectivity and yet be able to deliver messages from source to destination. In particular, routing schemes that leverage relays' memory and mobility area customary solution in order to improve message delivery delay. When large files need to be transferred from source to destination, not all packets may be available at the source prior to the first transmission. This motivates us to study general packet arrivals at the source, derive performance analysis of replication based routing policies and study their optimization under two-hop routing. In particular, we determine the conditions for optimality in terms of probability of successful delivery and mean delay and we devise optimal policies, so-called piecewise-threshold policies. We account for linear block-codes and rateless random linear coding to efficiently generate red undancy, as well as for an energy constraint in the optimization. We numerically assess the higher efficiency of piecewise-threshold policies compared with other policies by developing heuristic optimization of the thresholds for all flavors of coding considered.

We propose a 3D action recognition algorithm which uses depth-based Gradient Local Auto-Correlations (GLAC) feature and Locality-constrained Affine Subspace Coding (LASC) to improve the discriminative ability of human actions in spatio-temporal subsequences of 3D depth videos. First, each entire depth video sequence is divided automatically into a set of subsequences (i.e., multi-scale sub-actions) by the normalized motion energy vector. Next Depth Motion Maps (DMMs) based GLAC features are employed to capture the shape information and motion cues of each sub-action. In order to obtain a more compact and discriminative representation, LASC is then proposed to encode the features extracted from the depth video. We show that the use of LASC exhibits better performance compared to existing methods such as Locality-constrained Linear Coding (LLC). On all three datasets we obtain competitive results compared to fifteen methods, while using fewer features and less complex models.

Bag-of-Words lies at a heart of modern object category recognition systems. After descriptors are extracted from images, they are expressed as vectors representing visual word content, referred to as mid-level features. In this paper, we review a number of techniques for generating mid-level features, including two variants of Soft Assignment, Locality-constrained Linear Coding, and Sparse Coding. We also isolate the underlying properties that affect their performance. Moreover, we investigate various pooling methods that aggregate mid-level features into vectors representing images. Average pooling, Max-pooling, and a family of likelihood inspired pooling strategies are scrutinised. We demonstrate how both coding schemes and pooling methods interact with each other. We generalise the investigated pooling methods to account for the descriptor interdependence and introduce an intuitive concept of improved pooling. We also propose a coding-related improvement to increase its speed. Lastly, state-of-the-art performance in classification is demonstrated on Caltech101, Flower17, and ImageCLEF11 datasets.

Although traditional bag-of-words model has shown promising results for action recognition, it takes no consideration of the relationship among spatio-temporal points; furthermore, it also suffers serious quantization error. In this letter, we propose a novel coding strategy called context-constrained linear coding (CLC) to overcome these limitations. We first calculate the contextual distance between local descriptors and each codeword by considering the spatio-temporal contextual information. Then, linear coding using contextual distance is adopted to alleviate the quantization error. Our method is verified on two challenging databases (KTH and UCF sports), and the experimental results demonstrate that our method achieves better results than previous methods in action recognition.

One of the current versions of additive schemes to estimate gas chromatographic retention indices of organic compounds not characterized by these constants assumes their structure to be assembled from simpler precursors, and subsequent arithmetical operations are performed with the known indices by the selected schemes. In order to optimize the use of such an algorithm (to confirm the correctness of the selection of structural analogs, to eliminate superposition uncertainty, etc.) it is proposed to use the linear coding system for not complete molecular structures, but only those fragments that mostly affect the chromatographic retention parameters.

We show the sufficient conditions for coding by nodes such that every sink can decode all information multicasted by the single source when there exists time-varying delay of information transmission at links in the network coding.

The most important issue of image classification algorithm based on feature extraction is how to efficiently encode features. Locality-constrained linear coding (LLC) has achieved the state of the art performance on several benchmarks, due to its underlying properties of better construction and local smooth sparsity. However, the negative code may make LLC more unstable. In this paper, a novel coding scheme is proposed by adding an extra non-negative constraint based on LLC. Generally, the new model can be solved by iterative optimization methods. Moreover, to reduce the encoding time, an approximated method called NNLLC is proposed, more importantly, its computational complexity is similar to LLC. On several widely used image datasets, compared with LLC, the experimental results demonstrate that NNLLC not only can improve the classification accuracy by about 1–4 percent, but also can run as fast as LLC.

