Public Safety Cognitive Radio (PSCR) is a Software Defined Radio(SDR) developed by the Center for Wireless Telecommunications (CWT) at Virginia Tech. PSCR can configure itself to interoperate with any public safety waveform it finds during the scan procedure. It also offers users the capability to scan/classify both analog and digital waveforms. The current PSCR architecture can only run on a general purpose processor and hence is not deployable to the public safety personnel. In the first part of this thesis an Android based control application for the PSCR on a Beagle Board(BB) and the GUI for the control application are developed. The Beagle Board is a low-cost, fanless single board computer that unleashes laptop-like performance and expandability. The Android based Nexus One connected to the Beagle Board via USB is used to control the Beagle Board and enable operations like scan, classify, talk, gateway etc. In addition to the features that exist in the current PSCR a new feature that enables interoperation with P25 (CPFSK modulation) protocol based radios is added. In this effort of porting the PSCR to Beagle Board my contributions are the following (i) communication protocol between the Beagle Board and the Nexus One (ii) PSCR control application on the Android based Nexus One (iii) detection/classification of P25 protocol based radios. In the second part of this thesis, a prototype testbed of a Dynamic Spectrum Access (DSA) broker that uses the Beagle Board PSCR based sensor/classifier is developed. DSA in simple terms is a concept that lets the user without license (secondary user) to a particular frequency access that frequency, when the licensed user (primary user) is not using it. In the proposed testbed we have two Beagle Board based sensor/classifiers that cooperatively scan the spectrum and report the results to the central DSA broker. The DSA broker then identifies the frequency spectrum without primary users and informs the secondary users about the free spectrum. The secondary users can then communicate among each other using the frequency band allocated by the DSA broker. When the primary user enters the spectrum occupied by the secondary user, the DSA broker instructs the secondary user to use a different spectrum. Based on the experiments conducted on the testbed setup in the CWT lab environment, the average time taken by the DSA broker to detect the presence of primary user is 0.636 secs and the average time taken for the secondary user to leave the frequency band that interferes with the primary user is 0.653 secs. / Master of Science

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.

Cognitive Radio Networks (CRNs) are a next generation network that is expected to solve the wireless spectrum shortage problem, which is the shrinking of available wireless spectrum resources needed to facilitate future wireless applications. The first CRN standard, the IEEE 802.22, addresses this particular problem by allowing CRNs to share geographically unused TV spectrum to mitigate the spectrum shortage. Equipped with reasoning and learning engines, cognitive radios operate autonomously to locate unused channels to maximize its own bandwidth and Quality-of-Service (QoS). However, their increased capabilities over traditional radios introduce a new dimension of security threats. In an NSF 2009 workshop, the FCC raised the question, “What authentication mechanisms are needed to support cooperative cognitive radio networks? Are reputation-based schemes useful supplements to conventional Public Key Infrastructure (PKI) authentication protocols?” Reputation-based schemes in cognitive radio networks are a popular technique for performing robust and accurate spectrum sensing without any inter-communication with licensed networks, but the question remains on how effective they are at satisfying the FCC security requirements. Our work demonstrates that trust-based Cooperative Spectrum Sensing (CSS) protocols are vulnerable to rogue signals, which creates the illusion of inside attackers and raises the concern that such schemes are overly sensitive Intrusion Detection Systems (IDS). The erosion of the sensor reputations in trust-based CSS protocols makes CRNs vulnerable to future attacks. To counter this new threat, we introduce community detection and cluster analytics to detect and negate the impact of rogue signals on sensor reputations.

