The possibility of expending high frequency bands, in specific millimeter and Terahertz (THz) bands, for high-speed information transmission beyond 100 Gbps has drawn a big attention in the last years. The extra-wide bandwidth available at THz band has the potential to realize the extremely high data rate needs. Nevertheless, the unique propagation characteristics at THz frequencies have to be first fully investigated. This work presents indoor ultra-wideband (UWB) channel sounding measurements across 60 GHz of bandwidth between 240-300 GHz using a channel sounder system based on Vector Network Analyzer (VNA) with frequency converters. The measurements are performed in Line-of-Sight (LoS) and reflected-non-LoS (RNLoS) transmission conditions with multiple scenarios. By matching the power delay profile (PDP) to the actual geometry, we distinguish between the reflections caused by the environment, and the undesired reflections produced due to the setup geometry. A further analysis of the channel mean excess delay, the root mean square (rms) delay spread and the coherence bandwidth is also provided.

The most common analytical model used in raytracing to estimate the scattered field by a random rough surface is the Kirchhoff approximation, where a normal distribution of the surface heights is assumed. When approximating the non-Gaussian surface roughness with a Gaussian distribution, the high peaks and valleys cause an overestimation of the heights’ standard deviation and yields to inaccurate valuation of the coherent component of specular reflection. In this work, for an accurate scattering modeling of a non-line-of-sight (NLOS) propagation scenario, a curve fitting method is applied to calculate the effective roughness from non-Gaussian surfaces and then implemented in a self-programmed three-dimensional ray-tracer. When considering the third moments of the heights distribution, a deviation of (0 – 3.5) dB is observed in the scattering power depending on the material roughness.

The work in this paper is trying to answer the question: In a full-duplex indoor base station with antenna selection, what are the benefits of using distributed antennas instead of co-located antennas? In other words, how much the use of distributed antennas would contribute in mitigating the self-interference in the full-duplex system. A ray-tracing tool is used to characterize the indoor multi-path channel environment. Subsequently, a baseband analysis is done to evaluate the performance of the full-duplex system. The distributed antenna system slightly outperforms the co-located system in the studied scenario.

Most commonly used frequencies in Synthetic Aperture Radar (SAR) imaging are from hundreds of MHz to tens of GHz. Nowadays, the unexplored mmWave and THz regimes are being investigated for SAR imaging to have higher resolution. Two research challenges for this technology motion compensation and real-time processing are addressed in this paper. SAR image reconstruction algorithms lack data dependencies. Therefore, it can be processed in a parallel computing environment to accelerate the image reconstruction. This paper presents the theoretical model of a distributed signal processing testbed. The performance enhancement with the inclusion of high performance computing for image processing is also explained. Another challenge for this technology comes from a motion compensation point of view. The SAR signal processing assumption is that the radar-carrying platform follows an ideal path. However, in reality, the platform deviates from this path. Especially at THz, very small deviations in the range of sub-mm produce motion errors because of wavelengths in the range of deviations. It degrades the image quality and provides wrong estimations of range and azimuth scatters. The paper summarizes motion error types, their characterization, and constraint on the amplitude of deviation. Also, the paper presents effects of these errors at 100 GHz with simulation results.

In Terahertz region, the diffuse scattering phenomenon by rough surfaces has a remarkable effect of non-line of sight (NLOS) channel propagation. Several parameters contribute to form a specific distribution of the scattered rays, and the angular distribution of the scattered field varies depending on the surface roughness parameters. Yet, a clear understanding of how these parameters adjust the scattered field distribution is still missing. In this work, we investigate the scattering properties of indoor materials at Terahertz band. This is done via full-wave simulation of the diffused scattering by rough samples varies in their roughness characteristics. Then, the role of each statistical parameter is investigated separately. It was proven that both the surface correlation function and the correlation length have an influence on the far field scattered energy. While the height standard deviation has a direct effect on the specular part, the correlation length seems to affect the power distribution out of the specular direction. The longer the correlation length, the more energy is accumulated closer to the specular direction. The resulting angular distribution can be then applied for obtaining more accurate NLOS channel predictions.

