Search

Segmenting meaningful visual structures from an image is a fundamental and most-addressed problem in image analysis algorithms. However, among factors such as diverse visual patterns, noise, complex backgrounds, and similar textures present in foreground and background, image segmentation still stands as a challenging research problem. In this article, the proposed method employs an unsupervised method that addresses image segmentation as subspace clustering of image feature vectors. Initially, an image is partitioned into a set of homogeneous regions called superpixels, from which Local Spectral Histogram features are computed. Subsequently, a feature data matrix is created whereupon subspace clustering methodology is applied. A single-stage optimization model is formulated with enhanced segmentation capabilities by the combined action of l ½ and l 2 norm minimization. Robustness of l ½ regularization toward both the noise and overestimation of sparsity provides simultaneous noise robustness and better subspace selection, respectively. While l 2 norm facilitates grouping effect. Hence, the designed optimization model ensures an improved sparse solution and a sparse representation matrix with an accurate block diagonal structure, which thereby favours getting properly segmented images. Then, experimental results of the proposed method are compared with the state-of-art algorithms. Results demonstrate the improved performance of our method over the state-of-art algorithms.

In forensic document analysis, the authenticity of a document must be properly checked in the context of suspected forgery. Hyperspectral Imaging (HSI) is a non-invasive way of detecting fraudulent papers in a multipage document. The occurrence of a forged paper in a multi-page document may have a substantial difference from rest of the papers in its age, type, color, texture, and so on. Each pixel in an HSI data can be used as the material fingerprint for the spatial point it corresponds to. Hence, hyperspectral data of paper samples made of the same substance have similar characteristics and can be grouped into a single cluster. Similarly, paper samples made of different substances have different spectral properties. This paper relies on this heuristic and proposes a tensor based clustering framework for hyperspectral paper data, with an application to detect the forged papers in multi-page documents. Information embedded in the hyperspectral patches of the papers to be clustered is arranged into individual lateral slices of a third-order tensor in this framework. Further, this work employs the self-expressiveness property of submodules and an objective function is formulated to extract self-expressive representation tensor with low multirank and f-diagonal structure. Objective function of the proposed method incorporates $l_{\frac {1}{2}}$ -induced Tensor Nuclear Norm (TNN) and $l_{\frac {1}{2}}$ regularization to impart better low rankness and f-diagonal structure to the representation tensor. Experimental results of the proposed method were compared to the state-of-the-art subspace clustering approaches. The results demonstrate improved performance of the proposed method over the existing clustering algorithms.

An Atmospheric Water Generator (AWG) is a device that extracts water from humid air by cooling it below the dew point, compressing the air, or exposing the air to hygroscopic substances. The AWG incorporates the concept of thermoelectric cooling (TEC) to condense water vapor from humid air and is designed to produce two liters of water in twelve hours from the air at 30 °C and 80% relative humidity (RH) to meet the average water need of a person. Experiments are conducted on a prototype with the intent to study the effect of airspeed on the amount of water condensed and thus infer the optimum speed for maximum water output. The use of TEC for condensation in AWG is economical on a small scale and does not compromise on the environment in the long run. The results obtained can influence future designs of the same.

Incomplete observations in the data are always troublesome to data clustering algorithms. In fact, most of the well-received techniques are not designed to encounter such imperative scenarios. Hence, clustering of images under incomplete samples is an inquisitive yet unaddressed area of research. Therefore, the aim of this article is to design a single-stage optimization procedure for clustering as well as simultaneous reconstruction of images without breaking the intrinsic spatial structure. The method employs the self-expressiveness property of submodules, and images are stacked as the lateral slices of a three-dimensional tensor. The proposed optimization method is designed to extract a sparse t -linear combination tensor with low multirank constraint, consisting of a unique set of linear coefficients in the form of mode-3 fibers and the spectral clustering is performed on these fibers. Simultaneously, the recovery of lost samples is accomplished by twisting the entire lateral slices of the data tensor and applying a low-rank approximation on each slice. The prominence of the proposed method lies in the simultaneous execution of data clustering and reconstruction of incomplete observations in a single step. Experimental results reveal the excellence of the proposed method over state-of-the-art clustering algorithms in the context of incomplete imaging data.

