Spectrum Policy for Wireless Innovators Michael J. Marcus (Marcus Spectrum Solutions, USA)
Wireless technology is heavily regulated around the world and innovative ideas often encounter regulatory barriers on the national and international level. However, some wireless technologies in some countries exist in environments that allows "permissionless innovation". This tutorial will allow attendees to understand the basic regulatory issues confronting wireless innovations and how to identify and evaluate "permissionless innovation" opportunities. The presentation will deal with ITU regulation as well as national spectrum policy is a few representative countries. It will address realistic options for dealing with regulations that must be changed to allow an innovation and time delays and regulatory costs that might be encountered. Regulations discussed will be allocations, service rules, experimental license issues, and equipment authorization. In all cases the focus will be on innovative technologies on their way to the marketplace.
RF Network on Chip Jonathon Pendlum (Ettus Research, USA) (presentation not available for download)
When developing software defined radio (SDR) platforms, the process is very different when the target is a general purpose processor (GPP) versus a FPGA based design. On GPP platforms, there are many powerful SDR suites, such as GNU Radio (GR), that provide a modular stream based processing infrastructure that allow the developer to focus on algorithm development. In these GPP SDR suites, the user is generally not required to be an expert in the low level infrastructure. Conversely, most FPGA platforms are monolithic implementations that require the user to be an expert in both their algorithm and the intimate details of the FPGA design's infrastructure. Furthermore, the FPGA platforms lack the GPP based SDR suite's ability to easily add, rearrange, and reconfigure processing blocks. However, FPGAs have tremendous parallel processing capability and can accelerate many SDR related algorithms. An ideal platform would allow heterogeneous processing with the GPP and FPGA while retaining the ease of design and flexibility of GPP based SDR. RF Network on Chip (RFNoC) is our implementation of such a platform. RFNoC is a new processing framework for Ettus Research third generation USRP devices that aims to make FPGA acceleration in SDR more accessible. RFNoC implements a packetized network infrastructure in the USRP's FPGA that handles the transport of control and sample data between the GPP and radio. Users implement their custom algorithms in FPGA based processing blocks, or Computation Engines (CEs), that attach to this network. CEs act as independent nodes on the network that can receive from and transmit data to any other node (such as another CE, radio frontend, or GPP.) This framework permits scalable designs that can distribute processing across many nodes. Users can create modular, FPGA accelerated SDR applications by chaining CEs into a flow graph in a fashion similar to many GPP SDR suites. One such suite, GNU Radio (GR), fully supports RFNoC. Users can create flow graphs containing both GR blocks and RFNoC CEs that seamlessly communicate. CE parameters, such as FFT size and FIR filter coefficients, can be set from within GNU Radio like any other GR block. By simplifying access to the FPGA, the same hardware can be more easily used for more demanding applications. We will present an in depth tutorial on RFNoC including a discussion on its design and capabilities, demos of several existing examples, and instructions on how to implement Computation Engines in RFNoC.
TS1: Signal Classification
Automatic Modulation Classification using a Waveform Signature William Clark, IV and Joseph M. Ernst (Virginia Tech, USA); Robert McGwier (Virginia Tech & Allied Communications, AMSAT, and Flex Radio System, Inc., USA)
Cognitive Radios (CRs) build upon Software Defined Radios (SDRs) to allow for autonomous reconfiguration of communication architectures. In recent years, CRs have been identified as an enabler for Dynamic Spectrum Access (DSA) applications in which secondary users opportunistically share licensed spectrum. A major challenge for DSA is accurately characterizing the spectral environment, which requires blind signal classification. Existing work in this area has focused on simplistic channel models; however, more challenging fading channels (e.g., frequency selective fading channels) cause existing methods to be computationally complex or insufficient. This paper develops a novel blind modulation classification algorithm, which uses a set of higher order statistics to overcome these challenges. The set of statistics forms a signature, which can either be used directly for classification or can be processed using big data analytical techniques, such as principle component analysis (PCA), to learn the environment. The algorithm is tested in simulation on both flat fading and selective fading channel models. Results of this blind classification algorithm are shown to improve upon those which use single value higher order statistical methods.
Detection and Classification of LPI Radar signals using Spectrum Correlation and Support Vector Machines Thomas Schucker and Garrett Vanhoy (University of Arizona, USA)
In modern radar systems Low probability of Intercept (LPI) modulation techniques make the radars emitted signal difficult to detect. LPI signals use wideband, frequency hopping and frequency modulated continuous waves (FMCW) to reduce the signal profile. The low signal profile of the LPI signal enables the radar to perform detection and or target tracking while the target remains unaware. Several modulation techniques such as polytime codes, polyphase codes, FSK, and FMCW are used to produce LPI signals for transmission. From the target side, the incoming LPI signal is wideband with unknown center frequency, and low SNR. This paper looks at the ability of spectrum correlation along with a support vector machine (SVM) in order to automatically detect and classify the different LPI signal types in a non-cooperative environment.
TS2: SDR, CR and DSA Applications
Integrating Peer-to-Peer Functionalities and Routing in Mobile Ad-Hoc Networks Grant Millar (Cylo Ltd, United Kingdom); Alexandros Ladas, Olayinka Adigun and Christos Politis (Kingston University, United Kingdom)
Mobile Ad-hoc Networks (MANETs) impose strict requirements in terms of battery life, communication overhead and network latency, therefore optimization should be made to applications and services such as domain name service (DNS), dynamic host configuration protocol (DHCP) and session initiation protocol (SIP) if they are to be considered for use on MANETs. Due to the decentralized and self-organizing nature of MANETs, such applications could utilize a distributed name resolution/data storage service. Distributed Hash Tables (DHTs) enable these features by virtually organizing the network topology in a peer-to-peer (P2P) overlay. P2P overlays have been designed to operate on the application layer without knowledge of the underlying network thus causing poor performance. To address this problem, we propose and evaluate two different DHTs integrated with MANET routing in order to optimize the overall performance of MANET communications when P2P applications and services are used. Both architectures share the same functionality such as decentralization, self-reorganization, and self-healing but differ in MANET routing protocol. Performance evaluation using the NS2 simulator shows that these architectures are suited to different scenarios namely increasing network size and peer velocity. Comparisons with other well-known solutions have proven their efficiency with regard to the above requirements.
