In this paper, challenges regarding the provision of channel state information (CSI) on a multi-user (MU) MIMO-OFDM downlink are addressed. Reference symbols (pilots) support coherent detection at the receiver and enable link adaptation at the transmitter, but add overhead and consume transmit power. As spatial transmit processing combined with multiple access essentially requires the provision of dedicated and common pilots, i.e. pilots that include respectively exclude user-specific transmit processing, the potential gains of MU-MIMO may partly be canceled by the incurred pilot overhead. Fortunately, spatial correlation at the transmitter has the potential to substantially reduce overheads, by introducing spatial reuse of dedicated pilots, and by utilizing spatial interpolation of common pilots. We demonstrate that a reduction in overhead is traded with compromised channel estimation accuracy, which particularly limits the attainable spectral efficiency in the high SNR regime. This implies that a bandwidth efficient pilot design is to be complemented by advanced channel estimation schemes, which enhance accuracy by utilizing data symbols together with pilots.

Reference symbols (pilots) support coherent detection, but add overhead and consume transmit power. A sophisticated pilot design needs to trade-off the attainable channel estimation accuracy with the induced losses in bandwidth & power efficiency. MIMO transmission, transmit beamforming, and spatial multiplexing further complicate pilot design. In this paper a unified analytical framework is developed to evaluate the attainable spectral efficiency of different pilot types for MIMO- OFDM employing fixed transmit beamforming.