The massive generation of data, which includes images and videos, has made data management, analysis, information extraction difficult in recent years. To gather relevant information, this large amount of data needs to be grouped. Real-life data may be noise corrupted during data collection or transmission, and the majority of them are unlabeled, allowing for the use of robust unsupervised clustering techniques. Traditional clustering techniques, which vectorize the images, are unable to keep the geometrical structure of the images. Hence, a robust tensor-based submodule clustering method based on l12 regularization with improved clustering capability is formulated. The l12 induced tensor nuclear norm (TNN), integrated into the proposed method, offers better low rankness while retaining the self-expressiveness property of submodules. Unlike existing methods, the proposed method employs a simultaneous noise removal technique by twisting the lateral image slices of the input data tensor into frontal slices and eliminates the noise content in each image, using the principles of the sparse and low rank decomposition technique. Experiments are carried out over three datasets with varying amounts of sparse, Gaussian and salt and pepper noise. The experimental results demonstrate the superior performance of the proposed method over the existing state-of-the-art methods.

The massive generation of data, which includes images and videos, has made data management, analysis, information extraction difficult in recent years. To gather relevant information, this large amount of data needs to be grouped. Real-life data may be noise corrupted during data collection or transmission, and the majority of them are unlabeled, allowing for the use of robust unsupervised clustering techniques. Traditional clustering techniques, which vectorize the images, are unable to keep the geometrical structure of the images. Hence, a robust tensor-based submodule clustering method based on l12 regularization with improved clustering capability is formulated. The l12 induced tensor nuclear norm (TNN), integrated into the proposed method, offers better low rankness while retaining the self-expressiveness property of submodules. Unlike existing methods, the proposed method employs a simultaneous noise removal technique by twisting the lateral image slices of the input data tensor into frontal slices and eliminates the noise content in each image, using the principles of the sparse and low rank decomposition technique. Experiments are carried out over three datasets with varying amounts of sparse, Gaussian and salt and pepper noise. The experimental results demonstrate the superior performance of the proposed method over the existing state-of-the-art methods.

Due to the presence of diverse and disparate patterns, the segmentation of natural images endures crucial as well as a challenging problem in image analysis algorithms. The proposed method addresses image segmentation as a subspace clustering of image feature vectors. Initially, an image is partitioned into superpixels and further, a feature data matrix is computed using the Local Spectral Histogram (LSH) features from individual superpixels. A single-stage optimization model is formulated which incorporates better subspace selection, excellent grouping effect and simultaneous noise robustness for the uncorrelated, correlated and corrupted data by the conjunctive venture of l 1 , l 2 and l 2,1 norm minimization. The proposed model is solved using Augmented Lagrangian technique. We compared the proposed method with state-of-the-art methods and the results demonstrate the improved performance of our proposed model over the existing counterparts.

Cell-free massive multiple-input multiple-output (MIMO) is a promising cellular technology in which a large number of distributed access points (APs) jointly serve a small number of user equipments (UEs). In this work, we investigate the impact of the spatial distribution of APs on the performance of a cell-free massive MIMO system, considering different downlink power control policies. Further, we analyze the performance for the case where only subsets of all APs serve the individual UEs: this scenario has lower backhaul requirements and associated CAPEX/OPEX costs. In this framework, we first propose a novel, tractable approximation for the average spectral efficiency (SE) of the transmission to a UE conditioned on the estimated channel gains. This approximation is then used to develop different downlink power control policies. Further, we extend the policies to the scenario when power control is coordinated only among subsets of APs and not across the subsets. Through extensive system-level simulations, we evaluate the improvement in SE by spreading out the APs and the SE loss when a subset of all APs serve a UE and coordination across subsets is absent.

