Numerous algorithmic optimization techniques have been proposed to alleviate the computational complexity of convolutional neural networks. Given the broad selection of AI accelerators, it is not obvious which approach benefits from which optimization most. The design space includes a large number of deployment settings (batch sizes, power modes, etc.) and unclear measurement methods. This research provides clarity into this design space, leveraging a novel benchmarking approach. We provide a theoretical evaluation of different CNNs and hardware platforms, focusing on understanding the impact of pruning and quantization as primary optimization techniques. We benchmark across a spectrum of FPGA, GPU, TPU, and VLIW processors for systematically pruned and quantized neural networks (ResNet50, GoogLeNetv1, MobileNetv1, a VGG derivative, a multilayer perceptron) over many deployment options, considering power, latency, and throughput at a specific accuracy. Our findings show that channel pruning is most effective and works across most hardware platforms, with speedups directly correlated to the reduction in compute load, while FPGAs benefit the most from quantization. Pruning and quantization are orthogonal, and yield optimal design points when combined. Further in-depth results can be found at our web portal, where we share all experimental data, provide data analytics, and invite the community to contribute.

Numerous algorithmic optimization techniques have been proposed to alleviate the computational complexity of convolutional neural networks (CNNs). However, given the broad selection of inference accelerators, it is not obvious which approach benefits from which optimization and to what degree. In addition, the design space is further obscured by many deployment settings such as power and operating modes, batch sizes, as well as ill-defined measurement methodologies. In this paper, we systematically benchmark different types of CNNs leveraging both pruning and quantization as the most promising optimization techniques leveraging a novel benchmarking approach. We evaluate a spectrum of FPGA implementations, GPU, TPU and VLIW processor, for a selection of systematically pruned and quantized neural networks (including ResNet50, GoogleNetv1, MobileNetv1, a VGG derivative, and a multilayer perceptron) taking the full design space into account including batch sizes, thread counts, stream sizes and operating modes, and considering power, latency, and throughput at a specific accuracy as figure of merit. Our findings show that channel pruning is effective across most hardware platforms, with resulting speedups directly correlated to the reduction in compute load, while FPGAs benefit the most from quantization. FPGAs outperform regarding latency and latency variation for the majority of CNNs, in particular with feed-forward dataflow implementations. Finally, pruning and quantization are orthogonal techniques and yield the majority of all optimal design points when combined. With this benchmarking approach, both in terms of methodology and measured results, we aim to drive more clarity in the choice of CNN implementations and optimizations.

Large scale multiple-input multiple-output (MIMO) or Massive MIMO is one of the pivotal technologies for future wireless networks. However, the performance of massive MIMO systems heavily relies on accurate channel estimation. While the acquisition of channel state information (CSI) in such systems requires an increasingly large amount of training overhead as the number of users grows. To tackle this issue, in this paper, we propose a deep learning assisted blind channel estimation technique for orthogonal frequency division multiplexing (OFDM) based massive MIMO systems. We prove that by exploiting the asymptotic orthogonality of the massive MIMO channels, the channel distortion can be averaged out without the prior knowledge of channel impulse responses, and after some mathematical manipulation, different users' transmitted data symbols can be extracted. Thus, by deploying a denoising convolutional neural network algorithm (DnCNN), we mitigate a remaining channel and noise effect to accurately detect the transmitted data symbols at the channel sounding stage. Using the detected data symbols as virtual pilots, we estimate the CSI of all the users at each BS antennas. Our simulation results testify the efficacy of our proposed technique and demonstrate that it can provide a mean square error (MSE) performance which coincides with that of the data-aided channel estimation technique.

Neural Networks have become one of the most successful universal machine-learning algorithms. They play a key role in enabling machine vision and speech recognition and are increasingly adopted in other application domains. Their computational complexity is enormous and comes along with equally challenging memory requirements in regards to capacity and access bandwidth, which limits deployment in particular within energy constrained, embedded environments. To address these implementation challenges, a broad spectrum of new customized and heterogeneous hardware architectures have emerged, often accompanied with co-designed algorithms to extract maximum benefit out of the hardware. Furthermore, numerous optimization techniques are being explored for neural networks to reduce compute and memory requirements while maintaining accuracy. This results in an abundance of algorithmic and architectural choices, some of which fit specific use cases better than others. For system-level designers, there is currently no good way to compare the variety of hardware, algorithm, and optimization options. While there are many benchmarking efforts in this field, they cover only subsections of the embedded design space. None of the existing benchmarks support essential algorithmic optimizations such as quantization, an important technique to stay on chip, or specialized heterogeneous hardware architectures. We propose a novel benchmark suite, QuTiBench , that addresses this need. QuTiBench is a novel multi-tiered benchmarking methodology ( Ti ) that supports algorithmic optimizations such as quantization ( Qu ) and helps system developers understand the benefits and limitations of these novel compute architectures in regard to specific neural networks and will help drive future innovation. We invite the community to contribute to QuTiBench to support the full spectrum of choices in implementing machine-learning systems.

Transmission power control in energy-constrained mobile wireless sensor nodes is a necessary means to extend the battery lifetime and reduce the overall operational expenditure. The body-wearable sensors are perhaps the best examples when they are used for tracking and monitoring health of individuals and to carry out data analysis for possible symptoms of any disease. In indoor environments, the challenge is increased manifold due to complex multi-path propagation and temporal variation of link quality. This paper has enhanced our previous work on the state based adaptive power control protocol (SAPC) to suit well in indoor dynamic radio environments. The proposed protocol adjusts the state-transition rate R (drop-off rate) from a higher to a lower state, depending on the radio link quality. This improvement is implemented in SAPC protocol and renamed as SAPC-R. This protocol is compared with an existing practical adaptive power control protocol (P-ATPC) that can track variations in the radio link quality, and select an appropriate transmission power level and the number of retransmissions. The aim is to keep the algorithm computationally simple as it will run on battery powered devices and therefore are energy-constrained. Simulation results show that SAPC-R can save at least 25% energy as compared to P-ATPC. Results from the experiments that were conducted inside a University building show that by using the SAPC-R algorithm, energy consumption per successful transmission can be reduced by at least 15% as compared to P-ATPC while the packet success rates are comparable. This is a significant improvement, given the small changes in current consumption corresponding to the large changes in transmission power in present-day low power wireless transmitters.