Reliable wireless networks for high speed trains require a significant amount of data communications for enabling safety features such as train collision avoidance and railway management. Cognitive radio integrates heterogeneous wireless networks that will be deployed in order to achieve intelligent communications in future railway systems. One of the primary technical challenges in achieving reliable communications for railways is the handling of high mobility environments involving trains, which includes significant Doppler shifts in the transmission as well as severe fading scenarios that makes it difficult to estimate wireless spectrum utilization. This thesis has two primary contributions: (1) The creation of a Heterogeneous Cooperative Spectrum Sensing (CSS) prototype system, and (2) the derivation of a Long Term Evolution for Railways (LTE-R) system performance analysis. The Heterogeneous CSS prototype system was implemented using Software-Defined Radios (SDRs) possessing different radio configurations. Both soft and hard-data fusion schemes were used in order to compare the signal source detection performance in real-time fading scenarios. For future smart railways, one proposed solution for enabling greater connectivity is to access underutilized spectrum as a secondary user via the dynamic spectrum access (DSA) paradigm. Since it will be challenging to obtain an accurate estimate of incumbent users via a single-sensor system within a real-world fading environment, the proposed cooperative spectrum sensing approach is employed instead since it can mitigate the effects of multipath and shadowing by utilizing the spatial and temporal diversity of a multiple radio network. Regarding the LTE-R contribution of this thesis, the performance analysis of high speed trains (HSTs) in tunnel environments would provide valuable insights with respect to the smart railway systems operating in high mobility scenarios in drastically impaired channels.

The focus of this thesis is on cooperative spectrum sensing and related security issues in multi-channel cognitive radio networks (MCCRNs). We first study the channel assignment for cooperative spectrum sensing in MCCRNs to maximize the number of available channels. In centralized implementation, a heuristic scheme is proposed along with a greedy scheme to reduce the reported information from the cognitive radios (CRs). In distributed scenario, a novel scheme with multi-round operation is designed following the coalitional game theory. Next, we focus on the physical layer security issues for cooperative spectrum sensing in MCCRNs, caused by Byzantine attacks. New counterattacks are proposed to combat attacks comprising coalition head and CRs as Byzantine attackers, which target to reduce the number of available channels for sensing in distributed MCCRNs. First, a new secure coalition head selection is proposed, by using statistical properties of the exchanged SNRs in the coalitions. Then, an iterative algorithm is proposed to block out attackers, if they continue attacking the system. The important problem of key management is considered next, and an energy-efficient identity-based and a certificate-based distributed key management schemes are proposed. First, a new elliptic curve cryptography (ECC)-based distributed private key generation scheme is proposed to combat the single point of failure problem along with novel distributed private key generator (DPKG) selection schemes to preserve security and energy-efficiency. Because of its importance in the proposed identity-based key management scheme, we further propose a low-complexity DPKG assignment, based on multi-objective programming, which can capture DPKG fairness in addition to energy-efficiency. Finally, a more powerful and intelligent distributed cooperative Byzantine attack on the proposed multi-channel cooperative spectrum sensing is proposed, where attackers collude by applying coalitional game theory to maximize the number of invaded channels in a distributed manner. As a remedy, a hierarchical identity-based key management scheme is proposed, in which CRs can only play on a certain number of requested channels and channel access for sensing is limited to the honest CRs selected in the coalitional game. Simulation results show that the proposed schemes can significantly improve cooperative spectrum sensing and secure the system against Byzantine attacks.