Cloud radio access network (C-RAN) and massive multiple-input-multiple-output (MIMO) are two key enabling technologies for 5G as they improve radio performance while lowering the cost of operation. In a C-RAN system with massive MIMO-based remote radio units (RRUs), fronthaul is often the bottleneck in practice due to its finite capacity. To reduce the capacity requirements on fronthaul, precoding is done at the RRUs. In this paper, we maximize the energy efficiency (EE) of such a system by optimizing the transmit powers while explicitly incorporating the capacity constraints on fronthaul. We develop a successive convex approximation (SCA) algorithm, which is guaranteed to converge to a local optimum. Towards this, we propose novel bounds on the user rate function, which facilitates a convex approximation of the EE maximization problem. The convex problem is solved in each SCA iteration through Dinkel-bach’s algorithm and dual decomposition. Numerical results show that the proposed algorithm significantly improves EE compared to the case with no power control and outperforms the weighted minimum mean square error algorithm.

Latency is the one of the critical performance metrics for 5G and beyond mobile networks, particularly for ultra-reliable and low-latency communications (URLLC). In URLLC applications, it is required that the transmitted packets reach the destination within a certain time and the packets that are unable to meet this strict latency requirement will be discarded. In this paper, we compute the waiting time in the packetized fronthaul at the Ethernet switch and compute packet loss rate (PLR) incurred due to the inability of the transmitted packets to meet the FH latency threshold. In addition, we derive the tractable closed-form solution for the waiting time distribution and verify it with simulation results. Our results show that PLR is affected mainly by packet size, spectral efficiency, switch speed and arrival rate.

With the emergence of cloud radio access network (C-RAN) architecture, latency in fronthaul (FH) network is a critical performance metric especially for ultra-reliable and low-latency communication applications. The stringent FH capacity and latency requirements of C-RAN can be relaxed by offloading some baseband functionalities to remote radio unit (RRU), referred to as functional splitting. This allows packetized FH network solutions such as ubiquitous Ethernet. In this paper, we calculate the FH latency in the uplink of a C-RAN system with massive MIMO-based RRUs and 3GPP functional Split 7, wherein MIMO equalization is done at the RRU. We derive tractable, closed-form expressions for the steady-state probabilities of queue length and sojourn time distribution at the output port of an Ethernet switch in the FH network. We first present these results for Poisson file arrivals from users in the network and exponential file size distribution. We then extend the results to general file size distribution. The numerical results show that the file size and spectral efficiency of the users are critical in determining the FH latency. Further, results show that switch speed can be decreased without incurring significant increase in FH latency revealing the possibility for statistical multiplexing gains.

In contemporary orthogonal frequency division multiplexing (OFDM) systems, such as Long Term Evolution (LTE), LTE-Advanced, and WiMAX, a codeword is transmitted over a group of subcarriers. Since different subcarriers see different channel gains in frequency-selective channels, the modulation and coding scheme (MCS) of the codeword must be selected based on the vector of signal-to-noise-ratios (SNRs) of these subcarriers. Exponential effective SNR mapping (EESM) maps the vector of SNRs into an equivalent flat-fading SNR, and is widely used to simplify this problem. We develop a new analytical framework to characterize the throughput of EESM-based rate adaptation in such wideband channels in the presence of feedback delays. We derive a novel accurate approximation for the throughput as a function of feedback delay. We also propose a novel bivariate gamma distribution to model the time evolution of EESM between the times of estimation and data transmission, which facilitates the analysis. These are then generalized to a multi-cell, multi-user scenario with various frequency-domain schedulers. Unlike prior work, most of which is simulation-based, our framework encompasses both correlated and independent subcarriers and various multiple antenna diversity modes; it is accurate over a wide range of delays.