Conventional fixed spectrum allocation cannot successfully cope with today's scarcity of radio frequency spectrum. Therefore, regulation of spectrum will undergo revolutionary changes in the future, allowing a less restricted and more flexible access to radio spectrum. Intelligent mobile radios, or the so-called cognitive radios, will improve the spectrum utilization by seeking and opportunistically utilizing radio resources in time, frequency and space domains on a real time basis. The cognitive radio technology poses many new technical challenges, and overcoming these issues becomes even more challenging due to non-uniform spectrum and other radio resource allocation policies, economic considerations, the inherent transmission impairments of wireless links and user mobility. Cognitive radio technology advances are based on interdisciplinary research, including among others: signal processing, information theory, computer science, communications engineering, artificial intelligence, game theory and economics. Cognitive radios are self-aware communication systems, which autonomously coordinate the usage of spectrum. They identify unused radio spectrum, the so-called 'spectrum opportunities' or 'white spaces', while observing their radio environment, and follow regulatory rules, which are known or are to be learnt. The classification of spectrum as being unused and the way it is used involves regulation, as this spectrum might be originally assigned to a licensed communication system. Cognitive radios will share spectrum either horizontally with distributed spectrum access such as 'listen-before-talk' and equal rights to access the radio spectrum, or vertically, where the so-called primary radio systems have higher priority to access the radio spectrum than secondary radio systems (i.e. cognitive radio systems). Advanced spectrum management can refer to the implementation of centralized means for coordinating dynamic spectrum access. Such a centralized approach has the advantage of facilitating the regulators' control of spectrum usage, and to allow them directing how the spectrum is used. Such a control is of particular interest in vertical spectrum sharing. Contrary, the decentralized paradigm of cognitive radio allows for a more flexible spectrum access, which is especially helpful in horizontal spectrum sharing scenarios. Joint efforts of academia, regulators, infrastructure suppliers and telecom operators are required to establish trust in cognitive radio technologies and to enable a sustainable go-to-market. The introduction of cognitive radios will be rather a smooth evolution than a radical revolution of communication technology. Standardization, spectrum regulation, confidence in business opportunities and a commercialization of products based on cognitive radio technologies will most likely be incremental on a step-by-step basis. The standardization efforts on cognitive radio, advanced spectrum management and its enablers by the IEEE in SCC41/P1900, 802.11y and 802.22 are an example of this. Cognitive radio technologies as those discussed in this special issue have the potential to considerably change the wireless communication market as barriers for new market entries are essentially lowered. This special issue presents one invited and seven peer-reviewed papers written by experts from academia and industry. In the first (invited) paper 'Dynamic Spectrum Management for Cognitive Radio' S. Haykin an overview of the dynamic spectrum management (DSM) problem in cognitive radio is given. The radio scene is first analysed (Radio Scene Analysis—RSA), on the basis of which the whole of cognitive radio is built. Attributes of the RSA should desirably include: model independence; accurate analysis of the radio scene; robustness of performance; optimality in the presence of noise. Two different approaches for DSM, one being price-based and the other being opportunistic are presented; bearing in mind that this paper is devoted to the DSM problem in a cognitive radio environment, attention is focused entirely on the opportunistic approach. Under the opportunistic approach, mathematical formulation of the DSM problem is discussed from the viewpoint of optimality, thereby setting the stage for a graph-theoretic framework of the problem. Finally centralized versus decentralized approaches to DSM are touched. In the second paper 'CDMA Underlay Network with Cognitive Interference-Minimizing Code Assignment and Semi-Blind Interference Suppression' by A. Elezabi et al., a code-division multiple-access (CDMA) secondary network, which operates cognitively as an underlay for a primary network is presented. A novel interference-minimizing code assignment (IMCA) for secondary users is proposed such that the interference from existing primary users is minimized. This scheme shows significant gains in the error probability performance compared to random code assignment. However, as the exact solution is complicated, the authors obtained a near-optimal solution of complexity O(N) and demonstrated by simulations that its performance were almost identical to those of the exact IMCA scheme. Finally, for the secondary network they propose a slightly modified minimum-output energy receiver that operates semi-blindly as a first stage followed by parallel interference cancellation among secondary users only. For this receiver structure the authors derive improved log-likelihood ratios for the single-user decoders based on the partial knowledge of the user codes and channel coefficients. Significant performance improvements are also obtained due to IMCA in the coded case. The third paper 'Efficient ARQ Protocols with Anti-Jamming Coding for Cognitive Radios' by Guosen Yue et al. studies efficient ARQ protocols in conjunction with two types of anti-jamming coding techniques—rateless coding and piecewise coding—for protecting secondary users' data transmissions in cognitive radio systems. It is shown for both anti-jamming coding schemes that the corresponding simple ARQ protocols provide significant improvement in secondary user throughput. Further, the comparison of these two approaches indicates the superiority of the piecewise coding with its ARQ protocols compared to the rateless counterpart. In the fourth paper 'Spectral efficiency comparison between OFDM/OQAM- and OFDM-based CR networks' by Haijian Zhang et al. a potential candidate technology for cognitive radios referred to as OFDM/Offset Quadrature Amplitude Modulation (OQAM) is introduced and compared to classical OFDM based cognitive radio networks. OFDM/OQAM allows the implementation of an opportunistic spectrum access of secondary users, that is, cognitive radios, to spectrum originally licensed to primary radio services without harmful interference to these licensed systems. A spectral efficiency comparison is performed in simulating results for different multicarrier systems and interpreting these by theoretically analysing the out-of-band radiation of their prototype pulses shaping. In the fifth paper 'Dynamic Spectrum Assignment in Multi-cell OFDMA Networks enabling a Secondary Spectrum Usage' by F. Bernardo et al., a set of novel dynamic spectrum assignment algorithms for multi-cell orthogonal frequency multiple access (OFDMA) scenarios is presented. It is shown that thanks to the proposed algorithms, licensed spectrum holders can release spectrum bands in large geographical areas to be leased to other secondary markets in a cognitive radio environment, while preserving the quality of service of the licensed users in the system. Results are obtained comparing the proposed schemes against other conventional frequency reuse strategies, revealing significant improvements in terms of both spectral efficiency and opportunities for secondary usage. In the sixth paper 'QoS Aware Admission and Power Control for Cognitive Radio Cellular Networks' by J. Xiang et al., a power control scheme for primary and secondary users is proposed, and the admission control problem as an instance of the classical multidimensional knapsack problem is modeled. Three types of solutions, including an exact solution using dynamic programming and two approximate solutions using heuristic algorithms are investigated. Finally, simulation results are presented to demonstrate the performance of the proposed schemes. In the seventh paper 'Performance Evaluation of Advanced Spectrum Functionalities for Future Radio Networks' by M. Bennis et al., novel spectral resource management (SRM) functionalities from the WINNER project are presented. The proposed SRM spectrum architecture uses more efficient resource allocation techniques where flexible spectrum access and usage lead to more capable and faster services with high quality-of-service, giving more user satisfaction than conventional networks. In this context, short-term (ST) spectrum assignment is proposed, where spectral resources are exchanged between WINNER Radio Access Networks (RANs) on a fast basis, leading to a better utilization of the spectrum. Moreover, multi-band scheduler and base station-to-base station communications are presented as key enablers for efficient DSM. Performance gains of ST spectrum assignment are analysed in a multi-network, multi-cell environment based on realistic traffic patterns while taking inter-cell interference into account. The results confirm that spectrum availability and utilization can be considerably enhanced. In the eighth paper 'TV White-Space Prototype' by S.W. Oh et al., a white-space device prototype for detecting vacant spectrum in TV bands is presented. The authors implemented the prototype on a real-time platform and tested it using the test methodologies outlined by the Federal Communications Commission (FCC) and IEEE 802.22, the Working Group on Wireless Regional Area Networks (WRANs). The conducted measurements have shown that the prototype consistently performs within the requirements of the WRAN standard, proving the suitability of the considered spectrum sensing algorithm in practical environments.

