There is no doubt that big data are now rapidly expanding in all science and engineering domains. While the potential of these massive data is undoubtedly significant, fully making sense of them requires new ways of thinking and novel learning techniques to address the various challenges. In this paper, we present a literature survey of the latest advances in researches on machine learning for big data processing. First, we review the machine learning techniques and highlight some promising learning methods in recent studies, such as representation learning, deep learning, distributed and parallel learning, transfer learning, active learning, and kernel-based learning. Next, we focus on the analysis and discussions about the challenges and possible solutions of machine learning for big data. Following that, we investigate the close connections of machine learning with signal processing techniques for big data processing. Finally, we outline several open issues and research trends.

/pdf/a-survey-of-machine-learning-for-big-data-processing-4nbp71qkyz.pdf

A survey of machine learning for big data processing

The wireless medium being inherently broadcast in nature and hence prone to interferences requires highly optimized medium access control (MAC) protocols. This holds particularly true for wireless sensor networks (WSNs) consisting of a large amount of miniaturized battery-powered wireless networked sensors required to operate for years with no human intervention. There has hence been a growing interest on understanding and optimizing WSN MAC protocols in recent years, where the limited and constrained resources have driven research towards primarily reducing energy consumption of MAC functionalities. In this paper, we provide a comprehensive state-of-the-art study in which we thoroughly expose the prime focus of WSN MAC protocols, design guidelines that inspired these protocols, as well as drawbacks and shortcomings of the existing solutions and how existing and emerging technology will influence future solutions. In contrast to previous surveys that focused on classifying MAC protocols according to the technique being used, we provide a thematic taxonomy in which protocols are classified according to the problems dealt with. We also show that a key element in selecting a suitable solution for a particular situation is mainly driven by the statistical properties of the generated traffic.

/pdf/mac-essentials-for-wireless-sensor-networks-7v5zc579kv.pdf

MAC Essentials for Wireless Sensor Networks

A resource allocation framework is presented for spectrum underlay in cognitive wireless networks. We consider both interference constraints for primary users and quality of service (QoS) constraints for secondary users. Specifically, interference from secondary users to primary users is constrained to be below a tolerable limit. Also, signal to interference plus noise ratio (SINR) of each secondary user is maintained higher than a desired level for QoS insurance. We propose admission control algorithms to be used during high network load conditions which are performed jointly with power control so that QoS requirements of all admitted secondary users are satisfied while keeping the interference to primary users below the tolerable limit. If all secondary users can be supported at minimum rates, we allow them to increase their transmission rates and share the spectrum in a fair manner. We formulate the joint power/rate allocation with proportional and max-min fairness criteria as optimization problems. We show how to transform these optimization problems into a convex form so that their globally optimal solutions can be obtained. Numerical results show that the proposed admission control algorithms achieve performance very close to that of the optimal solution. Also, impacts of different system and QoS parameters on the network performance are investigated for the admission control, and rate/power allocation algorithms under different fairness criteria.

Resource allocation for spectrum underlay in cognitive radio networks

Cognitive radio (CR) has emerged as a promising technology to exploit the unused portions of spectrum in an opportunistic manner. The fixed spectrum allocation of governmental agencies results in unused portions of spectrum, which are called "spectrum holes" or "white spaces". CR technology overcomes this issue, allowing devices to sense the spectrum for unused portions and use the most suitable ones, according to some pre-defined criteria. Spectrum assignment is a key mechanism that limits the interference between CR devices and licensed users, enabling a more efficient usage of the wireless spectrum. Interference is a key factor that limits the performance in wireless networks. The scope of this work is to give an overview of the problem of spectrum assignment in cognitive radio networks, presenting the state-of-the-art proposals that have appeared in the literature, analyzing the criteria for selecting the most suitable portion of the spectrum and showing the most common approaches and techniques used to solve the spectrum assignment problem. Finally, an analysis of the techniques and approaches is presented, discussing also the open issues for future research in this area.

Spectrum Assignment in Cognitive Radio Networks: A Comprehensive Survey

Spectrum decision is the ability of a cognitive radio (CR) to select the best available spectrum band to satisfy secondary users' (SUs') quality of service (QoS) requirements, without causing harmful interference to licensed or primary users (PUs). Each CR performs spectrum sensing to identify the available spectrum bands and the spectrum decision process selects from these available bands for opportunistic use. Spectrum decision constitutes an important topic which has not been adequately explored in CR research. Spectrum decision involves spectrum characterization, spectrum selection and CR reconfiguration functions. After the available spectrum has been identified, the first step is to characterize it based not only on the current radio environment conditions, but also on the PU activities. The second step involves spectrum selection, whereby the most appropriate spectrum band is selected to satisfy SUs' QoS requirements. Finally, the CR should be able to reconfigure its transmission parameters to allow communication on the selected band. Key to spectrum characterization is PU activity modelling, which is commonly based on historical data to provide the means for predicting future traffic patterns in a given spectrum band. This paper provides an up-to-date survey of spectrum decision in CR networks (CRNs) and addresses issues of spectrum characterization (including PU activity modelling), spectrum selection and CR reconfiguration. For each of these issues, we highlight key open research challenges. We also review practical implementations of spectrum decision in several CR platforms.