With the rapid proliferation of new wireless communication devices and services, the demand for the radio spectrum is increasing at a rapid rate, which leads to making the spectrum more and more crowded. The limited available spectrum and the inefficiency in the spectrum usage have led to the emergence of cognitive radio (CR) and dynamic spectrum access (DSA) technologies, which enable future wireless communication systems to exploit the empty spectrum in an opportunistic manner. To do so, future wireless devices should be aware of their surrounding radio environment in order to adapt their operating parameters according to the real-time conditions of the radio environment. From this viewpoint, spectrum sensing is becoming increasingly important to new and future wireless communication systems, which is designed to monitor the usage of the radio spectrum and reliably identify the unused bands to enable wireless devices to switch from one vacant band to another, thereby achieving flexible, reliable, and efficient spectrum utilisation. This thesis focuses on issues related to local and cooperative spectrum sensing for CR networks, which need to be resolved. These include the problems of noise uncertainty and detection in low signal to noise ratio (SNR) environments in individual spectrum sensing. In addition to issues of energy consumption, sensing delay and reporting error in cooperative spectrum sensing. In this thesis, we investigate how to improve spectrum sensing algorithms to increase their detection performance and achieving energy efficiency. To this end, first, we propose a new spectrum sensing algorithm based on energy detection that increases the reliability of individual spectrum sensing. In spite of the fact that the energy detection is still the most common detection mechanism for spectrum sensing due to its simplicity. Energy detection does not require any prior knowledge of primary signals, but has the drawbacks of threshold selection, and poor performance due to noise uncertainty especially at low SNR. Therefore, a new adaptive optimal energy detection algorithm (AOED) is presented in this thesis. In comparison with the existing energy detection schemes the detection performance achieved through AOED algorithm is higher. Secondly, as cooperative spectrum sensing (CSS) can give further improvement in the detection reliability, the AOED algorithm is extended to cooperative sensing; in which multiple cognitive users collaborate to detect the primary transmission. The new combined approach (AOED and CSS) is shown to be more reliable detection than the individual detection scheme, where the hidden terminal problem can be mitigated. Furthermore, an optimal fusion strategy for hard-fusion based cognitive radio networks is presented, which optimises sensing performance. Thirdly, the need for denser deployment of base stations to satisfy the estimated high traffic demand in future wireless networks leads to a significant increase in energy consumption. Moreover, in large-scale cognitive radio networks some of cooperative devices may be located far away from the fusion centre, which causes an increase in the error rate of reporting channel, and thus deteriorating the performance of cooperative spectrum sensing. To overcome these problems, a new multi-hop cluster based cooperative spectrum sensing (MHCCSS) scheme is proposed, where only cluster heads are allowed to send their cluster results to the fusion centre via successive cluster heads, based on higher SNR of communication channel between cluster heads. Furthermore, in decentralised CSS as in cognitive radio Ad Hoc networks (CRAHNs), where there is no fusion centre, each cognitive user performs the local spectrum sensing and shares the sensing information with its neighbours and then makes its decision on the spectrum availability based on its own sensing information and the neighbours’ information. However, cooperation between cognitive users consumes significant energy due to heavy communications. In addition to this, each CR user has asynchronous sensing and transmission schedules which add new challenges in implementing CSS in CRAHNs. In this thesis, a new multi-hop cluster based CSS scheme has been proposed for CRAHNs, which can enhance the cooperative sensing performance and reduce the energy consumption compared with other conventional decentralised cooperative spectrum sensing modes.

Cognitive radio networks (CRNs) have emerged as a solution for the looming spectrum crunch caused by the rapid adoption of wireless devices over the previous decade. This technology enables efficient spectrum utility by dynamically reusing existing spectral bands. A CRN achieves this by requiring its users – called secondary users (SUs) – to measure and opportunistically utilise the band of a legacy broadcaster – called a primary user (PU) – in a process called spectrum sensing. Sensing requires the distribution and fusion of measurements from all SUs, which is facilitated by a variety of architectures and topologies. CRNs possessing a central computation node are called centralised networks, while CRNs composed of multiple computation nodes are called decentralised networks. While simpler to implement, centralised networks are reliant on the central node – the entire network fails if this node is compromised. In contrast, decentralised networks require more sophisticated protocols to implement, while offering greater robustness to node failure. Relay-based networks, a subset of decentralised networks, distribute the computation over a number of specialised relay nodes – little research exists on spectrum sensing using these networks. CRNs are vulnerable to unique physical layer attacks targeted at their spectrum sensing functionality. One such attack is the Byzantine attack; these attacks occur when malicious SUs (MUs) alter their sensing reports to achieve some goal (e.g. exploitation of the CRN’s resources, reduction of the CRN’s sensing performance, etc.). Mitigation strategies for Byzantine attacks vary based on the CRN’s network architecture, requiring defence algorithms to be explored for all architectures. Because of the sparse literature regarding relay-based networks, a novel algorithm – suitable for relay-based networks – is proposed in this work. The proposed algorithm performs joint MU detection and secure sensing by large-scale probabilistic inference of a statistical model. The proposed algorithm’s development is separated into the following two parts. • The first part involves the construction of a probabilistic graphical model representing the likelihood of all possible outcomes in the sensing process of a relay-based network. This is done by discovering the conditional dependencies present between the variables of the model. Various candidate graphical models are explored, and the mathematical description of the chosen graphical model is determined. • The second part involves the extraction of information from the graphical model to provide utility for sensing. Marginal inference is used to enable this information extraction. Belief propagation is used to infer the developed graphical model efficiently. Sensing is performed by exchanging the intermediate belief propagation computations between the relays of the CRN. Through a performance evaluation, the proposed algorithm was found to be resistant to probabilistic MU attacks of all frequencies and proportions. The sensing performance was highly sensitive to the placement of the relays and honest SUs, with the performance improving when the number of relays was increased. The transient behaviour of the proposed algorithm was evaluated in terms of its dynamics and computational complexity, with the algorithm’s results deemed satisfactory in this regard. Finally, an analysis of the effectiveness of the graphical model’s components was conducted, with a few model components accounting for most of the performance, implying that further simplifications to the proposed algorithm are possible. / Dissertation (MEng)--University of Pretoria, 2020. / Electrical, Electronic and Computer Engineering / MEng / Unrestricted

