To meet the ever increasing spectrum demand, developing a mechanism for dynamic spectrum access seems inevitable. Spectrum sensing enables cognitive radios (CRs) to identify and use frequency bands (channels) that are not being used by primary users (PUs) at a particular place and time. However, sensing errors and limited sensing resources, such as sensing hardware and sensing time, introduce significant technical challenges to the development of such an ideal capability. Adaptive sensing strategies allow the sensing resources to be spent on more promising primary channels. This is achieved by exploiting past sensing outcomes of one secondary user (SU), or, as proposed in this research, multiple spatially distributed SUs. We propose adaptive sensing strategies for three different scenarios. First, we assume that a SU sequentially senses a number of primary channels to find the first available channel. We propose a two-stage spectrum detection strategy that allows the spectrum detector to quickly detect and skip though most of busy channels and spend most of its time on channels that are more likely to be idle. Second, we consider the case where multiple SUs jointly try to locate idle channels within a given sensing time, which itself is divided into a number of sensing slots. We propose a cooperative spectrum search strategy that specifies the channel to be sensed by each SU in each slot in such a way to maximize the expected number of identified idle channels. Third, we consider a primary network that operates in a synchronous time-framed fashion. We assume that the occupancy state of each primary channel over different time frames follows a discrete-time Markov process. We propose a cooperative sensing strategy that decides which channel should be sensed by which SU in each frame. The goal is to maximize a utility function that accounts for both the number of detected idle channel-frames and the number of miss-detected busy channel-frames. We present analytical and numerical results to demonstrate the effectiveness of the proposed sensing strategies in increasing identified time-frequency spectrum opportunities and/or reducing interference with licensed systems.

The successful implementation and employment of various cognitive radio services are largely dependent on the spectrum sensing performance of the cognitive radio terminals. Previous works on detection of cognitive radio have suggested the necessity of user cooperation in order to be able to detect at low signal-to-noise ratios experienced in practical situations. This report provides a brief overview of the impact of different fusion strategies on the spectrum hole detection performance of a fusion center in a distributed detection environment. Different decision or detection rule and fusion strategies, like single sensor scenario, counting rule, and linear decision metric, were used to analyze their influence on the spectrum sensing performance of the cognitive radio network. We consider a system of cognitive radio users who cooperate with each other in trying to detect licensed transmissions. Assuming that the cooperating nodes use identical energy detectors, we model the received signals as correlated log-normal random variables and study the problem of fusing the decisions made by the individual nodes. The cooperating radios were assumed to be designed in such a way that they satisfy the interference probability constraint individually. The interference probability constraint was also met at the fusion center. The simulation results strongly suggests that even when the observations at the individual sensors are moderately correlated, it is important not to ignore the correlation between the nodes for fusing the local decisions made by the secondary users. The thesis mainly focuses on the performance measurement of linear decision combiner in detecting primary users in a cognitive radio network.

abstract: This thesis investigates three different resource allocation problems, aiming to achieve two common goals: i) adaptivity to a fast-changing environment, ii) distribution of the computation tasks to achieve a favorable solution. The motivation for this work relies on the modern-era proliferation of sensors and devices, in the Data Acquisition Systems (DAS) layer of the Internet of Things (IoT) architecture. To avoid congestion and enable low-latency services, limits have to be imposed on the amount of decisions that can be centralized (i.e. solved in the ``cloud") and/or amount of control information that devices can exchange. This has been the motivation to develop i) a lightweight PHY Layer protocol for time synchronization and scheduling in Wireless Sensor Networks (WSNs), ii) an adaptive receiver that enables Sub-Nyquist sampling, for efficient spectrum sensing at high frequencies, and iii) an SDN-scheme for resource-sharing across different technologies and operators, to harmoniously and holistically respond to fluctuations in demands at the eNodeB' s layer. The proposed solution for time synchronization and scheduling is a new protocol, called PulseSS, which is completely event-driven and is inspired by biological networks. The results on convergence and accuracy for locally connected networks, presented in this thesis, constitute the theoretical foundation for the protocol in terms of performance guarantee. The derived limits provided guidelines for ad-hoc solutions in the actual implementation of the protocol. The proposed receiver for Compressive Spectrum Sensing (CSS) aims at tackling the noise folding phenomenon, e.g., the accumulation of noise from different sub-bands that are folded, prior to sampling and baseband processing, when an analog front-end aliasing mixer is utilized. The sensing phase design has been conducted via a utility maximization approach, thus the scheme derived has been called Cognitive Utility Maximization Multiple Access (CUMMA). The framework described in the last part of the thesis is inspired by stochastic network optimization tools and dynamics. While convergence of the proposed approach remains an open problem, the numerical results here presented suggest the capability of the algorithm to handle traffic fluctuations across operators, while respecting different time and economic constraints. The scheme has been named Decomposition of Infrastructure-based Dynamic Resource Allocation (DIDRA). / Dissertation/Thesis / Doctoral Dissertation Electrical Engineering 2017