Power Efficient Vehicular Ad Hoc Networks Sterling Holcomb (Georgia Southern University, USA); Audrey Knowlton (New York Institute of Technology, USA); Juan Guerra (University of California Merced, USA); Hamed Asadi (University of Arizona, USA); Haris Volos (DENSO International America, USA); Jonathan Sprinkle and Tamal Bose (University of Arizona, USA)
Inter-vehicular communication is a growing platform for improving roadway safety. The highly mobile nature of Vehicle to Vehicle communications causes rapid changes in network topologies and propagation conditions. Since the advent of Vehicular Ad-Hoc Networks (VANETs), over fifty routing protocols with attendant topologies have been proposed. Despite these protocols' merits, many of them are not optimized for power management and frequency reuse. Our approach utilizes the one dimensional dynamic of roadways to simplify the routing problem and reduce energy consumption. Since each car is aware of only two types of connections, up road and down-road, we can form low power, line of sight links between adjacent vehicles. We also utilize a fuzzy logic algorithm that predicts the location of up-road cars to reduce interference from request for link signals. Once these links have been established, up-road vehicles send data down-road for a length of time based on the relative speed of the two vehicles. After this time period has expired the down-road vehicle must request additional information, restarting the timer. Data sent through the network will include information on up-road vehicles, and when required, messages such as accident notifications, alerts, and traffic warnings. Through simulation, we show an approach to VANETs that enables greater frequency reuse by reducing power consumption with a simplified network topology and a power aware link formation protocol.
Considerations for the Next Revision of STRS Sandra Johnson (NASA & Glenn Research Center, USA); Louis Handler and Janette Briones (NASA Glenn Research Center, USA)
NASA's Software Defined Radio architecture, the Space Telecommunication Radio System (STRS), was initiated in 2004. The goal of STRS is to enable waveform portability between NASA mission Software Defined Radios (SDRs) with limited effort. To meet NASA's needs, STRS is lighter and less complex than the Joint Tactical Radio System's Software Communication Architecture. STRS was implemented on three SDR platforms. These SDRs were installed on an external pallet on the International Space Station (ISS) in 2012. The SDRs have been reconfigured multiple times with new waveforms.
Blackspot: Using Tensor Decompositions to Guide Inspection of Source Code David Bruns-Smith, James Ezick, Janice McMahon and Jonathan Springer (Reservoir Labs, USA)
In this paper we introduce Blackspot, an extension to R-Check SCA that uses unsupervised machine learning based on tensor decompositions to organize and highlight sections of source code for more systematic inspection. Using markers identified by R-Check SCA's Pitchfork rule language, multi-dimensional decompositions are used to cluster code so as to group similar structures for accelerated manual inspection and, when seeded with examples of known weaknesses, to prioritize code fragments for rigorous review based on similarity derived from latent features. We show how multi-dimensional analysis provides a precision advantage over matrix SVD-based approaches and enables both accelerated compliance testing and more directed discovery of potentially critical software weaknesses. Utilizing high-performance tensor decomposition techniques provided by Reservoir's ENSIGN Tensor Toolbox, Blackspot scales to millions of lines of code, making it practical for application to complex, large-scale cyber-physical systems. Using an open SCA radio waveform as a first example, we illustrate how Blackspot can be applied to guide inspection for SCA compliance testing and weakness discovery in the software radio domain.
TS3: Spectrum Sharing 1
Reliable Spectrum Sharing Management for Cognitive Radio Networks Ahmed Fakhrudeen (University of Salford & College of Science and Technology, United Kingdom); Omar Younis Alani (University of Salford, United Kingdom)
Cognitive Radio Network (CRN) is a promising network that aims to improve the utilization of the wireless spectrum by enabling unlicensed (secondary) users to reuse the underutilized bands. CRN utilization of residual spectrum bands of Primary (licensed) Networks (PNs) must avoid harmful interference to the users of PNs and other overlapping CRNs. Numerous Internetwork spectrum sharing frameworks have been proposed in the literature; however, spectrum sharing among overlapping CRNs presents significant challenges. This paper comprises two major contributions; firstly, it proposes a Novel CRNs management framework, CogMnet, which regulates the operation of centralized CRNs. CogMnet aims to ensure the reliability of CRNs' spectrum sharing by tackling the Primary User Emulation Attack (PUEA) issue and avoiding an overcrowded CRNs scenario. Secondly, it proposes CRN Admission Control (CRNAC) capable of determining the maximum number of CRNs allowed in any location. To the best of our knowledge CogMnet is the first Internetwork framework able to distinguish an attacker CRN that may perform PUEA. Furthermore, CRNAC is the first network admission decision making algorithm in CRNs literature. Analytical results are presented to demonstrate the performance of the algorithm. Assigning the number of CRNs is very important to avoid saturated spectrum situation.
Analysis of dynamic capabilities in the citizens broadband radio service Seppo Yrjölä (Nokia Networks, Finland); Marja Matinmikko (University of Oulu, Centre for Wireless Communications, Finland); Miia Mustonen (VTT Technical Research Centre of Finland, Finland); Petri Ahokangas (University of Oulu, Finland)
This paper seeks to identify and analyze the sources of value creation and capture by key stakeholders in the new Citizens Broadband Radio Service (CBRS) shared spectrum access framework introduced by the FCC. More flexible and dynamic use of the 3.5GHz spectrum aims to increase the efficiency of spectrum use in delivering fast growing and converging mobile broadband and media services while paving way to new innovations in technology and business models. In this paper we focus on key stakeholders' capability to deal with combined internal and external resources and capabilities in doing business, referred as Dynamic Capability. Spectrum sharing introducing a rapid change in the technology and business environments requires dynamic capabilities from spectrum offering, spectrum utilization and spectrum management perspectives. We focus on defining key CBRS functional domains and identifying their key antecedents, processes, and outcomes. The DC analysis highlights the key role of the regulator in creating a sharing framework with incentives for all the key stakeholders having different operational and business requirements and enabling scaling ecosystem. Increased system dynamics in spectrum sharing will introduce needs for big data analytics, near real time network management capabilities and low cost 3rd tier general authorized access radios leveraging dominant technology ecosystems. This study provides viewpoints for stakeholders about additional ingredients and actions which may be relevant to further promote spectrum sharing in the form of the CBRS. The concept of dynamic capabilities was found useful to analyze the sources of competitive advantage regarding CBRS spectrum sharing.