Ultradense Internet of Things (IoT) mesh networks and machine-to-machine communications herald an enormous opportunity for new computing paradigms and are serving as a catalyst for profound change in the evolution of the Internet. The collective computation capability of these IoT devices is typically neglected in favor of cloud or edge processing. We propose a framework to tap into this resource pool by coupling data communication and processing. Raw data captured by sensing devices can be aggregated and transformed into appropriate actions as it travels along the network toward actuating nodes. This paper presents an element of this vision, whereby we map the operations of an artificial neural network onto the communication of a multihop IoT network for simultaneous data transfer and processing. By exploiting the principle of locality inherent to many IoT applications, the proposed approach can reduce the latency in delivering processed information. Furthermore, it improves the distribution of energy consumption across the IoT network compared to a centralized processing scenario, thus mitigating the “energy hole” effect and extending the overall lifetime of the system.

In this paper, we present a spectrum monitoring framework for the detection of radar signals in spectrum sharing scenarios. The core of our framework is a deep Convolutional Neural Network (CNN) model that enables Measurement Capable Devices (MCDs) to identify the presence of radar signals in the radio spectrum, even when these signals are overlapped with other sources of interference, such as commercial Long-Term Evolution (LTE) and Wireless Local Area Network (WLAN). We collected a large dataset of RF measurements, which include the transmissions of multiple radar pulse waveforms, downlink LTE, WLAN, and thermal noise. We propose a pre- processing data representation that leverages the amplitude and phase shifts of the collected samples. This representation allows our CNN model to achieve a classification accuracy of 99.6% on our testing dataset. The trained CNN model is then tested under various SNR values, outperforming other models, such as spectrogram-based CNN models.

We perform an asymptotic study on the performance of filter bank multicarrier (FBMC) in the context of massive multi-input multi-output (MIMO). We show that the signal-to-interference-plus-noise ratio (SINR) cannot grow unboundedly by increasing the number of base station (BS) antennas, and is upper bounded by a certain deterministic value. This is a result of the correlation between the multi-antenna combining tap values and the channel impulse responses between the terminals and the BS antennas. To solve this problem, we introduce a simple FBMC prototype filter design method that removes this correlation, enabling us to achieve arbitrarily large SINR values by increasing the number of BS antennas.

Ultra-dense Internet of Things (IoT) networks and machine type communications herald an enormous opportunity for new computing paradigms and are serving as a catalyst for profound change in the evolution of the Internet. We explore leveraging the communication within IoT to serve data processing by appropriately shaping the aggregate behavior of a network to parallel more traditional computation methods. This paper presents an element of this vision, whereby we map the operations of an artificial neural network onto the communication of an IoT network for simultaneous data processing and transfer. That is, we provide a framework to treat a network holistically as an artificial neural network, rather than placing neural networks within the network. The operation of components of a neural network, neurons and connections between neurons, are performed by the various elements of the IoT network, i.e., the devices and their connections. The proposed approach reduces the latency in delivering processed information and supports the locality of information inherent to IoT by removing the need for transfer to remote data processing sites.

In general, the Low Power Wide Area Internet-of-Things (IoT) wireless sensor networks operate in very narrow radio channels. This causes huge ambiguity when time-difference-of-arrival (TDoA) is used for localization of the sensors. To reduce this ambiguity we propose a solution which, without signal oversampling, approaches the Cramer-Rao lower bound (CRLB) in additive white noise for all useful signal-to-noise ratios (SNRs) by separating estimation of timestamps in two stages. First, the receiver cross-correlates the low bandwidth signal to determine a coarse timestamp estimate. Second, to finetune the coarse estimate, the same narrow band signal is correlated again with its template having much higher time granularity, but only over a narrow interval around the coarse estimate. Further improvement is achieved if the signal is oversampled just enough to accommodate for its slight frequency offset before the consecutive correlations.

This paper presents a novel Long Term Evolution (LTE) mobile wireless testbed for streaming multimedia Quality of Experience (QoE) experimentation. The software-radio based design of the testbed provides an extremely flexible architecture which offers detailed insights into all elements of the entire LTE network protocol stack from physical to application layer. With the testbed, we aim to bridge the gap between video content providers such as YouTube and commercial LTE operators by identifying and testing viable approaches to enhance end-user quality of experience (QoE). In this work, we focus on the design and implementation of the automated and instrumented testbed, and share results from our initial experiments.

We study the possibility of removing the cyclic prefix (CP) overhead from orthogonal frequency division multiplexing (OFDM) in massive multiple-input multiple-output (MIMO) systems. We consider the uplink transmission, while our results are applicable to the downlink as well. The absence of CP increases the spectral efficiency in expense of intersymbol interference and intercarrier interference. It is known that in massive MIMO, the effects of uncorrelated noise and multiuser interference vanish as the number of base station antennas tends to infinity. To investigate if the channel distortions in the absence of CP fade away, we study the performance of the standard maximum ratio combining receiver. Our analysis reveals that in this receiver, there always remains some residual interference leading to saturation of signal-to-interference-plus-noise ratio. To resolve this problem, we propose using the time reversal (TR) technique. Moreover, in order to further reduce the multiuser interference, we propose a zero-forcing equalization to be deployed after the TR combining. We compare the achievable rate of the proposed system with that of the conventional CP-OFDM. We show that in realistic channels, a higher spectral efficiency is achieved by removing the CP from OFDM, while reducing the computational complexity.

In this work we propose a virtualized network architecture for an infrastructure provider that shares the physical resources of a Massive MIMO cell among several virtual network operators (VNOs) using spatial multiplexing. In this architecture the infrastructure provider allocates spatial streams to the VNOs, which enables each VNO to select its own scheduling policy and user priority to differentiate its service from the other VNOs. To assign the spatial streams to the VNOs that value them the most, we propose an auction-based spatial stream allocation approach. We show that the proposed auction- based approach performs very close to the optimal (fixed) approach in the case of homogeneous static VNOs demand. In case of heterogeneous demands, the auction mechanism is able to dynamically allocate the resources according to the needs of different VNOs.