State-of-the-art wireless technology is reaching its technical limits and future technologies such as millimeter wave (mmWave), massive MIMO and full duplex transmission included in upcoming 5G standards will be key in fulfilling demands for higher data rate and spectrum efficiency. Massive MIMO systems operating in the mmWave spectrum have many advantages in terms of e.g. beamforming and channel capacity, not only for communication systems but also for radar systems. Therefore, an experimental real time massive MIMO testbed utilizing sub-6GHz and the mmWave spectrum is to be constructed at Institute of Digital Signal Processing (DSV), University of Duisburg-Essen, Germany to investigate this technology in a practical manner for indoor radar system applications. The proposed testbed will initially be implemented for sub-6GHz frequencies with more than hundred antennas and later extended to the mmWave region. Obviously, the testbed design is very complex and a detailed design study is required before implementation. This paper summarizes findings of the design study. Firstly, a theoretical model of a signal processing architecture for a massive MIMO testbed is presented. This is followed by requirements and comparison of parameters processing power, bandwidth, data throughput etc. for sub-6GHz and mmWave spectrum. Also, mapping of algorithms on heterogeneous computational resources, such as FPGA, GPU and CPU is discussed. Based on the study, parameters for the proposed testbed are selected.

This work proposes a practical method that combines wideband antenna selection and cross-polarization to enable full-duplex in an LTE femto base-station. A ray-tracing tool is used to characterize the indoor multi-path channel environment. Subsequently, a baseband analysis is done to evaluate the performance of the full-duplex system. Two antenna selection algorithms are tested: an optimal exhaustive method, and a practical low-complexity approach for real-case scenarios. It is shown that the high complexity, and the additional hardware, required for the optimal antenna-selection may be avoided without big drop in the performance of the full-duplex system.

As recommended by 5th generation Public-Private Partnership (5G-PPP) and Next Generation Mobile Networks (NGMN), one of the most important 5G requirements is to minimize the delay within the network for delay-sensitive services. Main objective of this work is to exploit massive MIMO technology to reduce Hybrid Automatic Repeat Query (HARQ) retransmission delay. Massive MIMO indoor environment is created by a 3D ray-tracing tool to simulate the received power, phase, delay, and angle of arrival of each ray in the channel. Two sizes of rectangular antenna array are used to evaluate the performance of beamforming in enhancing the bit error rate. This reflects directly on the mean and maximum numbers of packet retransmission. The angle between user-equipments is considered to evaluate the effect of inter-user interference on HARQ. This work opens the door to develop smarter resource allocation algorithms in 5G for delay-sensitive services based on the location and the angle of arrival of the users.

Distributed Antenna Systems (DAS) promise to deliver solutions for the future wireless challenges using multiple remote radio heads (RRHs) in a distributed architecture, thereby supporting high frequency re-use, increased coverage, decreasing interference and latency. DAS architecture consists of multiple RRHs connected to a centralized base station via an optical fiber network. This work explains different DAS deployment techniques on a real-time LTE system. It focuses on the implementation details of LTE PHY on a TI (Texas Instruments) DSP and corresponding processing blocks that need to be adapted to support the functionality of DAS. Different DAS use case scenarios of deployment, their advantages and key performance indicators, are analyzed. This work also includes the measurement results of deploying antenna selection scenario to investigate and validate DAS on a real-time LTE system.

In cognitive ratio (CR), spectrum sensing is a crucial technique aiming at exploiting spectrum white spaces. This sensing task is more difficult for multiple channels in wide bands. In this paper, we investigate spectrum sensing tasks for long-term evolution (LTE) networks in large frequency-bands. We propose an effective spectrum sensing scheme to detect LTE signals and classify the cell-identities transmitted in multi-channels in parallel. To prove the concept and validate the performance of the proposed scheme, a secondary LTE transmission in TV bands is modeled in our simulations. At the perspective of realistic scenarios, the identified information could be utilized for some functions such as the co-ordination among secondary networks or initial-cell-search procedures between secondary user-equipments and their serving base-stations in the secondary LTE transmission. The simulation results show that the detection and classification performance of the proposed scheme is close tightly to that in the case of a single channel. The scheme works well in multi-path fading environment with carrier frequency offset (CFO) and is tolerant to noise uncertainty.

In cognitive radio networks, the task of spectrum sensing is required to be reliable at low signal-to-noise ratios (SNRs). Spectral correlation is an effective approach to satisfy the requirement. The algorithms based on statistic spectral correlation profiles are a good method as shown in some previous works. In this paper, we propose an algorithm with maximum ratio combination for the profiles to enhance the method. We construct a formula of statistic test and describe an implementation for our algorithm in practice. Extensive simulations are carried out to verify the performance of algorithms. As a result, the proposed algorithm outperforms the existing algorithms with a neglectful cost of additional complexity.