Cognitive radios are currently presented as the solution to the ever-increasing spectrum shortage problem. However, their increased capabilities over traditional radios introduce a new dimension of security threats. Cooperative Spectrum Sensing (CSS) has been proposed as a means to protect cognitive radio networks from the well known security threats: Primary User Emulation (PUE) and Spectrum Sensing Data Falsification (SSDF). I demonstrate a new threat to trust-based CSS protocols, called the Rogue Signal Framing (RSF) intrusion. Rogue signals can be exploited to create the illusion of malicious sensors which leads to the framing of innocent sensors and consequently, their removal from the shared spectrum sensing. Ultimately, with fewer sensors working together, the spectrum sensing is less robust for making correct spectrum access decisions. The simulation experiments illustrate the impact of RSF intrusions which, in severe cases, shows roughly 40\% of sensors removed. To mitigate the RSF intrusion's damage to the network's trust, I introduce a new defense based on community detection from analyzing the network's Received Signal Strength (RSS) diversity. Tests show a 95\% damage reduction in terms of removed sensors from the shared spectrum sensing, thus retaining the benefits of CSS protocols.

The idea that the radio spectrum is growing ever more scarce has become commonplace, and is being reinforced by the recent bidding wars among telecom operators. New wireless applications tend to be deployed in the relatively narrow unlicensed frequency bands, worsening the problem of interference for all users.  However, not all frequency bands are in use in every location all the time, creating temporal and spatial gaps (also known as white spaces) that cognitive radio systems aim to take advantage of. In order to achieve that, such systems need to be able to constantly scan large chunks of the radio spectrum to keep track of which frequency bands are locally available any given moment, thus allowing users to switch to one of these unoccupied frequency bands once the current band becomes unusable (or less useful). This requirement of wideband sensing capabilities often translates into the need to install specialized radio components, raising the costs of such systems, and is often at odds with the focus on monitoring the current band as is done by traditional wireless devices. The goal of this master’s thesis project is to simplify cognitive radio systems by shifting the wideband sensing functionality to a specialized and inexpensive embedded platforms that will act as a white space sensor, thus freeing cognitive radio users from this task and making it easier to integrate dynamic spectrum management techniques into existing systems. To do that a wireless sensor gateway platform developed by a previous master’s thesis has been repurposed as a prototype white space detector and tested against several wireless transmitters.  The aim is to develop a standalone platform that can be deployed all around an area to collect data that can be used to create a geographical map of the use of the spectrum. Such a system should require as little maintenance as possible, thus auto-update and self-configuring features have been implemented in the detector, as well as a simple scanning protocol that allows for remote configuration of the wideband sensing parameters. Furthermore, a basic server has been developed to aggregate and display the data provided by the different sensors. / Tanken att radiospektrum blir allt knappare har blivit vardagsmat, och förstärks av de senaste budgivning krig mellan teleoperatörer. Nya trådlösa applikationer tenderar att sättas i de relativt smala olicensierade frekvensband, förvärrade problemet med störningar för alla användare. Men inte alla frekvensband som används i varje plats hela tiden, skapar tidsmässiga och rumsliga luckor (även känd som vita fläckar) som kognitiva radiosystem syftar till att dra nytta av.  För att uppnå detta, sådana system måste hela tiden kunna scanna stora delar av radiospektrum för att hålla reda på vilka frekvensband är lokalt tillgängliga varje givet ögonblick, vilket gör omkopplaren när den nuvarande bandet blir obrukbar.  Det här kravet på bredbands avkänning kapaciteter översätter ofta in behovet av att installera specialiserade radiokomponenter, höja kostnaderna för sådana system, och är ofta i strid med fokus på övervakning av strömmen band med traditionella trådlösa enheter. Målet med detta examensarbete är att förenkla kognitiva radiosystem med wideband avkänning funktionalitet till en specialiserad och billig inbäddad plattform som kommer att fungera som ett vitt utrymme sensor, vilket frigör kognitiva radio användare från denna uppgift och gör det enklare att integrera dynamiskt spektrum förvaltning tekniker i befintliga system. För att göra det en trådlös sensor gateway plattform som utvecklats av ett tidigare examensarbete har apterat som en prototyp blanktecken detektor och testas mot flera trådlösa sändare. Målet är att utveckla en fristående plattform som kan sättas runt för att skapa en geografisk karta av användningen av spektrum och kräva så lite underhåll som möjligt, har automatisk uppdatering och självkonfigurerande funktioner implementerats i detektorn, samt som en enkel scanning protokoll som möjliggör fjärrkonfiguration av den bredbandiga avkänningsparametrarna. Dessutom har en grundläggande server utvecklats för att aggregera och visa uppgifterna från de olika sensorerna.