Flexible licensing model is a necessary enabler of the technical and procedural complexities of Spectrum Access System (SAS)-based sharing framework. The purpose of this study is to explore the effectiveness of 3.5GHz Licensing Framework which is based on census tracts as license area units. Their main characteristic is population - as such, the boundary of census tract does not follow the edge of wireless network coverage. We demonstrate why census tracts are not suitable for small cell networks licensing, by: (1) gathering and analysing the official census data, (2) exploring the boundaries of census tracts which are in the shape of non-convex polygons, and (3) giving a measure of effectiveness of the licensing scheme through metrics of area loss and the number of people per census tract with access to spectrum. Results show that census tracts severely impact the effectiveness of the licensing framework since almost entire strategically important cities in the U.S. will not avail from spectrum use in 3.5 GHz band. Our paper does not seek to challenge the core notion of geographic licensing concept, but seeks a corrective that addresses the way the license is issued for a certain area of operation. The effects that inappropriate size of the license has on spectrum assignments lead to spectrum being simply wasted in geography, time and frequency. The corrective is necessary since the main goal of promoting innovative sharing in 3.5 GHz band is to put spectrum to more efficient use.

Head portraits are popular in traditional painting. Automating portrait painting is challenging as the human visual system is sensitive to the slightest irregularities in human faces. Applying generic painting techniques often deforms facial structures. On the other hand portrait painting techniques are mainly designed for the graphite style and/or are based on image analogies; an example painting as well as its original unpainted version are required. This limits their domain of applicability. We present a new technique for transferring the painting from a head portrait onto another. Unlike previous work our technique only requires the example painting and is not restricted to a specific style. We impose novel spatial constraints by locally transferring the color distributions of the example painting. This better captures the painting texture and maintains the integrity of facial structures. We generate a solution through Convolutional Neural Networks and we present an extension to video. Here motion is exploited in a way to reduce temporal inconsistencies and the shower-door effect. Our approach transfers the painting style while maintaining the input photograph identity. In addition it significantly reduces facial deformations over state of the art.

GFDM (Generalized Frequency Division Multiplexing) is a multicarrier modulation technique that is being proposed as a potential candidate for the fifth generation of wireless communication systems (5G). Due to the fact that GFDM uses only one cyclic prefix (CP) for a group of symbols rather than a CP per symbol and it has a well contained spectral properties, it is more bandwidth efficient than the widely used OFDM modulation. In this paper, we propose novel receiver designs for GFDM by taking advantage of the particular structure in the modulation matrix. A unified receiver structure for matched filter (MF), zero forcing (ZF) and minimum mean square error (MMSE) receivers is derived. Our proposed MF receiver is based on sparsification of the modulation matrix using block discrete Fourier transform (DFT) matrix. The proposed ZF and MMSE receiver algorithms in this paper harness the special block circulant property of the matrices involved in the demodulation stage to reduce the computational cost of the system implementation. In addition, our algorithms do not incur any performance loss as no approximation is involved. The computational costs of our proposed techniques are analyzed in detail and are compared with the existing solutions that are known to have the lowest complexity. It should be highlighted that a substantial computational complexity reduction can be achieved by adopting our techniques.

The need to carry out analytical studies of wireless systems often motivates the usage of simplified models which, despite their tractability, can easily lead to an overestimation of the achievable performance. In the case of dense small cells networks, the standard single slope path-loss model has been shown to provide interesting, but supposedly too optimistic, properties such as the invariance of the outage/coverage probability and of the spectral efficiency to the base station density. This paper seeks to explore the performance of dense small cells networks when a more accurate path-loss model is taken into account. We first propose a stochastic geometry based framework for small cell networks where the signal propagation accounts for both the Line-of-Sight (LOS) and Non-Line-Of-Sight (NLOS) components, such as the model provided by the 3GPP for evaluation of pico-cells in Heterogeneous Networks. We then study the performance of these networks and we show the dependency of some metrics such as the outage/coverage probability, the spectral efficiency and Area Spectral Efficiency (ASE) on the base station density and on the LOS likelihood of the propagation environment. Specifically, we show that, with LOS/NLOS propagation, dense networks still achieve large ASE gain but, at the same time, suffer from high outage probability.

As wireless networks shift towards denser deployments, simulations take longer time to run and consume more resources. Simulation of the physical (PHY) layer is particularly time consuming and, as such, its implementation requires the use of abstracted models with low computational complexity, in order to provide accurate results in reasonable time. Although Mutual Information Effective SINR Mapping (MIESM) is commonly used as the standard method to abstract the PHY layer, the popular "modified Shannon's equation" - proposed by Mogensen et al. - has been gaining ground as a simpler and lighter implementation of the PHY layer in System Level (SL) simulations. This paper aims at comparing MIESM and the modified Shannon's formula for SL simulators in terms of accuracy and computational complexity. Results show that Shannon's method proves beneficial for networks with a large number of users, as it is less complex and allows us to reduce up to 30 times the simulation time required by the PHY layer abstraction. On the other hand, the accuracy is in the range of 80-87% for the instantaneous throughput compared to MIESM; the accuracy improves up to 95% for the time-averaged user throughput. In scenarios involving fast-varying channels, the modified Shannon's formula needs to be calibrated considering the actual user speed; in this case, accuracy of 90% can be achieved for a user speed of 60kmph.

Public Protection and Disaster Relief (PPDR) services can benefit greatly from the availability of mobile broadband communications in disaster and emergency scenarios. While undoubtedly offering full control and reliability, dedicated networks for PPDR have resulted in high operating costs and a lack of innovation in comparison to the commercial domain. Driven by the many benefits of broadband communications, PPDR operators are increasingly interested in adopting mainstream commercial technologies such as Long Term Evolution (LTE) in favour of expensive, dedicated narrow-band networks. In addition, the emergence of virtualization for wireless networks offers a new model for sharing infrastructure between several operators in a flexible and customizable manner. In this context, we propose a virtual Public Safety (PS) operator that relies on shared infrastructure of commercial PPDR networks to deliver services to its users. We compare several methods of allocating spectrum resources between virtual operators at peak times and examine how this influences differing traffic services. We show that it is possible to provide services to the PS users reliably during both normal and emergency operation, and examine the impact on the commercial operators.