Convolutional Neural Network (CNN) features have demonstrated outstanding performance as global representations for image classification, but they lack invariance to scale transformation, which makes it difficult to adapt to various complex tasks such as scene classification. To strengthen the scale invariance of CNN features and meanwhile retain their powerful discrimination in scene classification, we propose a framework where cross-level Locality-constrained Linear Coding and cascaded fine-tuned CNN features are combined, which is shorted as cross-level LLC-CNN. Specifically, this framework first fine-tunes multi-level CNNs in a cascaded way, then extracts multi-level CNN features to learn a cross-level universal codebook, and finally performs locality-constrained linear coding (LLC) and max-pooling on the patches of all levels to form the final representation. It is experimentally verified that the LLC responses on the universal codebook outperform the CNN features and achieve the state-of-the-art performance on the two currently largest scene classification benchmarks, MIT Indoor Scenes and SUN 397.

An equivalence between multiple multicast and multiple unicast network coding is proven in a 2007 paper by Dougherty and Zeger. The result is extended to index coding under the assumption of linear coding in recent work by Maleki et al. In this work, we extend the result to arbitrary (e.g. possibly noisy) memoryless networks under an asymptotic definition of reliability. We show the equivalence for non-linear index codes as a special case. Our result implies that solving the multiple unicast communication problem is sufficient to solve all communication problems in memoryless networks.

We study index-coding problems (one sender broadcasting messages to multiple receivers) where each message is requested by one receiver, and each receiver may know some messages a priori. This type of index-coding problems can be fully described by directed graphs. The aim is to find the minimum codelength that the sender needs to transmit in order to simultaneously satisfy all receivers' requests. For any directed graph, we show that if a maximum acyclic induced subgraph (MAIS) is obtained by removing two or fewer vertices from the graph, then the minimum codelength (i.e., the solution to the index-coding problem) equals the number of vertices in the MAIS, and linear codes are optimal for this index-coding problem. Our result increases the set of index-coding problems for which linear index codes are proven to be optimal., accepted and to be presented at the 2014 International Symposium on Network Coding (NetCod)

We propose using linear mappings over finite rings as encoders in the Slepian-Wolf and the source coding for computing problems. It is known that the arithmetic of many finite rings is substantially easier to implement than the one of finite fields. Hence, one of the advantages of using linear mappings over rings, instead of its field counterparts, is reducing implementation complexity. More importantly, the ring version dominates the field version in terms of achieving strictly better coding rates with strictly smaller alphabet size in the source coding for computing problem [1]. This paper is dedicated to proving an achievability theorem of linear source coding over finite rings in the Slepian-Wolf problem. This result includes those given by Elias [2] and Csiszar [3] saying that linear coding over finite fields is optimal, i.e. achieves the Slepian-Wolf region. Although the optimality issue remains open, it has been verified in various scenarios including particularly many cases use non-field rings [1], [4].

Human group behaviors are usually composed of several sub-groups. Considering the interaction between groups, this paper presents an algorithm to recognize human group behavior with multi-group causalities. It has two main contributions: (1) we introduce inter-group causality to reflect the interaction between human groups, (2) an improved coding scheme (i.e. Locality-constrained Linear Coding) is used for encoding the causality to go beyond Vector Quantization (VQ). Finally, a simple linear SVM is adopted to learn this model. Our experiment results demonstrate that inter-group causality feature and LLC methods can significantly boost behavior recognition performance.

This paper studies linear coding (LC) techniques in the setting of computing functions of correlated memoryless sources. Instead of linear mappings over finite fields, we consider using linear mappings over finite rings as encoders. It is shown that generally the region c×R, where c ≥ 1 is a constant and R is the Slepian-Wolf (SW) region, is achievable with LC over ring (LCoR) when the function to compute is the identity function. c = 1 if the ring used is a field. Hence, LCoR could be suboptimal in terms of achieving the best coding rates (the SW region) for computing the identity function. In spite of that, the ring version shows several advantages. It is demonstrated that there exists a function that is neither linear nor can be linearized over any finite field. Thus, LC over field (LCoF) does not apply directly for computing such a function unless the polynomial approach [1], [2] is used. On the contrary, such a function is linear over some ring. Using LCoR, an achievable region containing the SW region can be obtained for computing this function. In addition, the alphabet sizes of the encoders are strictly smaller than using LCoF. More interestingly, LCoF is not useful if some special requirement is imposed.

Motivated by the presence of an implicit communication channel in the asymptotic version of Witsenhausen's counterexample, implicit discrete memoryless channels (IDMC) with discrete memoryless (DM) states are considered. Information-theoretic lower and upper bounds (based respectively on the ideas from rate-distortion theory and hybrid-coding) are derived on the optimal distortion in estimating the input of the implicit channel. The intuition gained from the DMIC with DM state model is then used to evaluate the optimal distortion for the asymptotic version of the Witsenhausen counterexample. The minimum distortion is characterized for the counterexample and it is shown that a combination of linear coding and dirty-paper coding (DPC) proposed in [1] achieves the minimum distortion.