Field trial of Licensed Shared Access (LSA) with enhanced LSA controller power control concept algorithms Seppo Yrjölä (Nokia Networks, Finland); Vesa Hartikainen and Lucia Tudose (Nokia Solutions and Networks, Finland); Jaakko Ojaniemi (Aalto University, Finland); Arto Kivinen (Turku University of Applied Sciences, Finland); Tero Kippola (Centria University of Applied Sciences, Finland)
This paper presents the results from an end-to-end ecosystem trial of the Licensed Shared Access (LSA) concept using over the air TD-LTE network in the 3GPP spectrum band 40 (2.3-2.4 GHz) in Finland. In the field trial, the LTE network shared the spectrum with Program Making and Special Events incumbent wireless cameras. New LSA concept elements, LSA Repository for incumbent protection information and LSA Controller for controlling the mobile broadband network in the same spectrum band were implemented in the trial environment. The trial utilized commercially available network elements like multimode multiband terminals, LTE base stations, core network and network management system. Incumbent spectrum usage data was collected to the LSA repository, which further converts it to spectrum availability information for the LSA controller. The trial goes beyond previous LSA demonstrations by presenting enhanced LSA controller power control concept algorithms that instead of maximizing the number of transmitting base stations (BS) as in the Protection Zone Optimization (PZO), formulate the optimization objective as a function of BS transmit power. An advantage of this procedure in contrast to the PZO is that adjusting the transmit power level do not result in abrupt changes in the received signal quality; therefore it is possible to reach better overall throughput. Furthermore enhanced incumbent protection is provided with algorithms considering aggregate interference from the LTE network to incumbent. The developed LSA controller was implemented as Self Organizing Network solution fully integrated into commercial Operational Support System. Incumbent users' rights were protected by evacuating and or reconfiguring the overlay LSA TD-LTE band and handing users over to coverage FDD LTE network if required when requested by the incumbent spectrum user. Numerical results are presented to quantify the duration of the LSA work flow steps. The trial showed that the LSA concept can be implemented with commercial available network elements and a minimum amount of new software and hardware components. The performance results on the LSA system workflow indicated that in the PMSE use case the usage of the LSA band can be managed timely manner and the incumbents' rights can be protected. Furthermore developed approach could be expanded to be utilized in the LSA evolution towards more dynamic US 3 tier Citizens Broadband Radio Service (CBRS) system.
TS4: Advanced Radio Architectures
Experimental Validation of an Iterative Receiver for Energy-Efficient Communications Qasim Chaudhari (Centre for Energy-Efficient Telecommunications, Australia); Brian Krongold (University of Melbourne, Australia)
Rising demand for wireless connectivity anywhere anytime has led the scientific community to focus on the energy efficiency of wireless networks. It has been realized that powerful error correcting codes like turbo and LDPC are making it possible to establish wireless communications at low SNRs. However, synchronization and channel estimation errors, which are inevitable at such SNRs, erode most of the achieved gains. Further expanding the turbo concept through an iterative receiver -- which brings synchronization and equalization modules inside the loop -- can help, but this solution is prohibitively complex and it is not clear what can and what cannot be a part of the iterative structure. This paper fills two important gaps in this field: (1) as compared to previous research which either focuses on a subset of the problem assuming perfect remaining parameters or is computationally too complex, we propose a proper partitioning of algorithm blocks in the iterative receiver for manageable delay and complexity, and (2) to the best of our knowledge, this is the first physical demonstration of an iterative receiver based on experimental radio hardware. We have found that for such a receiver to work, (1) iterative timing synchronization is impractical, iterative carrier synchronization can be avoided by using our proposed approach, while iterative channel estimation is essential, and (2) the SNR gains claimed in previous publications are validated in indoor channels.
Design Considerations for Channelizer Based Receivers in Generalized Frequency Division Multiplexed OFDM Systems frederic j harris (San Diego State Univ, USA); Chris Dick (Xilinx, USA); Elettra Venosa (Space Micro, USA); Xiaofei Chen (San Diego State University, USA)
Polyphase filter banks permit multiple access OFDM signaling waveforms without need for time alignment of received signals nor frequency offset alignment to maintain their mutual orthogonality. The modulation bandwidth, channel spacing, and channels sample rates are coupled and their specifications interact. We will show how to design the modulation and demodulation filter banks to minimize the interaction of the filter bank composite impulse response with the transmitted and received signal components. A pleasant surprise at the end of the design process is that the minimization results in zero interaction of the channelizers with the channel except for a short causal delay.
Quantifying the Impact of RF Front End Nonlinearity on Receiver Performance Aditya V Padaki (Virginia Tech, USA); Ravi Tandon (University of Arizona, USA); Jeffrey Reed (Virginia Tech, USA)
RF front end nonlinearity can potentially impair receiver performance by harmful Adjacent Channel Interference (ACI). The vulnerability to ACI significantly increases in shared spectrum environments, which allow a free-style of spectrum access with diverse receiver technologies accessing the same band of spectrum. ACI is a function of both the RF environment in which the receiver operates and the front end nonlinearity. In this presentation we first describe tractable representations to capture RF nonlinear distortions. Second, we outline a framework to quantify the implications of RF front end on receiver performance using information theoretic metrics. Third, we comment on techniques to model the RF environment. Finally, we present certain numerical results to depict the impact of RF nonlinearity for practical scenarios and configurations of wireless networks.
A 50MHz+ bandwidth real-time FPGA implementation of OFDM-based PHY transceiver for 5G Carlos M. Ribeiro (Instituto de Telecomunicações / Instituto Politécnico de Leiria, Portugal); Atílio Gameiro (Instituto de Telecomunicações / Universidade de Aveiro, Portugal)
This paper introduces a fully pipelined implementation architecture for OFDM-based transceivers that moves one step closer to fulfilling the envisioned 5G bitrate demands. The implementation uses high level abstraction tools to develop and test the algorithms, significantly increasing the productivity and reducing costs and time-to-market. The proposed architecture defines a common interface between blocks, with FIFOs interconnecting the blocks and a soft-handshake, which present significant advantages over the other reported architectures. The proposed architecture is implemented in a fully working real-time platform, using COTS development boards. The demonstrator has a real-time scalable bandwidth from 20MHz to 54MHz. Using 64-QAM constellations and 1024 sub-carriers, the demonstrator attains over 300Mbits/s of raw bitrate in each direction.