In this paper we present a case study of a project carried out in a telecommunications research center. The Postcards from the near Future project sought to begin to reflexively address the difficulties faced by telecommunications researchers in communicating their specialist research with colleagues and stakeholders across academia, industry and with a wider public. We devised an experimental practice based research project, Postcards from the near Future (PFTNF), as a pedagogical structure to support reflexive practical learning within the research centre to this effect. The PFTNF project invited PhD students and Post-Doctorate researchers within the telecommunications research lab to express their technical research through making postcards. The postcards created visual and narrative fictions of a future where research carried out within the telecoms lab might actually become deployed. Our case study of the PFTNF project provides insights into the nature of the challenge of communicating complex technical research. We show how the experimental communicative practice developed through the PFTNF project influenced researchers and catalyzed change in the communicative culture of the telecommunications lab. We argue that the kind of experimental communicative practice engendered within the PFTNF project merits further research and development, and we make suggestions to this effect.

Current wireless research is dominated by the need to satisfy ever-increasing demands for high-speed mobile wireless data. To satisfy these demands, researchers employ flexible, agile software-based radio systems which efficiently use the resources available to them. Dynamic Spectrum Access (DSA) and cognitive radio are key concepts for achieving these research goals. This paper presents Spectrum Wars, a game designed to inform about DSA and cognitive radio but which has also proven to be an effective tool in educating about many of the basic principles of wireless telecommunications. The game pits two teams of players against each other to create and maintain wireless connections in a dynamic environment in order to transfer data as quickly as possible between a transmitter and receiver. Built upon Iris, an open-source framework for software radio, Spectrum Wars is freely available to be recreated, reused, expanded and improved. The paper explains the background of DSA and cognitive radio and gives an overview of the Iris framework. It discusses related work including the MANIAC challenge and the DARPA spectrum challenge before presenting the game itself, examining the feedback of players and exploring the possibilities for future work and improvement.

In this paper we investigate the effect of Line-of-Sight (LOS) and Non-Line-of-Sight (NLOS) propagation on the Area Spectral Efficiency (ASE) and on the energy efficiency of dense small-cell networks. We show that including both LOS and NLOS propagation in the path-loss model provides a completely different picture of the behaviours of ASE and energy efficiency than what would be observed in case of either LOS or NLOS propagation only. In particular, with combined LOS/NLOS path-loss, the ASE exhibits superlinear and sublinear behaviour at low and high cell densities, respectively. In addition, the energy efficiency as a function of the cell density has a global maximum and is not a monotonically increasing function like in case of LOS or NLOS propagation only. Based on our findings, we claim that Line-of-Sight (LOS) and Non-Line-of-Sight (NLOS) propagations play an important role in studying the performance of extremely dense small-cell networks.

Telecommunication systems have evolved, from being simple monolithic structures to complex ones. The present paper is an attempt to recast telecom systems in the language of complex systems. As a specific application of complex systems science to telecom systems, we tackle the problem of self-organizing frequency allocation. For this purpose, we model the problem of frequency assignment within the framework of cellular automata and we present an algorithm that reaches an interference-free allocation of the channels in a finite number of steps. By applying a measure of spatial structure and pattern to the two-dimensional lattice representing the frequency allocation, our study shows that autonomous networks assigning frequencies to their cells in a self-organized way, can be defined and therefore studied as complex systems. On the contrary, traditional networks based on centralized pre-planned frequency allocation cannot be considered as complex, according to the meaning given to the word by complex systems science. The present paper is to be considered as a step towards a comprehensive and rigorous study of Complex Communication Systems (CCS), adopting in communication networks design and analysis, the results and philosophy underlying the emerging multi-disciplinary field of Complex Systems Science.

Wireless network virtualization enables multiple virtual wireless networks to coexist on shared physical infrastructure. However, one of the main challenges is the problem of assigning the physical resources to virtual networks in an efficient manner. Although some work has been done on solving the embedding problem for wireless networks, few solutions are applicable to dynamic networks with changing traffic patterns. In this paper we propose a dynamic greedy embedding algorithm for wireless virtualization. Virtual networks can be re-embedded dynamically using this algorithm, enabling increased resource usage and lower rejection rates. We compare the dynamic greedy algorithm to a static embedding algorithm and also to its dynamic version. We show that the dynamic algorithms provide increased performance to previous methods using simulated traffic. In addition we formulate the embedding problem with multiple priority levels for the static and dynamic case.

This paper introduces filter bank multicarrier (FBMC) as a potential candidate in the application of massive MIMO communication. It also points out the advantages of FBMC over OFDM (orthogonal frequency division multiplexing) in the application of massive MIMO. The absence of cyclic prefix in FBMC increases the bandwidth efficiency. In addition, FBMC allows carrier aggregation straightforwardly. Self-equalization, a property of FBMC in massive MIMO that is introduced in this paper, has the impact of reducing (i) complexity; (ii) sensitivity to carrier frequency offset (CFO); (iii) peak-to-average power ratio (PAPR); (iv) system latency; and (v) increasing bandwidth efficiency. The numerical results that corroborate these claims are presented.

Orthogonal frequency division multiplexing (OFDM) systems suffer from large out-of-band (OOB) emissions. These OOB emissions are undesirable and may cause harmful interferences to adjacent channels (users). In this paper, we investigate the ability of combining two OOB emissions reduction techniques (the advanced subcarrier weighting (ASW) and the advanced cancellation carriers (ACC) techniques). The combination is proposed to provide significant OOB emissions reduction without degrading the performance of the communication system. Both techniques do not require performing complex optimizations or excessive computations which makes the combination suitable for practical systems. Simulation results show that significant OOB emissions reduction can be achieved using the proposed combination at the expense of slight BER degradation. Moreover, the combination results in negligible peak-to-average power ratio (PAPR) increase. The simplicity of the chosen techniques allows us to implement the combination in a software defined radio. Implementation results prove that the proposed combination can be introduced for real-time applications.

Multiple carrier frequency offsets (CFOs) present in the uplink of orthogonal frequency division multiple access (OFDMA) systems adversely affect subcarrier orthogonality and impose a serious performance loss. In this paper, we propose the application of time domain receiver windowing to concentrate the leakage caused by CFOs to a few adjacent subcarriers with almost no additional computational complexity. This allows us to approximate the interference matrix with a quasi-banded matrix by neglecting small elements outside a certain band which enables robust and computationally efficient signal detection. The proposed CFO compensation technique is applicable to all types of subcarrier assignment techniques. Simulation results show that the quasi-banded approximation of the interference matrix is accurate enough to provide almost the same bit error rate performance as that of the optimal solution. The excellent performance of our proposed method is also proven through running an experiment using our FPGA-based system setup.