/pdf/spectrum-decision-in-cognitive-radio-networks-a-survey-4zyh1oas7j.pdf

Spectrum Decision in Cognitive Radio Networks: A Survey

The primary function of a cognitive radio is to detect idle frequencies or sub-bands, not used by the primary users (PUs), and allocate these frequencies to secondary users. The state of the sub-band at any time point is either free (unoccupied by a PU) or busy (occupied by a PU). The states of a sub-band are monitored over L consecutive time periods, where each period is of a given time interval. Existing research assume the presence of a Markov chain for sub-band utilization by PUs over time, but this assumption has not been validated. Therefore, in this paper we validate existence of a Markov chain for sub-band utilization using real-time measurements collected in the paging band (928–948 MHz). Furthermore, since the detection of idle sub-bands by a cognitive radio is prone to errors, we probabilistically model the errors and then formulate a spectrum sensing paradigm as a Hidden Markov model that predicts the true states of a sub-band. The accuracy of our proposed method in predicting the true states of the sub-band is substantiated using extensive simulations.

Markov chain existence and Hidden Markov models in spectrum sensing

In this paper, we propose a novel spectrum occupancy model designed to generate accurate temporal and frequency behavior of various wireless transmissions. Our proposed work builds upon existing concepts in open literature in order to develop a more accurate time-varying spectrum occupancy model. This model can be employed by wireless researchers for evaluating new wireless communication and networking algorithms and techniques designed to perform dynamic spectrum access (DSA). Using statistical characteristics extracted from actual radio frequency measurements, first- and second-order parameters are employed in a statistical spectrum occupancy model based on a combination of several different probability density functions (PDFs) defining various features of a specific spectrum band with several concurrent transmissions. To assess the accuracy of the model, the output characteristics of the proposed spectrum occupancy model are compared with realtime radio frequency measurements in the television and paging bands.

A framework for statistical wireless spectrum occupancy modeling

A wireless sensor network typically consists of a dense deployment of sensor nodes to achieve higher resolution and better network coverage. Having a dense network increases the fault-tolerance and robustness of the system. However, if not properly handled, it can lead to more collisions during transmission and also network congestion. Furthermore, wireless communication is inherently unpredictable and error-prone. Hence, it is imperative to design an efficient medium access control (MAC) protocol that facilitates guaranteed delivery of data over unreliable wireless links. In this paper, we have designed an on-demand reliable MAC (RMAC) protocol that enables timely delivery of data. We have demonstrated its superior performance over existing reliability-enforcing approaches in terms of reliability, latency, scalability and energy-efficiency.

On-demand reliable medium access in sensor networks

In this paper, we present three novel priority-based spectrum allocation techniques for enabling dynamic spectrum access (DSA) networking for non-contiguous orthogonal frequency division multiplexing (NC-OFDM) transmission. With each communication link in the network possessing a specified pair of bit error rate (BER) and throughput requirements for supporting a specific application, the proposed technique assigns one or more blocks of wireless spectrum to these applications in an attempt to simultaneously satisfy these requirements. Specifically, the proposed techniques assigns blocks of spectrum possessing aggregate bandwidth that is sufficient for supporting the intended wireless data service over the communication link. Moreover, since several portions of the wireless spectrum may be heavily attenuated due to frequency-selective fading resulting from multipath propagation, communication links requiring high error robustness are assigned frequency bands located further away from these attenuated regions of spectrum. Thus, the proposed spectrum allocation techniques aims at accommodating communication links supporting several wireless services possessing different performance requirements.

Priority-based spectrum allocation for cognitive radio networks employing NC-OFDM transmission

In this paper, we provide a new perspective for an- alyzing spectrum occupancy by introduced an M/M/1 queueing model to generate accurate temporal and frequency behavior of various wireless transmissions. Our proposed research builds upon existing concepts in the open literature in order to develop a more accurate time-varying spectrum occupancy model. This model can be employed by wireless researchers for evaluating new wireless communication and networking algorithms and techniques designed to perform dynamic spectrum access (DSA). Using statistical characteristics extracted from actual radio fre- quency measurements, first- and second-order parameters are employed in a statistical spectrum occupancy model based on a combination of several different probability density functions (PDFs) defining various features of a specific spectrum band with several concurrent transmissions. To assess the accuracy of the model, the output characteristics of the proposed spectrum occupancy model are compared with real-time radio frequency measurements in the paging and ISM bands.

Chittabrata Ghosh

Papers

Markov chain existence and Hidden Markov models in spectrum sensing

A framework for statistical wireless spectrum occupancy modeling

On-demand reliable medium access in sensor networks

Priority-based spectrum allocation for cognitive radio networks employing NC-OFDM transmission

Queueing Theory Representation and Modeling of Spectrum Occupancy Employing Radio Frequency Measurements