Dynamic spectrum access (DSA) wireless networks focus on using RF spectrum more efficiently and dynamically. Significant progress has been made during the past few years. For example, many measurements of current spectrum utilization are available. Theoretical analyses and computational simulations of DSA networks also abound. In sharp contrast, few network systems, particularly those with a decentralized structure, have been built even at a small scale to investigate the performance, behavior, and dynamics of DSA networks under different scenarios. This dissertation provides the theory, design, and implementation of a software radio-based decentralized DSA network prototype, and its enabling technologies: software radio, signal detection and classification, and distributed cooperative spectrum sensing. By moving physical layer functions into the software domain, software radio offers an unprecedented level of flexibility in radio development and operation, which can facilitate research and development of cognitive radio (CR) and DSA networks. However, state-of-the-art software radio systems still have serious performance limitations. Therefore, a performance study of software radio is needed before applying it in any development. This dissertation investigates three practical issues governing software radio performance that are critical in DSA network development: RF front end nonlinearity, dynamic computing resource allocation, and execution latency. It provides detailed explanations and quantitative results on SDR performance. Signal detection is the most popular method used in DSA networks to guarantee non-interference to primary users. Quickly and accurately detecting signals under all possible conditions is challenging. The cyclostationary feature detection method is attractive for detecting primary users because of its ability to distinguish between modulated signals, interference, and noise at a low signal-to-noise ratio (SNR). However, a key issue of cyclostationary signal analysis is the high computational cost. To tackle this challenge, parallel computing is applied to develop a cyclostationary feature based signal detection method. This dissertation presents the method's performance on multiple signal types in noisy and multi-path fading environments. Distributed cooperative spectrum sensing is widely endorsed to monitor the radio environment so as to guarantee non-interference to incumbent users even at a low SNR and under hostile conditions like shadowing, fading, interference, and multi-path. However, such networks impose strict performance requirements on data latency and reliability. Delayed or faulty data may cause secondary users to interfere with incumbent users because secondary users could not be informed quickly or reliably. To support such network performance, this dissertation presents a set of data process and management schemes in both sensors and data fusion nodes. Further, a distributed cooperative sensor network is built from multiple sensors; together, the network compiles a coherent semantic radio environment map for DSA networks to exploit available frequencies opportunistically. Finally, this dissertation presents the complete design of a decentralized and asynchronous DSA network across the PHY layer, MAC layer, network layer, and application layer. A ten-node prototype is built based on software radio technologies, signal detection and classification methods, distributed cooperative spectrum sensing systems, dynamic wireless protocols, and a multi-channel allocation algorithm. Systematic experiments are carried out to identify several performance determining factors for decentralized DSA networks. / Ph. D.