One of the drawbacks of orthogonal frequency division multiplexing (OFDM) systems is the large out-of-band (OOB) emissions. These emissions are undesirable and may cause harmful interference to adjacent channels (users). Reducing the OOB emissions may allow more users to access the spectrum especially in the case of dynamic spectrum access networks. In this paper, we investigate the ability of combining two practical OOB emissions reduction techniques. The combination is proposed to mitigate the drawbacks of each individual technique while providing significant OOB emissions reduction. The chosen techniques do not require performing complex optimizations which makes the combination suitable for practical implementations. The effectiveness of the proposed combination is verified via computer-based simulations as well as practical implementation. The implementation is done using Iris software defined radio along with the universal software radio peripheral (USRP-N210). Implementation results prove that this combination can be used in real-time applications.

This paper suggests that the regulatory tool known as the Block Edge Mask (BEM), which is used to define the technical conditions governing a spectrum license, can be used as an instrument to jointly drive spectrum and energy efficiency in a network. The paper explores this concept and goes on to show how networks can generate waveforms which comply with the energy limits of the BEM and at the same time make best use of the spectral resources available to them. The focus of the paper is on orthogonal frequency division multiplexing (OFDM) waveforms. The spectrum- and energy-efficient BEM-compliant waveforms are generated using a process that optimally combines symbol shaping and subcarrier-power loading to create a waveform that minimizes the difference between the transmitted wave and BEM. Practical considerations relating to peak-to-average-power-ratio (PAPR) reduction and time and frequency synchronization issues are explored to ensure the BEM-compliant waveforms can be transmitted with minimal changes to the transceiver. It is envisaged that these techniques will be used by transmitters that change their frequency of operation and in doing so need to shape their transmitted waveforms to comply with the different BEMs governing the use of the different spectrum blocks they wish to access.

The purpose of this paper is to present a vision for future mobile and wireless networks. The vision, which we call Networks without Borders (NwoB), is based on a marketplace of virtual network operators which construct networks from a pool of shared resources (e.g., base stations, spectrum, core network components, cloud resources, processing capabilities, etc.). The resources will be sourced from traditional industry players as well as crowdsourced from individuals. The paper describes this approach from a value-chain perspective. The proposed value chain is substantially different from the value-chain models that are currently used to illustrate mobile and wireless networks. The economic imperatives and innovation drivers for this approach are discussed. Early work showing the promise of this vision is presented. This work focuses on diverse examples which advocate the removal of traditional and historical restrictions on spectrum and infrastructure and move toward more dynamic use of shared resources. In the first example, we look at how frequency-division-duplexing (FDD) and time-division-duplexing (TDD) restrictions on spectrum usage can be relaxed; we remove the borders between TDD and FDD. In the second example, we look at the aggregation and pooling of corporate infrastructure which uses exclusive spectrum and removes the borders between different mobile operators. Finally, we look at the aggregation of user-deployed or crowdsourced infrastructure that opportunistically uses spectrum and removes the borders between independently deployed hotspots. These are starting points, and the full realization of the vision will involve more dynamic access to spectrum and more extensive sharing of infrastructure. Hence, the final part of the paper describes the resulting research challenges.

In this paper we explore a means of allowing independent service providers using whatever spectrum they can find, to come together in a loose form of coalition which supports mobility between service providers. The service providers are not required to use similar network configurations- i.e. they can configure to best suit their needs and in line with what spectrum they find available. Our solution involves the use of what we term a cyclostationary codec which in essence is a physical pattern that can be placed in the control channel of the service provider. This pattern is the key enabling mechanism for identifying the coalition, enabling rendezvous, supporting handover and conveying details of the service provider configuration.

Orthogonal frequency-division multiplexing (OFDM) systems suffers from large out-of-band (OOB) emissions caused by high sidelobes of the modulated subcarriers. In the literature, performing considerable OOB emissions reduction at all frequencies in the OOB regions can be done at the expense of wasting the available resources drastically. In this paper, we introduce the idea of combining two efficient OOB emissions reduction techniques (the shaping and the advanced cancellation carriers (ACC)). These two techniques are chosen to enable considerable reduction at all frequencies in the OOB regions while at the same time to utilize the available resources. Also, these techniques do not require performing complex optimizations (which is essential for practical implementations). The proposed combination is implemented in a software defined radio and tested for over-the-air transmissions.

This paper focuses on a solution for a spectrum access system that can be deployed in the 3.5 GHz band to enable sharing of these bands between existing incumbents and new small cell systems. The FCC envisages some kind of dynamic data base as a means of supporting this sharing. In this paper we recommend that the database should support the use of adaptive spectrum masks, defined at the cell level, to control the emissions of the systems sharing the bands. The severity of the mask is context dependent and hence the least restrictive mask rather than one that represents a worst case scenario is used. Calculating the least restrictive spectrum mask in a context dependent manner involves examining an infinite number of potential scenarios. Our solution makes use of a pixelated database which results in large but finite no of possibilities and therefore converts the problem to a tractable one. A library of masks, each one corresponding to a specific scenario, is then stored in theh database. The fact that users must register with the database to access spectrum means it is possible to determine the distribution and density of users (i.e. the context). and retrieve the best mask. This will allow for far more efficient use of the spectrum then in a traditional system while still guaranteeing a certain QoS for the users.

Orthogonal frequency division multiple access (OFDMA), where different subcarriers are allocated to different users, has been adopted for the uplink of several standards and has attracted a great deal of attention as a result. However, OFDMA is highly sensitive to carrier frequency offset (CFO) between the transmitter and receiver. In the uplink, different carrier frequency offsets due to different users can adversely affect subcarrier orthogonality. We propose a low complexity CFO compensation approach that addresses this problem while maintaining optimal performance. This approach is based on the least squares and minimum mean square error criteria applicable to interleaved and block interleaved carrier assignment schemes. The proposed algorithms use the special block circulant property of the interference matrix. In contrast to existing CFO compensation techniques, our algorithms do not rely on iterations or approximations. We present our approach in this paper and describe how a considerable reduction in computational complexity can be achieved by adopting it.

The concept of cancellation carriers (CCs) has been proposed in the literature for sidelobe suppression for orthogonal frequency division multiplexing (OFDM) systems. Subcarriers at the edges of the OFDM spectrum are used for sidelobe reduction while the remaining subcarriers are used for data transmission. Existing CCs techniques require performing complex optimization that should be applied for each OFDM symbol which is not suitable for real-time applications. In this paper, we propose a heuristic approach for CCs. The proposed algorithm involves few computations compared with all other techniques proposed in the literature. Simulation results show that sidelobe reduction performance can be close to the optimal. Moreover, the proposed algorithm is implemented in a software defined radio and implementation results prove that it can be introduced for real-time applications.