TS5: Spectrum Sharing 2
An LTE-Based Wideband Distributed Spectrum Sharing Architecture Mingming Cai and J. Nicholas Laneman (University of Notre Dame, USA)
This paper describes a radio architecture for distributed spectrum sharing among secondary users (SUs) in a localized area and a wide band of frequencies. Based upon an orthogonal frequency-division multiplexing (OFDM) physical layer, the architecture allows multiple pairs of SUs to utilize one or more sub-channels within the band without causing harmful interference to each other. The spectrum utilized by a SU pair may be contiguous or discontiguous; it can be changed dynamically based upon spectrum sensing at the transmitter, and the receiver tracks using synchronization and control messages from the transmitter. A prototype implementation of the architecture has been developed using National Instruments hardware and software, specifically USRP RIOs as well as the LabVIEW Communications System Design Suite (CSDS) and associated LTE Application Framework, respectively. Preliminary results from experimental validation of the prototype implementation will also be reported.
Co-existence of MIMO Radar and Communications Systems: Real-Time Interference Control Mohammed Hirzallah and Tamal Bose (University of Arizona, USA)
Controlling interference in a time-varying wireless channels is essential in the spectrum sharing and co-existence solutions. We develop a robust tool that controls the interference generated from a Multiple Input Multiple Output (MIMO) radar and received by a MIMO base station. The tool is able to confine the interference up to any tolerated value, and it can be used by any MIMO based platform. The presented tool manage the interference at the MIMO radar transmitter through a pre-coder design. It configures the radar's pre-coder in an iterative manner based on a Steepest-Descent (SD) and a Normalized SD (NSD) approaches. The tool has a low computational complexity. As expected, our results show that the NSD has a shorter convergence time and a better convergence stability when compared with the SD approach. Our results reveal that the initial settings influence the convergence speed and the random assignment of initial settings results in a longer convergence time when compared with the other option.
ESC Sensor Nodes Placement and Location for Moving Incumbent Protection in CBRS Satya Krishna Joshi (University of Oulu, Finland); Kapuruhamy Badalge Shashika Manosha (Centre for Wireless Communications, Department of Communications Engineering, University of Oulu, Finland); Markku Jokinen (University of Oulu & Centre for Wireless Communications, Finland); Tuomo Hänninen and Pekka Pirinen (University of Oulu, Finland); Harri Posti (Centre for Wireless Communications, Finland); Matti Latva-aho (UoOulu, Finland)
We consider environmental sensing capability sensor node placement and location problem in coastal areas for the moving incumbent protection in citizens broadband radio service (CBRS). We have considered two criteria for sensor nodes placement to protect the moving incumbents from the harmful interference that might occur due to the citizens broadband radio service devices. First, we consider the protection of the incumbents by deploying a minimum number of sensors in the coastal area. Then, we consider the robustness against the sensor failures.
Quality of Service Assurance-based Auction for Spectrum Sharing Systems Munawwar Sohul (Virginia Tech, USA); Miao Yao (1710 Emerald Street & Virginia Tech, USA); Vuk Marojevic and Jeffrey Reed (Virginia Tech, USA)
Wireless communities throughout the world have recognized the shortage of spectrum for commercial broadband uses and acknowledged the urgent need for a global effort to make additional spectrum available for broadband data. Spectrum sharing has captured the center stage as the solution to this issue of spectrum scarcity [1]. The Federal Communications Commission (FCC) proposed spectrum sharing framework introduces a Spectrum Access System (SAS) as the governing entity that manages desired coexistence among the primary and secondary users (PU and SUs), as well as the availability of a wide range of radio spectrum. An important aspect of this dynamic spectrum management is the pricing of spectrum from the perspective of both the PU and SU [2-4]. Existing auction based spectrum sharing models failed to take into consideration an important aspect of successful SU operation: the duration of the available spectrum opportunity. In order to achieve any desired level of Quality of Service (QoS), the service providers need information about the QoS predictability of a shared spectrum which is highly influenced by the duration of the available spectrum opportunity. Based on the QoS Assurance (QoSA) approach presented in our previous work [5], an auction based spectrum sharing framework that accounts for the quality of the available spectrum opportunity is proposed in this paper. The proposed auction process allows both the PU and SU to iteratively adjust their evaluation about the available spectrum opportunities and over time achieve a price combination that maximizes both PU and SU objectives.
TS6: SDR, CR and DSA Applications 2
Analysis of Spread-Spectrum Algorithms to Prevent Jamming Attacks in Safety-Critical Applications Aysegul Aglargoz and Valentin Bartkowiak (German Aerospace Center, Germany)
Rapid development in the field of wireless technologies makes the wireless systems feasible for high-data-rate and power-efficient applications. As one of the applications, wireless sensor and actuator networks have an expanding application field. In field of aeronautics, the substitute of cable-based sensory and actuator network by wireless technologies in aircraft is a novel approach, which aims to reduce the cable weight and enhance system functionalities, ease of layout and maintenance [1]. However, the replacement of wired sensor networks with wireless systems raises several problems in terms of safety and reliability. During wireless data-transmission of sensory and actuator information in aircraft, the transmission channel is shared with several applications and encounter various types of noise and interferences. This shared wireless transmission medium requires regulations and robust methods to fulfill the safety and reliability requirements. One of the significant threats in such a propagation medium is jamming attacks. Therefore, jamming-resistant wireless communication methods are examined in this work, specifically for safety-critical applications in aircraft. The proposed solution is spread-spectrum methods, which expand the signal in frequency domain to obtain a wider bandwidth. These methods are commonly used in wireless military and commercial applications, such as IEEE 802.15.4 standards for wireless personal area networks and IEEE 802.11b for Wi-Fi, to counter against jamming attacks and overcome interferences in a transmission channel [2]. In this paper, we examine the anti-jamming performance of a type of spread-spectrum technique: Frequency-Hopping Spread-Spectrum (FHSS). To analyze the performance of FHSS against jamming attacks in terms of safety and reliability, we develop a transceiver on a real-time platform, which is based on a Xilinx Virtex-6 FPGA. We use Xilinx System Generator libraries to develop a Software-Defined Radio (SDR) in Simulink environment. This provides high-level and prompt DSP development capabilities. Furthermore, we model several jammer characteristics: tone, multitone, narrowband, partial-band and broadband jammers. Multitone and partial-band jammers are classified as worst-case jammers [3]. By using these jammer models, we further simulate the data-link under worst-case jamming attack and test the performance of FHSS regardless of baseband modulation and error- correction algorithms. [1] ITU-R, 2010, Technical characteristics and operational objectives for wireless avionics intra-communications (WAIC), Report ITU-R M.2197, International Telecommunication Union. [2] Magill, D.T., Natali, F.D., Edwards, G.P., 1994, Spread-spectrum technology for commercial applications, Proceedings of the IEEE, vol.82, no.4, pp.572-584. [3] Simon, M., Omura, J.K., Scholtz R.A., Levitt, B.K., 1994, Spread Spectrum Communications Handbook, Revised Edition, Mc-Graw Hill.