In orthogonal frequency division multiplexing systems, reduced feedback schemes provide essential channel state information from the users to the base station (BS) without overwhelming the uplink. For the practically important best- $M$ scheme, in which each user feeds back only its $M$ strongest subchannels and their indices to the BS, we derive a novel, throughput-optimal scheduling and rate adaptation policy that enables the BS to schedule the best user and its data rate for all the subchannels. The policy exploits the structure of the information fed back by the best- $M$ scheme and the correlation among subchannel gains. We present it in closed-form for the widely studied exponential correlation model. Using insights gleaned from the optimal policy, we propose a novel, low-complexity two subchannel reduction approach, which is seen empirically to be near-optimal and easily handles practically important general channel correlation models, quantized feedback, and co-channel interference in multi-cell scenarios. Compared with several ad hoc approaches, the proposed approaches improve the cell throughput without any additional feedback. A modified gradient-based opportunistic scheduler is also proposed to ensure user fairness.

channel which convey a major part of information transfer in our everyday life. Hand gestures are the natural way of interactions when one person is communicating with one another and therefore hand movements can be treated as a non verbal form of communication. Hand gesture recognition is a process of understanding and classifying meaningful movements by the human hands. Nowadays vehicles launched from the industry offers an increasing number of infotainment systems as well as comfort functions that can be controlled by the driver. Though they are feature rich which demands more attention of the driver and degrade the driving performance and thereby reducing the safety. The gestural interaction is a promising means to cover the full range of driver’s operational needs while minimizing the visual workload. Hand gesture recognition is of great importance for human computer interaction (HCI) because of its extensive applications in virtual reality and sign language recognition etc. Human hand is very smaller with very complex articulations comparing with the entire human body and therefore errors can be easily affected. It is thus a very challenging problem to recognize hand gestures. This paper comprises of the existing methods in detecting and recognizing hand gestures and a detailed study on their performances, accuracy, convenience, operational range and design challenges etc.

In rate-adaptive orthogonal frequency division multiplexing (OFDM) systems, limited feedback schemes are essential to reduce the number of subchannels for which the channel state information is fed back by the users. For the practically important best-M scheme, in which each user feeds back only its M strongest subchannels and their indices to the base station (BS), we derive a throughput-optimal rate adaptation policy that enables the BS to assign rates to the subchannels of every user. We present it in closed-form for the widely used exponential correlation model. The novelty of the policy lies in its exploitation of the structure of the information fed back by the best-M scheme and the correlation among subchannel gains. We also present a near-optimal, lower computational complexity approach. In effect, our approach facilitates rate adaptation and scheduling by the BS even on subchannels that are not fed back by a user due to feedback constraints. For various schedulers, we show that it improves the downlink throughput compared to several conventional approaches, without requiring any additional feedback.

Time-variations in wireless channels affect opportunistic selection in a multi-node wireless system in two ways. First, the selected node can become sub-optimal by the time data transmission commences. Second, the channel changes during data transmission. We develop a comprehensive and tractable analytical framework that accurately accounts for both these effects. It differs from the extensive existing literature that primarily focuses on time-variations until the data transmission starts. We first develop a novel concept of a time-invariant effective signal-to-noise ratio (TIESNR), which tractably and accurately captures the time-variations during the data transmission phase with partial channel state information available at the receiver. Thereafter, we model the joint distribution of the signal-to-noise ratio at the time of selection and TIESNR during the data transmission and analyze the average packet error rate (PER). Extensive numerical results verify the accuracy of each step of our approach and show that ignoring the correlated time-variations during the data transmission phase can significantly underestimate the average PER.

Images convey relevant data and information in biological sciences. Digital image processing and the image analysis technology have a vital role in biology and agricultural sectors. Automatic detection of plant diseases and cultivation of healthy plants is of great importance and agricultural automation.[1] The case of a plant, the term disease is defined as any impairment happening to the normal physiological function, producing characteristic symptoms. The studies of plant diseases refer to studying the visually observable patterns of a particular plant. The identification of plants, leaves, stems and finding out the pests or diseases, or its percentage is found very effective in the successful cultivation of crops. The naked eye observation is the approach adopted by many of the farmers for the detection and identification of plant diseases. It requires continuous monitoring and found less useful on large farms. Also, the farmers are unaware of non-native diseases. [1,2] With the aid of imaging technology the plant disease detection systems automatically detect the symptoms that appear on the leaves and stem of a plant and helps in cultivating healthy plants in a farm. These systems monitor the plant such as leaves and stem and any variation observed from its characteristic features, variation will be automatically identified and also will be informed to the user. This paper provides an evaluative study on the existing disease detection systems in plants.