Cognitive Radio (CR) aims to access the wireless spectrum in an opportunistic manner while the licensed user is not using it. To accurately determine the licensed user's existence, spectrum sensing procedure is vital to CR system. Energy detection-based spectrum sensing techniques is favourable due to its simplicity and low complexity. In addition, to improve the detection performance, cooperative spectrum sensing technique exploits multi-user diversity and mitigates detection uncertainty. In this thesis, we investigate several energy detection based cooperative spectrum sensing techniques.First, the closed-form analysis for the Equal Gain Combining based Soft Decision Combining (EGC-SDC) scheme, in which all CR users forward its observation to the fusion center, is derived. In order to reduce the communication overhead between CR users and the fusion center, we proposed quantized cooperative spectrum sensing technique, in which CR users quantize its local observation before forwarding to the fusion center. Next, the Double Threshold scheme, where some users only forward its local decision while other users forward its observation, is considered and analyzed. To further reduce the communication overhead, we also proposed that quantization is applied to the users who forward its observation. Later on, three sequential cooperative spectrum sensing schemes in time-varying channel are considered. By aggregating past local observations from previous sensing slots, CR users can improve the detection performance. The Weighted Sequential Energy Detector (SED) scheme simply takes fixed number of past local observations, while the other two schemes, Two-Stage SED and Differential SED, adaptively determine the number of observations, based on its decision towards primary user's existence.Simulation results show that the analysis on EGC-SDC scheme is accurate and the quantized cooperative spectrum sensing technique can improve the performance and approach the detection performance of EGC-SDC scheme with much less bandwidth requirement. Also, the Double Threshold scheme can help improve the detection performance over the conventional technique. Furthermore, the analysis on Double Threshold provides a closed-form for the probability of false alarm and detection. Additionally, the sequential spectrum sensing schemes are shown to improve the detection performance and enable CR system to work in scenarios that the conventional technique can not accommodate.

No description available.

Cognitive radio had been proposed as a methodology for overcoming the inefficiency of the conventional static allocation of the available spectrum in wireless communication networks. The majority of opportunistic spectrum access schemes in cognitive radio networks (CRNs) rely on the Listen-Before-Talk (LBT) model due to the half-duplex nature of conventional wireless radios. However, LBT su ers from the problem of high collision rates and low secondary user throughput if time is misaligned among the secondary users (SUs) and the primary users (PUs). This problem can be mitigated by leveraging full-duplex (FD) communications that facilitate concurrent sensing and transmission. This thesis considers the problem of optimal opportunistic multi-channel spectrum sensing and access using FD radios in the presence of uncertain primary user (PU) activity statistics. A joint learningand spectrum access scheme is proposed. To optimize its throughput, the SU sensing period has to be carefully tuned. However, in absence of exact knowledge of the PU activity statistics, the PU's performance may be adversely a ected. To address this problem, a robust optimization problem is formulated. Analysis shows that under some non-restrictive simplifying assumptions, the robust optimization problem is convex. The impact of sensing periods on the PU collision probability and the SU throughput are analyzed, and the optimal sensing period is found via convex optimization. An "\epsilon-greedy algorithm is proposed for use by the SU to learn the PUs' activity statistics in multichannel networks. It is shown that sublinear regrets can be attained by the proposed estimation and robust optimization strategy. Simulation studies demonstrate that the resulting robust solution achieves a good trade-o between optimizing the SU's throughput and protecting the PU. / Thesis / Master of Applied Science (MASc)

No / Abstract: The scarcity of frequency spectrum used for wireless communication systems has attracted a considerable amount of attention in recent years. The cognitive radio (CR) terminology has been widely accepted as a smart platform mainly aimed at the efficient interrogation and utilization of permitted spectrum. Non-parametric spectrum sensing, or estimation, represents one of the prominent tools that can be proposed when CR works under an undetermined environment. As such, the periodogram, filter bank, and multi-taper methods are well considered in many studies without relying on the transmission channel's characteristics. A unified approach to all these non-parametric spectrum sensing techniques is presented in this paper with analytical and performance comparison using simulation methods. Results show that the multi-taper method outperforms the others.

No / Eigenvalue based spectrum sensing schemes such as Maximum Minimum Eigenvalue (MME), Maximum Energy Detection (MED) and Energy with Minimum Eigenvalue (EME) have higher spectrum sensing performance without requiring any prior knowledge of Primary User (PU) signal but the decision hypothesis used in these eigenvalue based sensing schemes depends on the calculation of maximum eigenvalue from covariance matrix of measured signal. Calculation of the covariance matrix followed by eigenspace analysis of the covariance matrix is a resource intensive operation and takes overhead time during critical process of spectrum sensing. In this paper we propose a new blind spectrum sensing scheme based on the approximation of the maximum eigenvalue using state of the art results from Random Matrix Theory (RMT). The proposed sensing scheme has been evaluated through extensive simulations on wireless microphone signals and the proposed scheme shows higher probability of detection (Pd) performance. The proposed spectrum sensing also shows higher detection performance as compared to energy detection scheme and RMT based sensing schemes such as MME and EME.