The emergence of cognitive radio and software defined networks are signs of a network world that is much more dynamic. These kinds of systems support reconfiguration of the radio or the network to suit user demands and prevailing conditions so that bandwidth and other resources can be directed to where they are needed. As communication systems become ever more sophisticated, achieving high computational capacity while supporting a high level of flexibility becomes more of a challenge. In this paper we focus on the computational complexity and flexibility that will be required of future cognitive radios. The challenge is to achieve sufficient computational capacity, in particular for processing wide-band high-bitrate waveforms, within acceptable size, weight and power consumption limits while supporting significant flexibility. One method of addressing this problem is through the use of hybrid processing architectures that provide both programmability and performance. In particular we focus on a cognitive radio framework called Iris and explore how Xilinx's Zynq system-on-chip (SoC) can deliver the kind of performance that is needed of future cognitive radio systems. The Zynq SoC has the flexibility offered by an ARM based processing system (PS), while also having performance-orientated programmable logic (PL). The challenges of multi-core programming on embedded platforms and the difficulties of accelerating the cognitive radio functionality in the PL are discussed.

The concept of subcarrier weighting (SW) has been proposed in the literature for sidelobe suppression for OFDM systems. Each OFDM subcarrier is multiplied with a real-valued weight that is determined in order not to interfere with adjacent users. The problem with the SW technique is involving a very complex optimization that has to be performed for each OFDM symbol which limits the suitability for real-time applications. In this paper, we propose a heuristic approach called advanced subcarrier weighting (ASW). It can achieve considerable sidelobe suppression while requiring significantly less computational resources than the optimal solution. Simulation results show that the performance of the proposed algorithm can be close to the optimal performance. Moreover, the proposed ASW algorithm is implemented in a software defined radio. Implementation results prove that it can be introduced for real-time transmissions.

This paper considers spatial spectrum sharing-based carrier aggregation (SSS-CA) from a game theoretic perspective. In SSS-CA, a network operator can not only transmit on its own licensed spectrum but it can also access and aggregate the licensed spectrum of other operators on payment of a certain price. The difference between operators and the aggregators in each licensed spectrum makes this network heterogeneous. We first model the pairing problem between potential operators as a pairing game and then derive the condition for which both operators are incentivized to form an SSS-CA pair. We then introduce the power control game to derive the optimal transmit power of each spectrum aggregator. Finally, we consider the pricing optimization problem by forming a pricing adjustment game. We observe that these three problems are linked by the price function of the operators and hence can be jointly optimized by using a hierarchical game theoretic framework. We derive the Stackelberg equilibrium for the pricing and power joint optimization problem and present the numerical results to compare the performance improvement brought by our proposed joint optimization method.

Heterogeneous Networks (HetNet) have been proposed as a means of boosting the coverage in areas where users experience weak wireless connectivity. In HetNets, due to the massive deployment of Low Power Nodes (LPN) along side the usual Macro Base Stations (MBS), inter-cell interference becomes a significant issue which, if not properly handled, may degrade the network throughput and the macro- cell edge users performance. In this paper, we focus on the problem of the inter-cell interference to the Macro User Equipments (MUE) generated by the LPNs. We propose an algorithm for flexible spectrum usage among LPNs that enhances macro cell-edge user throughput. By means of a dynamic redistribution of the frequency resources among MBSs and LPNs, this algorithm reduces the harmful interference generated by the LPNs to the macro cell-edge users, hence increasing the throughput of the latter. In addition to helping the terminals which experience high interference, the proposed solution also limits the overall reduction in network throughput, which is usually one of the side effects of interference coordination in HetNets. The simulation results show that, compared to a full frequency reuse for the LPNs, the proposed algorithm improves the throughput of the macro cell-edge users, which are usually the ones experiencing the lowest performance within the network. Overall, with limited impact in terms of HetNet throughput reduction and with good benefit to the cell-edge users performance, this algorithm represents an effective solution to the inter-cell interference problem affecting the HetNets.

We report on an emerging application focused on the design of resilient long reach passive optical networks using combinatorial optimisation techniques. The objective of the application is to determine the optimal position and capacity of a set of metro nodes. We specifically consider dual parented networks whereby each customer must be associated with two metro nodes. An important property of such a placement is resilience to single node failure. Therefore excess capacity should be provided at each metro node in order to ensure that customers can be redistributed amongst the metro sites. Our application, as well as finding optimal node placements, can compute the minimum level of excess capacity on all metro nodes. In this paper we present three alternative approaches to optimal metro node placement. We present a detailed analysis of the impact of different placement approaches on the distribution of excess capacity throughout the network. We show that preferential distributions occur in practice, based on a case-study in Ireland. Finally we show that load and excess capacity provision are independent of each other.

We present an application focused on the design of resilient long-reach passive optical networks. We specifically consider dual parented networks whereby each customer must be connected to two metro sites via a local exchange sites. An important property of such a placement is resilience to single metro node failure. The objective of the application is to determine the optimal position of a set of metro-nodes such that the total optical fibre length is minimised. We prove that the decision variant of this problem is NP-Complete. We present three alternative combinatorial optimisation approaches to finding an optimal metro node placement using: a mixed integer linear programming formulation of the problem, a hybrid approach that uses clustering as a preprocessing step, and, finally, a local search approach. We consider a detailed case-study based on a network for Ireland. The hybrid approach scales well and finds solutions that are close to optimal, with a runtime that is two orders-of-magnitude better than the MIP model. The local search approach is consistently good on all benchmarks.

Cognitive radio is a promising technology for fulfilling the spectrum and service requirements of future wireless communication systems. Real experimentation is a key factor for driving research forward. However, the experimentation testbeds available today are cumbersome to use, require detailed platform knowledge, and often lack high level design methods and tools. In this paper we propose a novel cognitive radio design technique, based on a high-level model which is implementation independent, supports design-time correctness checks, and clearly defines the underlying execution semantics. A radio designed using this technique can be synthesised to various real radio platforms automatically; detailed knowledge of the target platform is not required. The proposed technique therefore simplifies cognitive radio design and implementation significantly, allowing researchers to validate ideas in experiments without extensive engineering effort. One example target platform is proposed, comprising software and reconfigurable hardware. The design technique is demonstrated for this platform through the development of two realistic cognitive radio applications.