Reduced feedback schemes play a critical role in orthogonal frequency division multiplexing-based cellular systems because they facilitate scheduling and rate adaptation by the base station (BS) while reducing the number of subchannels for which channel state information is fed back by the users. We address the problem of reliable transmission even on subchannels that are fed back by a few users due to feedback constraints. For the practically relevant best- $M$ feedback scheme, in which each user reports only its $M$ strongest subchannels to the BS, we derive a nonlinear constrained minimum mean square error estimator that enables the BS to estimate the signal-to-noise-ratios of all the subchannels of every user. We then propose two lower computational complexity approaches that incur a negligible loss in performance. The novelty of these approaches lies in their exploitation of the structure of the best- $M$ feedback information and the correlation among subchannel gains. Applications to general channel models and to quantized feedback are also shown. In terms of system-level impact, the proposed approaches improve the cell throughput compared to several conventional approaches – without requiring any additional feedback – for uncorrelated and correlated subchannels, and for various schedulers.

Managing inter-cell interference is one of the main challenges in current and next generation wireless systems that aggressively reuse the frequency. Cooperation between interfering cells has been sought to mitigate interference. In this paper, we address the problem of jointly optimizing the transmit powers, user scheduling, and user association in a cellular network to maximize the weighted sum rate (WSR). To this end, we develop a distributed interference penalty algorithm in which the cells update their transmit powers and user schedule to maximize its utility minus an interference cost. The proposed algorithm involves only limited exchange of information via backhaul and has convergence guarantees. Furthermore, we propose a sub-optimal algorithm with lower computational and backhaul overhead. In it, the users are first associated to the base stations (BSs) based on their signal-to-interference-plus-noise-ratios (SINRs). It is then followed by joint optimization of BS transmit powers and user scheduling, for which we develop an interference penalty algorithm as well. We show that the proposed algorithms outperform the computationally complex weighted minimum mean squared error (WMMSE) algorithm.

In orthogonal frequency division multiplexing (OFDM) systems, such as long term evolution (LTE) and WiMAX, a codeword is transmitted over a group of subcarriers. Since the subcarriers see different channel gains in a frequency-selective channel, the modulation and coding scheme (MCS) of the codeword must be selected based on the vector of signal-to-noise-ratios (SNRs) of these subcarriers. Exponential effective SNR mapping (EESM) simplifies this problem by mapping the vector of SNRs into a single, equivalent flat-fading SNR. We develop a new analytical framework to characterize the throughput of EESM-based rate adaptation in such wideband channels in the presence of feedback delays, which make the choice of the MCS partially outdated by the time data transmission takes place. To this end, we first propose a novel bivariate gamma distribution to model the joint statistics of EESM at the times of estimation and data transmission. We then derive a novel expression for the throughput as a function of feedback delay. Our framework works for both correlated and independent subcarriers and for various multiple antenna diversity modes, and is accurate over a wide range of delays.

In contemporary wideband orthogonal frequency division multiplexing (OFDM) systems, such as Long Term Evolution (LTE) and WiMAX, different subcarriers over which a codeword is transmitted may experience different signal-to-noise-ratios (SNRs). Thus, adaptive modulation and coding (AMC) in these systems is driven by a vector of subcarrier SNRs experienced by the codeword, and is more involved. Exponential effective SNR mapping (EESM) simplifies the problem by mapping this vector into a single equivalent flat-fading SNR. Analysis of AMC using EESM is challenging owing to its non-linear nature and its dependence on the modulation and coding scheme. We first propose a novel statistical model for the EESM, which is based on the Beta distribution. It is motivated by the central limit approximation for random variables with a finite support. It is simpler and as accurate as the more involved ad hoc models proposed earlier. Using it, we develop novel expressions for the throughput of a point-to-point OFDM link with multi-antenna diversity that uses EESM for AMC. We then analyze a general, multi-cell OFDM deployment with co-channel interference for various frequency-domain schedulers. Extensive results based on LTE and WiMAX are presented to verify the model and analysis, and gain new insights.