This paper describes the implementation of a scanning signal detector and automatic modulation classification system. The classification technique is a completely blind method, with no prior knowledge of the signal's center frequency, bandwidth, or symbol rate. An energy detector forms the initial approximations of the signal parameters. The energy detector used in the wideband sweep is reused to obtain fine estimates of the center frequency and bandwidth of the signal. The subsequent steps reduce the effect of frequency offset and sample timing error, resulting in a constellation of the modulation of interest. The cumulant of the constellation is compared to a set of known ideal cumulant values, forming the classification estimate. The algorithm uses two platforms that together provide high speed parallel processing and flexible run-time operation. High-rate spectral scanning using an energy detector is run in parallel with a variable down sampling path; both are highly pipelined structures, which allows for high data throughput. A pair of processing cores is used to record spectral usage and signal characteristics as well as perform the actual classification. The resulting classification system can accurately identify modulations below 5 dB of signal-to-noise ratio (SNR) for some cases of the phase shift keying family of modulations but requires a much higher SNR to accurately classify higher-order modulations. These estimates tend toward classifying all signals as binary phase shift keying because of limits of the noise power estimation part of the cumulant normalization process. Other effects due to frequency offset and synchronization timing are discussed. / Master of Science

Cognitive radios require spectral occupancy information in a given location, to avoid any interference with the existing licensed users. This is achieved by spectrum sensing. Existing narrowband, serial spectrum sensors are spectrally inefficient and power hungry. Wideband spectrum sensing increases the number of probable fre-quency candidates for cognitive radio. Wideband RF systems cannot use analog to digital converters (ADCs) for spectrum sensing without increasing the sampling rate and power consumption. The use of ADCs is limited because of the dynamic range of the signals that need to be sampled and the frequency of operation. In this work, we have presented a CMOS based area efficient, dedicated and scalable wideband parallel/serial spectrum sensor for cognitive radio. The key contributions of the thesis are: 1. An injection locked oscillator cascade (ILOC) for parallel LO synthesis. An area-efficient, wideband RF frequency synthesizer, which simultaneously gen-erates multiple local oscillator (LO) signals, is designed. It is suitable for parallel wideband RF spectrum sensing in cognitive radios. The frequency synthesizer consists of an injection locked oscillator cascade where all the LO signals are derived from a single reference oscillator. The ILOC is implemented in a 130-nm technology with an active area of 0.017 mm2. It generates 4 uni-formly spaced LO carrier frequencies from 500 MHz to 2 GHz. 2. A wideband, parallel RF spectrum sensor for cognitive radios has been de-signed. This spectrum sensor is designed to detect RF occupancy from 250 MHz to 5.25 GHz by using an array of CMOS receivers with envelope detec-tors. A parallel LO synthesizer is implemented as an ILOC. The simulated sensitivity is around -25 dBm for 250 MHz wide bandwidth. 3. A mitigation technique for harmonic downconversion in wideband spectrum sensors. The downconversion of radio frequency (RF) components around the harmonics of the local oscillator (LO), and its impact on the accuracy of white space detection using integrated spectrum sensors, is (are) studied. We propose an algorithm to mitigate the impact of harmonic Down conversion by utilizing multiple parallel downconverters in the system architecture. The proposed algorithm is validated on a test-board using commercially avail-able integrated circuits (IC) and a test-chip implemented in a 130-nm CMOS technology. The measured data shows that the impact of the harmonic down-conversion is closely related to the LO characteristics, and that much of it can be mitigated by the proposed technique. 4. A wideband spectrum sensor for narrowband energy detection. A wideband spectrum sensing system for cognitive radio is designed and implemented in a 130-nm RF mixed-mode CMOS technology. The system employs an I-Q downconverter, a pair of complex filters and a pair of envelope detectors for energy detection. The spectrum sensor works from 250 MHz to 3.25 GHz. The design makes use of the band pass nature of the complex filter to achieve two objectives : i) Separation of upper sideband (USB) and lower sideband (LSB) around the local oscillator (LO) signal and ii) Resolution of smaller bands within a large detection bandwidth. The measured sensitivity is close to -45 dBm for a single tone test over a bandwidth of 40 MHz. The measured Image reject ratio (IRR) is close to 30 dB. The overall sensing bandwidth is 3.5 GHz and the overall wideband detection bandwidth is 250 MHz which is partitioned into 40 MHz narrowband chunks with 8 such overlapping chunks.