This paper describes a method to mitigate interference when the cyclic prefix (CP) is totally removed for shaped orthogonal frequency division multiplexing (OFDM) signals. The CP is removed to increase system throughput while shaping is applied to increase the spectral efficiency. Samples in the shaping extension are used to restore part of the cyclicity information missing in the $i$th received symbol. Furthermore, optimally weighted samples of the $(i+1)$th symbol are used to restore the remaining part of the cyclicity information with some residual error. Simulation results show that the bit error rate (BER) can be very close to the optimal bound in certain scenarios, while the complexity added to the standard OFDM receiver is low.

Adaptive systems are set to become more mainstream, as numerous practical applications in the communications domain emerge. FPGAs offer an ideal implementation platform, combining high performance with flexibility. While significant research has been undertaken in the area of FPGA partial reconfiguration, it has focussed primarily on low-level architecture-specific implementations. Building upon this previous work, we present a system model and software architecture for implementing runtime adaptive applications on FPGAs, separating the control and processing planes and abstracting away the details of hardware reconfiguration from the system designer. Hardware processing components appear as software components in the runtime system, enabling their inclusion in adaptive applications. We present an adaptive wireless application, demonstrating the use of the model and software architecture.

This paper identifies important features a cognitive radio framework should provide, namely a virtual architecture for hardware abstraction, an adaptive run-time system for managing cognition, and high level design tools for cognitive radio development. We evaluate a range of existing frameworks with respect to these, and propose a novel FPGA-based framework that provides all these features. By abstracting away the details of hardware reconfiguration, radio designers can implement FPGA-based cognitive nodes much as they would do for a software implementation. We apply the proposed framework to the design and implementation of a receiver node that works in two modes: discovery, where it uses spectrum sensing to find a radio transmission, and communication, in which it receives and demodulates the said transmission. We show how the whole design process does not require any hardware experience on behalf of the radio designer.

In this paper the feasibility of femtocells partially reusing GSM spectrum is investigated. Due to the increasing scarcity and cost of spectrum, partial spectrum reuse by secondary networks is being considered to improve spectral efficiency. Additionally, new revenue streams can be extracted from existing resources by utilising partial spectrum reuse. The partial reuse of spectrum depends largely on the ability of the secondary network to mitigate the negative interference impact upon the primary network. This can be achieved by the establishment of acceptable interference criteria. This paper presents a solution to the problem of meeting the interference criteria with regards to frequency allocation and power configuration within the femtocell network. We demonstrate that these solutions can be applied successfully to allow for partial spectrum reuse in the GSM band. Furthermore we present an analysis of both the theoretical performance of femtocells and the resulting impact on the macrocellular network.

Cognitive radio promises better spectrum utilisation through decentralised control, sensing and decision making. The devices are intended to be inexpensive, upgradeable and adaptable. However, introducing upgraded devices into a decentralised environment may cause unexpected results, the upgrades may require a critical mass to show any benefit, and may have an adverse impact on existing devices, which may deter their adoption. We study, in simulation, the gradual introduction of cognitive devices into an existing population using a spectrum access etiquette, in three different scenarios for the device capability. We show that the new devices gain immediate benefits, and that there is always an incentive to upgrade. We show that overall spectrum efficiency is improved. Finally, we show that there is no negative impact on existing devices, and that the introduction of cognitive devices may even improve the success rate of lesser devices.

We define a cyclostationary signature as a feature which may be intentionally embedded in a digital communications signal, detected through cyclostationary analysis and used as a unique identifier. The purpose of this paper is to demonstrate how cyclostationary signatures can be exploited to overcome a number of the challenges associated with network coordination in emerging cognitive radio applications and spectrum sharing regimes. In particular we show their uses for signal detection, network identification and rendezvous and discuss these in the context of dynamic spectrum access. We present a theoretical discussion followed by application-oriented examples of the cyclostationary signatures used in practical cognitive radio and dynamic spectrum usage scenarios. We focus on orthogonal frequency division multiplexing (OFDM) based systems and present an analysis of a transceiver implementation employing these techniques developed on a cognitive radio test platform.

Current research trends in networks and telecommunications suggest that future architectures will aim for the convergence not only of fixed and mobile infrastructures, but also of different network layer technologies. Proposals like OCALA from UC Berkeley and our own TRANSIT can be used to converge different architectures and network layer overlays while supporting legacy applications. In this paper we analyse and assess the security threats against such systems. Our contribution constitutes the first step in designing security models for convergence architectures as part of their design and deployment phases, rather than as retrofitted mechanisms.

The Centre for Telecommunications Value-Chain Research (CTVR) and the Center for Wireless Telecommunications (CWT) are undertaking a joint collaborative research project investigating the potential of different software-defined radio (SDR) and cognitive radio (CR) systems that can interoperate and co-exist in common frequency bands.This paper presents some key results from this experience so far, including an over-the-air interoperability analysis of two different SDR architectures in the same frequency band.An initial analysis of actual co-existence experiences involving primary users and secondary opportunistic spectrum users in a common frequency band is also presented.

With the advent of cognitive radio technology, new paradigms for spectrum access can achieve near-optimal spectrum utilisation by letting each user sense and utilise available spectrum opportunistically while regulating the interference it imposes on other users through interference constraints. However, the simplest and most common forms of such constraints are binary and transmitter-centric, which are often inefficient since they only consider pair-wise sets of transmitters. Hence, we propose a non-binary receiver-centric constraint model for spectrum access in cognitive radio networks. Such a model is in line with the recently proposed interference temperature metric that constraints whole subsets of transmitters, thereby permitting interfering signals to be introduced and enabling additional communication, leading to improved spectrum utilisation. These constraints are easy to generate and check, and are currently being used to devise a co-operative negotiated etiquette for cognitive radios offering heterogeneous services in a wireless office networking scenario.