An Open and Free ISDB-T full_seg Receiver Implemented in GNU Radio Federico Larroca and Pablo Flores Guridi (Universidad de la República, Uruguay); Gabriel Gómez Sena (Universidad de la República & Facultad de Ingeniería, Uruguay); Victor Gonzalez Barbone (Universidad de la República, Uruguay); Pablo Belzarena (Universidad de la Republica, Uruguay)
Almost every country in Latin America has adopted the ISDB-T standard for free-to-air television broadcasting. The so-called "analogical blackout" is about to be performed, so broadcast engineers and technicians have to be prepared for such a challenging task. Key to the success of this blackout is a deep understanding of the chosen technology. However, no true expertise can be achieved without dealing with actual implementations. Sadly, most of South American countries do not have either analogical nor digital television receiver or transmitter industries, so these devices have to be bought to technology suppliers outside the region. Software defined radio seems a good alternative for, at least, making cheap and flexible prototypes and getting to know how things really work. In this paper we present the first open, free and fully software-based ISDB-T full_seg receiver (https://github.com/git-artes/gr-isdbt), entirely implemented in GNU Radio, a free and open-source software development toolkit for implementing software defined radios. The block-based architecture of this framework allows broadcasting professionals and researchers to get in touch with a real-time working receiver. Moreover, it also offers the possibility to replace particular blocks or simulate channel conditions, in order to test different algorithms and implementations. It is important to highlight that the receiver outputs the corresponding transport stream, which can be fed into any typical video player (e.g. MPlayer or ffmpeg). The implementation was tested on a regular personal computer with an Intel i5 processor and 4 GB RAM memory, which was very capable of receiving, decoding and playing the transport stream online. This possibility further allows our receiver to be used to test the channel effects on the actual decoded audio/video. Regarding the receiver itself, OFDM synchronization was the most challenging task we had to solve (i.e. symbol timing, carrier frequency and sampling clock offsets compensation). To that end, we implemented two different synchronization methods: a one-shot (or feed-forward) which corrects only frequency and symbol timing, and a more sophisticated feedback method which also corrects the sampling clock offset (and further corrects the carrier frequency offset). As a toy example of the possibilities brought by our framework, we compare both methods in terms of the resulting BER in a typical scenario. The article also highlights notable differences between ISDB-T and the European DVB-T norm. For instance, the ISDB-T standard sends a single continual pilot while working with coherent modulation schemes. This fact impacts the method that must be used to perform post-FFT integer frequency offset estimation, which generally relies on the presence of several such pilots. Instead, we used the ISDB-T's TMCC carriers, which despite transmitting information we still do not know at this point of the receiving chain, we perfectly know in which carrier positions they should be, and also that they must all carry the same redundant information. These carriers, which deliver very useful information for the hierarchical division process, the Viterbi decoder, the symbol demapper and the time interleaving which were implemented later on in the receiving chain, were also used to for ISDB-T frame synchronization as they start with a pre-defined sync word that starts with each frame.
Combining Artificial Noise Beam Forming and Concatenated Coding schemes to effectively secure wireless communications Christiane L. Kameni Ngassa (Thales Communications & Security, France); Jean-Claude Belfiore (Telecom Paristech, France); Renaud Moliere and François Delaveau (Thales Communications & Security, France); Nir Shapira (Celeno Communications Ltd, France)
Existing security mechanisms for wireless communication rely on pre-shared cryptographic keys to encrypt exchanged data. However, recent news revealed that attackers can have access to these encryption keys by exploiting weakness of SS7 protocol and international roaming. Furthermore, the late hacking of SIM card manufacturers to get encryption keys proves that the cryptographic key distribution approach can no longer be considered as completely secure for public networks. Physical layer security (Physec) appears therefore as a crucial help to strengthen wireless communication security as it leverages inherent properties of the wireless channel to provide secrecy by remaining key-free. Secrecy (or wiretap) Coding is one of the main Physec techniques. Its goal is to provide both reliability and secrecy without using any secret key. Note that secrecy coding is independent of eavesdropper's computational capability, and thus remaining secure from attacks with unlimited computational power. Nevertheless secrecy coding requires a radio advantage for the legitimate nodes and terminals. In addition, the design of a practical wiretap code is very challenging: despite numerous available theoretical results, secrecy coding schemes proposed in the literature cannot be readily implemented in existing wireless communication systems. In this paper we propose a practical implantation of secrecy coding schemes helped by Artificial Noise (AN) and Beam Forming (BF) - The Artificial Noise scheme provides the required radio advantage to legitimate users, while the Beam Forming allows reliable link to the legitimate terminal - A concatenation of an "outer" polar code and an "inner" LDPC code is applied to the legitimate messages to build a wiretap coding scheme. The outer polar code provides secrecy while the LDPC inner code provides reliability. We study this scheme in a typical scenario where two legitimate users (Alice and Bob) attempt to securely communicate in presence of an eavesdropper (Eve). Artificial Noise (AN) and Beamforming (BF): Such schemes combine Beamforming of data towards the legitimate receiver and emission of interfering signals elsewhere. The power of the artificial noise is controlled and steered to limit the link budget of eavesdroppers while optimizing the MIMO transmission scheme for legitimate links. AN schemes proceed as follows: - Estimation of the legitimate Channel Frequency Response (CFR) or Channel Impulse Response (CIR), from Alice to Bob, and extraction of orthogonal directions of the legitimate CFR or CIR. - Transmission of noise streams on orthogonal directions. Eve is thus forced into low Signal to Interference Noise Ratio (SINR) regime and is unable to decode. - Beamforming of the Alice-Bob data stream for Bob to maximize legitimate link budget. Bob extracts Alice's channel and suppresses orthogonal noisy channel directions thanks to Beamforming. In ideal cases, the Interference at Bob's side completely vanishes and the SINR at Bob's side reduces to a Signal to Noise Ratio (SINRBob=SNRBob). When Artificial Noise and Beamforming techniques are established, a better SINR is provided to Bob than to Eve in any case. This radio advantage is then exploited by the legitimate link to compute secrecy codes. Secrecy coding scheme: As mentioned before, the design of secrecy codes for continuous channels is challenging. However since Polar codes