Network coding, introduced by Ahlswede et al. in their pioneering work [1], has generated considerable research interest in recent years, and numerous subsequent papers, e.g., [2–6], have built upon this concept. One of the main advantages of network coding over traditional routed networks is in the area of multicast, where common information is transmitted from a source node to a set of terminal nodes. Ahlswede et al. showed in [1] that network coding can achieve the maximum multicast rate, which is not achievable by routing alone. When coding is used to perform multicast, the problem of establishing minimum cost multicast connection is equivalent to two effectively decoupled problems: one of determining the subgraph to code over and the other of determining the code to use over that subgraph. The latter problem has been studied extensively in [5, 7–9], and a variety of methods have been proposed, which include employing simple random linear coding at every node. Such random linear coding schemes are completely decentralized, requiring no coordination between nodes, and can operate under dynamic conditions [10]. These papers, however, all assume the availability of dedicated network resources. In this chapter, we focus on the former problem, which is to find the min-cost subgraph that allows the given multicast connection to be established (with appropriate coding) over coded packet networks. This problem has been studied in [11,12]. The analogous problem for routed network is the Steiner tree problem, which is known

This contribution gives an introduction to algebraic coding theory over rings. We will start with a historical sketch and then present basics on rings and modules. Particular attention will be paid to weight functions on these, before some foundational results of ring-linear coding will be discussed. Among these we will deal with code equivalence, and with MacWilliams’ identities about the relation between weight enumerators. A further section is devoted to existence bounds and code optimality. An outlook will then be presented on the still unsolved problem of the construction of large families of ring-linear codes of high quality.

We introduce the concept of average linear coding throughput when the alphabet is restricted to the binary field. We further investigate the maximum achievable coding throughput compared to integer routing throughput, average integer routing throughput and average linear coding throughput over GF(2) for some networks. We also show that there exist a multicast network solvable over the binary field whose linear coding throughput is a factor Theta (radic|V|) greater than its integer routing throughput, where |V| is the number of nodes in the network

A generalization of the Gaussian dirty-paper problem to a multiple access setup is considered. There are two additive interference signals, one known to each transmitter but none to the receiver. The rates achievable using Costa's strategies (i.e. by a random binning scheme induced by Costa's auxiliary random variables) vanish in the limit when the interference signals are strong. In contrast, it is shown that lattice strategies ("lattice precoding") can achieve positive rates independent of the interferences, and in fact in some cases - which depend on the noise variance and power constraints - they are optimal. In particular, lattice strategies are optimal in the limit of high SNR. It is also shown that the gap between the achievable rate region and the capacity region is at most 0.167 bit. Thus, the dirty MAC is another instance of a network setup, like the Korner-Marton modulo-two sum problem, where linear coding is potentially better than random binning. Lattice transmission schemes and conditions for optimality for the asymmetric case, where there is only one interference which is known to one of the users (who serves as a "helper" to the other user), and for the "common interference" case are also derived. In the former case the gap between the helper achievable rate and its capacity is at most 0.085 bit., 42 pages, 14 figures

This article advances linear coding construction method for DNA codes based on three-letter alphabets. Further more it also advances the new conception of Constrain Strength and algorithm of Construction and Search. We use the former to evaluate the coding amount of combinatorial constraints and use the latter to validate the evaluated results. Experiments show that experiment result accords with the theoretically evaluated value very well.

We develop a method for the linear coding of discrete time continuous amplitude source signals, on both SISO and MIMO FIR channels that minimizes the end-to-end mean square error. This is a joint source channel coding procedure. We develop two coding approaches and derive asymptotic performance limits for both of them.

A new triangulation-based optical profilometry-linearly coded profilometry (LCP)-is presented. In this method, a linear coding technique is introduced. It uses a periodic sawtoothlike structure light to code the surface being measured and the phase-shifting technique to decode the profile. A coding light that has isosceles triangle teeth is proposed. To realize the decoding, at least three samples are needed. When more than three samples are used, higher accuracy can be obtained. A decoding method of LCP with an even number of samples is provided. This method has been realized in my laboratory and the experimental results verify the reliability of LCP. Because the coding method used in LCP is simpler than that used in phase-measuring profilometry, LCP has a faster measuring speed.

The authors describe the multiband linear predictive (MB-LPC) vocoder and its operation at 2.4 kb/s and 1.2 kb/s. The MB-LPC vocoder uses mixed excitation and exploits the advantages of both time and frequency domain speech coding techniques to produce natural sounding, good quality speech. Subjective performance of speech at 2.4 kb/s produced by the MB-LPC is very close to that for the 4.15 kb/s INMARSAT-M IMBE speech coder. Informal listening tests have shown that in most cases people could not tell the difference between the new 2.4 kb/s MB-LPC coder and the 4.15 kb/s INMARSAT-M IMBE coder.