The reliability of spectrum sensing is a challenging issue in cognitive radio (CR) systems. In this paper, we validate the reliability of two spectrum sensing algorithms for pilot-added OFDM signals: time-domain symbol cross-correlation (TDSC) and periodical peaks of autocorrelation (PPA) with a real system in real environments. To validate, these two algorithms carry out detection function for real signals captured by a test-bed of cognitive Long Term Evolution Advanced (LTE-A) systems. Moreover, we use a transmission between a vector generator and a spectrum analyzer to cross-check with results by the test-bed. The experimental results agree with each other and with simulated results in previous works. Two algorithms work well in real-environments and are insensitive to noise-uncertainty. The results show that PPA algorithms outperform TDSC algorithms by 1 dB - 2.5 dB with the observation durations in experiments. Additionally, PPA algorithms are suitable for short observations. For example, PPA algorithms can work with a 5 ms duration of 8K mode Digital Video Broadcasting Terrestrial (DVB-T) signals, but TDSC algorithms cannot. The results also show the performance of TDSC and PPA algorithms by a test-bed of cognitive LTE-A systems. They give clues to apply suitable algorithms for different operations such as in-band and out-band sensing modes in cognitive cellular systems.

Wideband spectrum sensing is an attractive topic in cognitive radio (CR). Filter-banks are a good approach for wideband sensing with different detection algorithms. A wideband sensing scheme with filter-bank realization was proposed as in [1] for long-term evolution advanced (LTE-A) signals. The scheme is able to detect and classify LTE-A signals in multi-channels in parallel. In this work, we validate the reliability of the scheme in realistic scenarios of mobile communications. Real wideband LTE signals are processed by the scheme with an integrated test-bed for LTE base-stations in emergency situations. The experimental results agree with previous works to approve the reliability of the spectrum sensing scheme. Besides, they lead a successful integration approach of the scheme into LTE base-stations.

Spectrum sensing is a challenging task in cognitive radio. A lot of research work in spectrum sensing aims to improve the performance of algorithms. In this paper, we present our exploitation on periodical peaks of autocorrelation in time-domain for pilot-added OFDM signals. Digital video broadcasting terrestrial (DVB-T) signal is taken as an example for the exploitation. The results of this exploitation lead to an approach to enhanced spectrum sensing including detection and classification. Therefore, we propose two enhanced sensing algorithms for DVB-T signal, namely, periodical peaks of autocorrelation (PPA)- based detection with Neyman Person (NP) solution and with maximum rate combination (MRC), respectively. It is found that PPA-based detection algorithms outperform the previous existing detection algorithms which are based on time-domain symbol cross-correlation (TDSC) by about 1dB to 2.5dB. Moreover, the simulation results show that the proposed algorithms are low sensitivity to noise uncertainty and work well in different environments.

Orthogonal frequency division multiplexing (OFDM) systems are sensitive to RF impairments that should be investigated carefully with an appropriate RF model. In this paper, a simplified RF model is studied, regarding nonlinearity, IQ imbalance, phase noise (PN), carrier frequency offset (CFO), DC offset and noise figure (NF). The simplified RF model has not only low complexity and high flexibility, but also high accuracy by taking into account the coherence between RF impairments.

This paper presents our design of a long-term evolution advanced (LTE-A) system enhanced with cognitive radio (CR) technology. Introductions are given on the key components and functionalities such as spectrum awareness, cognitive engine, location awareness, digital front-end and baseband spectrum shaping. Besides, a research platform implementing the CR-enhanced LTE-A system is presented. According to the goals of our ongoing projects kogLTE and ABSOLUTE, specific considerations on TV white space (TVWS) and disaster relief scenario are addressed in the system design.

In the research on spectrum sensing, many works focus on TV bands, in which primary user signals follow digital video broadcasting-terrestrial (DVB-T) standard. However, DVB-T2 is the next generation of this standard. Some characteristics of the primary signal are changed including the pilot pattern and the cyclic prefix (CP). This issue results that the existing sensing algorithms based on the characteristics of DVB-T are not suitable for DVB-T2 signal, or become complex. In this paper, we propose a new sensing algorithm for DVB-T2 signal based on the first preamble symbol of DVB-T2 frame. We derive analytical forms for the detection. The detector works well at the SNR of -10dB with the false alarm of 0.01, with very short sensing time (0.224ms) and with all configurations of DVB-T2. Moreover, we propose a sensing scheme based on this algorithm for practical cognitive radio systems.

In this paper, we give an introductory presentation on the state of the art of the key technologies in the field of cognitive radio, including spectrum awareness, cognitive engine, location awareness, digital front-end and baseband spectral shaping. The technical feasibility of opportunistically operating the future evolution of long-term evolution (LTE) over cognitive white spaces is discussed. A platform implementing these new concepts is briefly introduced.