provide strong security for discrete channels, our idea is to concatenate them to an inner code which is a capacity approaching code (in practice any channel code already present in the RATs - Turbo, LDPC, RS, etc.). In this way, the channel between the outer polar encoder and decoder can be viewed as a Binary Symmetric Channel. For our simulations, we use an LDPC code defined in the 802.11 standard as inner code and a polar code with compatible length as outer code. Simulations apply to several channel models: - main Order of magnitude and performance results are provided in theoretical Gaussian model, - then explorations of the SC performance is achieved by taking into account real field channel estimates recorded in several indoor and outdoor location within real Wifi and LTE networks. The main results of this study are the following. First our simulation results show, as expected, that the bit error rate at the output of the polar decoder is 0.5 up to a given "attacker threshold" of the Signal to Interference + noise ratio (SINREve), depending on the modulation and of concatenated coding scheme, that ensures no information leakage. When the SINR grows, as expected the bit error rate at the output of the polar decoder vanishes. When SINR is high enough (meaning greater than an "user threshold" SINRuser) the bit error rate at the output of the polar decoder approaches zero. Thus the artificial noise intensity I should be tuned in order to achieve suitable SINR at Bob's side thanks to the BF capabilities and un-practicable SINR at Eve's side who cannot apply BF. Finally, it appears that two basic radio parameters mainly drive the efficiency of the secrecy scheme - a "minimum SNRuser" for the legitimated link, which is relevant to the performance of the Beam-Forming modulation and coding schemes at Bob's receiver. This depends on the Channel estimation and on the energy budget of the legitimate link, all these parameter being part of the equalization processing and of the Quality of Service Management. - a "SINR Security gap" that represents the lower bound of the radio advantage to be provided at legitimate link by improving the interference at Eve's side (using artificial noise as interference power I), on the efficiency of the BF efficiency being controlled by Alice and Bob in the established AN scheme, the SINR Security gap drives the tuning of the Artificial Noise power by the Node to ensure the radio advantage at any Eve's location. To the best of our knowledge, this is the first work on a full practical secrecy coding scheme and complete simulation of the scheme with theoretical and real field channel models. Our promising results are evidence that the proposed secrecy coding scheme should be practically implemented in existing wireless MIMO or MISO communication systems and in emerging radiocells and WLAN Standards with only minor modifications of the software architecture of the nodes and terminals.
Exploring Cognition using Software Defined Radios for NASA Missions Sandra Johnson (NASA & Glenn Research Center, USA); Dale Mortensen, P. E. (NASA Glenn Research Center, USA); Richard Reinhart (National Aeronautics and Space Administration & Glenn Research Center, USA)
NASA missions typically operate using a communication infrastructure that requires significant schedule planning with limited flexibility when the needs of the mission change. Parameters such as modulation, coding scheme, frequency, and data rate are fixed for the life of the mission. This is due to antiquated hardware and software for both the space and ground assets and a very complex set of mission profiles. Automated techniques in place by commercial telecommunication companies are being explored by NASA to determine their usability by NASA to reduce cost and increase science return. Adding cognition - the ability to learn from past decisions and adjust behavior - is also being investigated.
TS7: Spectrum Sharing 3
Spectrum Consumption Model Builder: A software tool to enhance spectrum sharing Carlos E. Caicedo Bastidas and Arnav Mohan (Syracuse University, USA)
The growing demand for Radio Frequency (RF) spectrum resources for commercial and government use has motivated changes to spectrum management regulatory frameworks. Many of these changes focus on moving away from rigid spectrum management policies and embracing dynamic spectrum access and spectrum sharing mechanisms. Spectrum sharing refers to "the use of automated techniques to facilitate the coexistence of disparate unaffiliated spectrum dependent systems that would conventionally require separate bands to avoid interference" (PCAST Report, 2012). Thus, radio interference control and management will be a key requirement for the successful operation of future RF spectrum sharing environments. To satisfy this requirement, there must be an effective means of communicating the characteristics and limits of spectrum use of an RF transmitter, receiver, system or collection of RF systems. Spectrum consumption models (SCMs) attempt to capture this information. The structure and characteristics of SCMs are being standardized by the 1900.5.2 working group of the IEEE Dynamic Spectrum Access Networks Standardization Committee (DySPAN-SC). As a data model for spectrum consumption modeling, the SCM aims to be minimal and effective at providing the means for spectrum managers to use their judgment and tools to capture relevant spectrum consumption details of their devices/systems and express them in a standardized way. The models attempt to capture spectral, spatial, and temporal characteristics, and boundaries of the consumption of spectrum by any specific transmitter or receiver device or RF system. The information contained in the models enables better RF spectrum management practices and allows for the identification of spectrum reuse opportunities. The IEEE 1900.5.2 standard aims to standardize the data model for SCMs and the mechanisms/algorithms to arbitrate compatibility among individual and/or combinations of transmitter/receiver devices. RF devices and/or systems are compatible if they can operate under the spectrum use boundaries detailed by their respective SCMs. The standard also aims to provide the means to generate machine-readable SCMs, which together with the standardized compatibility calculation mechanisms would provide the means to automate the identification of spectrum reuse opportunities and coordinate spectrum access. With these capabilities SCM and the systems that are built around them would provide sufficient information to enable dynamic and flexible spectrum management decisions that can increase spectrum reuse (i.e. spectrum sharing, spectrum trading and policy-based spectrum management interactions). Thus, SCMs are well in line with moving forward the goals of many regulators for modernizing spectrum management. This paper presents our results in developing an open source tool for the construction and analysis of SCMs - The Spectrum Consumption Model Builder. Among its main functions, the tool provides a means for the construction of standard compliant SCMs and for the analysis of compatibility between systems and/or devices in which an SCM describes the boundaries of spectrum use. The tool is intended to promote the use of SCMs and educate radio engineers, wireless service providers, researchers and regulators on their benefits. The tool also serves as a mechanism to test the completeness and viability of using SCMs in scenarios of interest to industry and regulators.