The ever increasing demands for access to frequency spectrum and more efficient methods of using this spectrum involving multi-carrier techniques are driving the development of reconfigurable and cognitive wireless devices. The potential for, and the need to accommodate the potentially time-varying entropy in these wireless systems presents several design and implementation challenges. However, significant opportunities for innovation extending from the wireless terminal implementation to the business and market models also exist. This paper focuses on the physical (PHY) and medium-access control (MAC) layer issues of these emerging systems, highlighting the following three main contributions: An introduction to reconfigurable multi-carrier code division multiple access (MC-CDMA), an exploration of the use of reconfigurable MC-CDMA and its impact on the design of the physical (PHY) and medium access control (MAC) layers in a cognitive radio context and overview of the practical design and development considerations of this novel system which is being implemented on a real cognitive radio development testbed.

Abstract — With the advent of cognitive radio technology, new paradigms for spectrum access can achieve near-optimal spectrum utilization by letting users sense and utilize available spectrum opportunistically. Recently, graph-coloring models have been used to produce fair, efficient and optimal spectrum allocations for a given network topology. However, topology changes (e.g., due to user mobility) can incur significant signaling traffic to reallocate spectrum, consuming valuable resources. We propose a frequency-spread multiple access (FSMA) scheme that is topology-transparent, and numerical results suggest that it is suited for opportunistic spectrum access in a sparsely-connected, low-density mobile ad-hoc network. I.

The performance evaluation of ad hoc network components through simulation allows for isolation of parameters, easy access to global quantities and a statistically significant number of repeatable trials. In designing network protocols, it is important to investigate the relationship between global network performance and the underlying locally observed network characteristics. Mobility models are used to define the movement of nodes in the simulation space. A wide variety of mobility models exist, and the choice of mobility model has significant influence over locally observed metrics. Unfortunately, many mobility models exhibit spatial and temporal non-stationarity of important metrics, such that global averages of certain metrics are not representative of local observations. This finding impacts on the development of adaptive ad hoc architectures. This paper explores the spatial stationarity of frequently used mobility models. It focuses on the importance of stationarity in relation to the evaluation of performance studies. The spatial non-stationarity of link statistics, and the impact of this artefact on network performance evaluation is examined for the Toroidal Random Waypoint and Random Direction mobility models. We show that the Toroidal Random Waypoint model exhibits spatial stationarity of link statistics. Furthermore, we demonstrate that network performance studies in spatially stationary environments produce dissimilar results to those produced by non-stationary environments. This shows the importance of spatial stationarity when investigating global network performance and locally observed network characteristics.

The dynamic nature of ad hoc networks advocates the use of adaptive schemes to optimize network performance. Such adaptive schemes require local observations of prevailing network conditions. This paper discusses the dependencies between locally observable metrics. The probability density functions for link statistic metrics are assessed, and the implications of the distribution model for parametric inference is investigated.

The purpose of this paper is to present an overview of a platform for exploring reconfigurability in wireless networks. The types of networks that are particularly of interest here are distributed and disaggregated networks. The term disaggregated is used to describe networks that are not alone distributed but also non-homogenous and controlled by different and possibly competing entities. In distributed and disaggregated networks, network-wide reconfiguration, if indeed possible, will not be achieved through a coordinating or controlling entity as a global view or global control of the network does not exist Therefore mechanisms for facilitating network-wide reconfiguration based on local knowledge only are needed. The platform presented in the paper combines research outcomes from ad hoc networking research and work carried out in the area of reconfigurable radio design in an effort to address these issues

This paper describes a self-stabilizing protocol which enables the reactive auto-configuration of multi-protocol mobile ad hoc network nodes. The spatially and temporally diverse nature of ad hoc networks suggests the development of a more sophisticated network-layer that enables the network's nodes to choose their preferred routing protocol from an array of protocols with varying abilities. However, the configuration of such a dependent variable, and one with network-wide significance, is a non-trivial task in a distributed system. This paper focuses on the problems that arise in developing a configuration protocol which does not compromise the inherent distributed, open and infrastructure-independent traits of an ad hoc networking system. Self-stabilization is identified in this paper as a suitable technique on which to base such a robust and distributed autonomous configuration protocol. The protocol takes the form of a stateless configuration initialisation (SCl) protocol which is coupled with a configuration conflict detection & resolution (CCDR) protocol.

In this paper, we present findings relating to frame synchronization algorithms for software radio based OFDM systems in addition to an overview of the design and implementation of an (OFDM) software radio based on a general-purpose processor (GPP) platform. The objectives of this work are to advance on existing research carried out in the area of reconfigurable wireless flexible architecture nodes, which are subject to multipath fading, Doppler effects, and adjacent channel interference and have low cost yet robust and reliable operation requirements. We also include findings dealing with bit error rate (BER) versus channel signal to noise ratio (SNR) for BPSK and QPSK modulation schemes, and also the effect of non real-time operating systems (OS) running on GPPs on the transmitter processing time.

An interactive narrative is a story that is shaped by digital technology and that allows the dynamic presentation of scenes or sequences based on input from the user. In this paper we present a new foundation for interactive storytelling that allows a mobile user to interact with a story. The user is placed at the center of the story and the story comes to the user in transit. The behavior and actions of the user influence the scenes and sequences of the story the user experiences. The framework for this system is an ad-hoc network. Ad-hoc networks allow localized presentation of story elements to users who are in transit and able to receive story elements on a mobile device. Furthermore ad-hoc networks have the capability to be context-aware and respond to the physical and social context of the user both on an individual and group level. This mobile context-aware story form is a powerful format in the fields of education and entertainment. It allows the story to be connected with the surrounding environment and it allows the user to see cause and effect of individual and group behavior. In this paper we present an ad-hoc network story system and examine a case study of its use for a prototype mobile context-aware story.

This paper describes an automatic modulation scheme recognition technique. The technique is designed for a real-time software radio using general-purpose processors and is based on modified pattern recognition and signal space approaches. It is robust and efficient with a processing time overhead that still allows the software radio to maintain its real-time operating objectives. Both digital and analogue modulation schemes can be identified. Tests and simulations using an AWGN channel show that the SNR threshold for correct analogue modulation scheme classification is approximately 6.5 dB.

Ad-hoc networking has been of increasing interest in recent years. It encapsulates the ultimate notion of ubiquitous communications with the absence of reliance on any existing network infrastructure. This paper presents a concept for robust operation of multimedia applications over such networks. Error resilient communication is achieved by using a new error detection and concealment technique that exploits information from the decoded image data itself as well as using information from the underlying network. This approach unifies information from both traditional computer science and signal processing domains. A layered architecture framework for the implementation of the proposed system is also described.