Improving Robustness of a Cognitive Radio Engine Hamed Asadi (University of Arizona, USA); Haris Volos (DENSO International America, USA); Michael Marefat and Tamal Bose (University of Arizona, USA)
Cognitive radio engines (CEs) are the brains behind the intelligent adaptations for cognitive radios (CRs). A CE is an intelligent agent which observes the radio environment and chooses the best communication settings that best meet the application's goal. In this process, providing reliable performance is one of the major tasks in designing CEs for wireless communication systems. The main purpose of this work is providing predictable performance and controlling the cost of intelligent algorithms based on the CE's experience and complexity analysis respectively. In our previous publication, we have proposed a meta-CE that is able to evaluate various CE algorithms' performance automatically. A meta-CE is generally considered to be comprised by a set of CE algorithms and meta-cognition module that provides the meta-abilities of the CE. We showed that Meta-CE is able to provide much higher performance than each individual CE algorithm due to its flexibility and ultimately predictability of the Cognitive Radio Systems (CRS). We also have proposed a several knowledge indicators to estimate the amount of knowledge that is obtained by various CE algorithms. Providing predictable performance at all times is of paramount importance in different CE techniques. In this work, we extend our meta-CE design to control the cost of computations and provide more reliable performance for providing the minimum requirement of the radio applications in different scenarios. To this end, we use robust training algorithm in two different levels alongside of the individual CE algorithms. The first level of RoTA operates with individual CEs to control the exploration and exploitation rate. The RoTA, in this level, enables the Meta-CE to guarantee some minimum output performance based on the learning stages. RoTA in this level, also uses confidence interval approximation for standard normal distribution to calculate the lower and upper bounds of CE's expected performance to analyze the reliability of decisions. The second level of RoTA operates in meta-level to control the amount of computation complexity of intelligent algorithms in all levels with respect to the obtained performance and operating scenarios (channel conditions, operating objectives, radio applications, etc.). To sum up, in this paper we continue our work in meta-CE design by presenting our work on RoTA and the development and definition of a learning stages that allows to control the performance and complexity of individual CEs and out Meta-CE.
Metrics-based Comparison of OWL and XML Based Approaches to Representing Cognitive Radio Capabilities Yanji Chen (Northeastern University, USA); Jakub Moskal (VIStology, Inc., USA); Mieczyslaw Kokar and Kaushik Chowdhury (Northeastern University, USA)
A wireless network of software-defined radios can be considered as a distributed computing system, since the radios that are on the network possess the capabilities of performing various sensing and computation tasks requested by the applications running on other nodes. In order to achieve such an objective, radios would need to inform the network of their capabilities, applications would issue requests for services, and the network would match the radio capabilities against the request. One of the research questions is what languages are good for describing the radio capabilities and the application requests? Since the solution to this problem needs to be flexible enough to address the scenarios in which radios with previously unknown capabilities and new applications can join the network dynamically, the language must be interpretable by the applications, radio devices and the network. One possibility would be to base the approach on the XML technology, i.e., use XML schema definition (XSD) as a base for expressing device capabilities and application requests and then use XML to describe instance XML data of device descriptions and application requests. Application requests are then sent to the network as XQuery expressions. The network then can answer the query by matching it with the radio capabilities using an XQuery processor. Another approach could use the Semantic Web based approach that includes Web Ontology Language (OWL) and the SPARQL query language. In this case an ontology for this domain would serve as the base for describing both the device capabilities and the application requests. Application requests would be expressed in SPARQL and the matching would be performed by a SPARQL processor. In this paper we will present our investigation into the comparison of the two approaches described above. We use various metrics, such as processing time, bandwidth usage, precision, recall and F-measures to evaluate the two approaches. In particular, we make use of the Spectrum Consumption Modeling Markup Language (SCMML) for the XML based approach, while for the OWL based approach we use the counterpart SCMML ontology and Cognitive Radio Ontology (CRO). The paper describes the experimentation framework that we developed for the purpose of this evaluation, including the query generator system that is able to assess the ground truth of each request. The paper also presents experiment results for both the XML and the OWL based approaches.
TS8: SDR, CR and DSA Algorithms
Timing Recovery Algorithm Options for Modems with Rectangle Shaping Filters frederic j harris (San Diego State Univ, USA); Richard Bell (SPAWAR, USA); Vamsi Krishna (SDSU, USA)
In many timing recovery implementations we form two filters, the matched filter to supply the output and time derivative matched filter to supply the output and then proceed to drive the product to zero. Here is the problem! The rectangle shaping filter doesn't fit this option. The rectangle shaped matched filter doesn't have a zero derivative at its peak, and the derivative of the shaping filter isn't very useful, it doesn't have a zero crossing at time 0. As it happens, there are many modulation formats that use rectangle shaping filters. Frequency Shift Key (FSK), Continuous Phase Modulation (CPM), Frequency Hopping (FH) are some simple examples. We revisit the timing recovery mechanism for this class of modulation.
Secret Key Generation scheme from WiFi and LTE reference signals Christiane L. Kameni Ngassa, Renaud Moliere and François Delaveau (Thales Communications & Security, France); Taghrid Mazloum (Telecom ParisTech, France); Alain Sibille (Telecom Paris Tech & ENSTA PARISTECH, France)
Recent news highlighting security failures of modern wireless communication systems have recalled the limits of the cryptographic key distribution approach and the urge to improve security of the information exchanged over the air interface. The emergence of Physical layer Security (Physec) has provided an alternative approach for designing robust secret keys by leveraging the intrinsic randomness of wireless channels. This technique is referred to as Secret Key Generation (SKG). The vast majority of existing work on SKG uses the Received Signal Strength Indication (RSSI) to generate secret keys since it is easily accessible. However, RSSI does not capture the entire richness of the channel as it ignores the phase of channel coefficients, which usually provide more randomness than the power of the signal. In this paper we present a full SKG scheme based on full Channel State Information (CSI). We consider the typical scenario where two legitimate users (Alice and Bob) attempt to securely communicate in presence of an eavesdropper (Eve). Our SKG protocol is composed of the following steps: Channel Coefficient decorrelation, Quantization of the CSI, Information Reconciliation and Privacy Amplification. Channel Coefficient decorrelation: in this step, we apply a new algorithm to select channel coefficients with low cross correlation. This allows us to extract secret keys with enough randomness even from stationary environments. Quantization: we use the Channel Quantization Alternate (CQA) algorithm introduced by Wallace to quantize selected channel coefficients. The main advantage of this algorithm is that it minimizes key mismatch between the legitimate users Alice and Bob. Information Reconciliation: the goal of this step is to eliminate the remaining mismatch between Alice and Bob keys. We employ secure sketch and error correcting codes to correct Bob's errors on Alice's key. To do so, Alice has to send the secure sketch over the public channel, possibly leaking a controlled amount of information to the eavesdropper Eve. Privacy Amplification: here, we use hash functions to remove the information leaked to Eve and to improve the randomness of the secret key by decreasing its length. This final step guarantees that the generated secret key is independent from the key computed by the eavesdropper. Note that we choose a simple algebraic forward error correcting (FEC) code to reconcile Alice and Bob keys and a classical family of 2-universal hash function in the privacy amplification step. In order to evaluate the performance of our scheme we apply our secret key generation protocol to real field WiFi and LTE networks, with signals captured in several indoor and outdoor locations. The keys are computed from Channel Frequency Responses extracted from real field records on which we perform our low complex channel coefficient de-correlation algorithm. The main results of this practical implantation are the following. In non-stationary environments (with some scatterers and some mobility), a significant number of keys (of hundreds of bits each) can be extracted in a very short time under Wifi carriers and under LTE carriers, and these keys have basically low cross correlation at the output and are quite robust to correlation attacks since the quantification step. In stationary environments (with very few scatterers and no mobility, such as encountered in IoT applications) and when no channel coefficient de-correlation algorithm is applied, the extracted keys may be highly correlated and this vulnerability can be exploited by Eve to recover Bob's key. Still in stationary environments, the quantization processing takes a large benefit of our channel coefficient de-correlation algorithm: the key rate is quite decreased but the extracted keys present low cross correlation and are robust to a correlation attack. In any case, the proposed simplified reconciliation step with classical FEC codes provides a significant resilience of the key agreement between Alice and Bob. In any case, the proposed simplified amplification step with classical 2-Universal hash functions provides significant resilience of the key randomness against Eve's attacks, with a limited reduction of the Key lengths: NIST statistical tests were used to show that the keys shared by Alice and Bob are no correlated to the keys extracted by Eve. To the best of our knowledge, this is the first work on a full secret key generation scheme with experimental CSI results using real field WiFi and LTE signals. Our promising results are evidence that the proposed secret key generation scheme can be practically implemented in existing wireless communication systems with minor modifications of the software architecture of nodes and terminals.
Performance Evaluation of a Faster-than-Nyquist System Based on Turbo Equalization and LDPC Codes Albert Abelló Barberán and Damien Roque (ISAE, France); Jean-Marie Freixe (Eutelsat, France); Sébastien Mallier (DGA-MI, France)
In the frame of digital satellite broadcasting by satellite - second generation (DVB-S2), a faster-than-Nyquist (FTN) system based on turbo equalization and low-density parity-check (LDPC) codes is introduced. Truncated maximum a posteriori (MAP) and minimum mean square error (MMSE) equalizers allow a reduced complexity implementation of the FTN system. On the other hand, LDPC codes allow us to demonstrate attractive performance results over an additive white Gaussian noise (AWGN) channel while increasing spectral efficiency beyond the Nyquist rate up to 40 % and keeping a complexity comparable to a current DVB-S2 modem.
Blind Channel Estimation for Massive MIMO Ture Peken, Garrett Vanhoy and Tamal Bose (University of Arizona, USA)
In order to scale with the demand of higher data rates and improved spectral efficiency in next generation wireless communication systems, a large-scale MIMO technology called massive MIMO has been proposed. In massive MIMO, high spectral efficiency can be achieved by the addition of base station antennas instead of increasing transmit power. Pilot-based channel estimation is widely used in conventional MIMO systems where pilot symbols are sent from the user terminals to the base station to estimate the channel. However, the channel estimation in a given cell will be impaired by the pilot symbols transmitted by users in other cells which is called pilot contamination. Therefore, the pilot symbols based channel estimation degrades the performance of massive MIMO systems. Semi-blind and blind channel estimation methods perform channel estimation with short pilot sequences and without pilot symbols respectively. Hence, blind channel estimation is one of the promising solutions to the pilot contamination problem in massive MIMO. However, the computational complexity of blind and semi-blind channel estimation methods are much higher than computational complexity of pilot symbols based methods. We improve the conventional ICA based channel estimation method by making it adaptive. Moreover, we propose a recursive blind adaptive channel estimation method for massive MIMO which has less computational complexity than ICA based channel estimation method. Finally, a MATLAB simulation is presented which compares performance of pilot-based, semi-blind, blind and adaptive blind channel estimation methods over a cluster-based COST 2100 channel model.
The Wireless Innovation Forum does not endorse or recommend the content of any of papers and presentations contained within these proceedings, rather these items are listed strictly as a service for the community. The information and materials contained in these proceedings are provided "as is" without warranty of any kind, either express or implied, including without limitation, any warranty of accuracy, adequacy or completeness of the information and materials, title, non-infringement of third party rights, merchantability, fitness for a particular purpose, etc.