Dynamic spectrum allocation is a key technique to face spectrum scarcity and inefficient pre-allocation of resources in an environment of rapidly growing demands for mobile communications. By allowing operators to adaptively exchange transmission resources of their communication systems with each other by way of trading, short-term traffic variations can be addressed based on local actual demand. Concatenating trading areas and coarsening the time scale, pre-allocation of resources can be adapted with respect to long-term estimations of traffic, thereby building a hierarchic market framework for trading transmission resources. Numerical simulations show, that our approach can lead to a significant increase in resource usage.

An efficient scheme for limiting the inherent high out-of-band radiation of the orthogonal frequency-division multiplexing signal has been studied by Brandes et al.. The scheme achieves sidelobe suppression by using few subcarriers, referred to as cancellation carriers (CCs), positioned at the edges of the spectrum and modulated by data-dependent complex-valued symbols that are designed so as to minimize the out-of-band radiation. Unfortunately, its performance has been investigated at the input of the power amplifier (PA) without considering the spectral re-growth induced by PA's non-linear region. Motivated by the above consideration, in this work we return to the aforementioned scheme and investigate its performance at the output of the PA. In addition, since reducing the time-domain signal peaks may alleviate the problem of spectral re-growth, we also adopt two well-known tone reservation (TR)-based peak reduction algorithms. Computer simulations are used to assess the performance of the investigated schemes under different operating conditions. It turns out that the CC-based solution may provide some performance gain at the PA output only for input power backoffs larger than 6 dB. Additional benefits can be achieved when used in conjunction with TR and some linearization of the PA characteristic.

In this paper, we consider the problem of high out-of-band (OOB) radiation in OFDM systems. OOB radiation is mainly caused by high peak-to-average power ratio (PAPR) of OFDM signal that drives high power amplifier (HPA) in nonlinear region. We investigate the OOB reduction capabilities of the most common PAPR reduction algorithms and compare their performance. We determine input back-off of HPA needed to achieve adjacent channel leakage ratio (ACLR) requirement assuming different PAPR reduction algorithms, HPA characteristics and modulation schemes. Also, we consider the factors that contribute to OOB in OFDM and show how to distinguish them.

In this paper synchronization of uncoordinated OFDM spectrum sharing systems that operate in the presence of narrowband interference (NBI) is analyzed. Conventional synchronization algorithms based on correlation of synchronization preamble in time domain suffer from considerable performance degradation when NBI is present. To overcome this degradation we propose and analyze a method to detect NBI presence and to suppress its influence on synchronization before estimation of synchronization parameters is performed at the OFDM receiver. We focus on an OFDM WLAN system operating in license-exempt band and analyze synchronization performance when NBI is active. The proposed method does not introduce any additional preamble overhead with respect to existing WLAN preamble and requires only slightly higher computational complexity at the receiver. To illustrate effectiveness of the proposed method, several numerical examples are provided.

In this paper we analyze the capacity of MIMO-OFDM systems that utilize pilot symbols to acquire estimation of the radio channel at the receiver. An analytical framework is developed to derive the optimal pilot boost and optimal pilot overhead over time-variant frequency selective channels. In addition, to illustrate the achievable capacity, several numerical examples are given assuming different pilot boosts, pilot overheads, and number of transmit antennas.

In this paper a method for sidelobe suppression in orthogonal frequency-division multiplexing (OFDM) spectrum sharing systems is proposed. This technique is denoted additive signal because a complex-valued sequence is added to the original transmit data sequence in order to minimize the sidelobes. To determine additive signal sequence an efficient convex optimization algorithm subject to constraints on transmission power and bit-error rate (BER) performance is employed. Numerical results show that the sidelobes are suppressed more than 10 dB with a moderate loss in BER performance.

In this paper, an analytical framework is established to dimension the pilot grid for OFDM in terms of overhead and power allocation. With the constraint that the receiver solely relies on pilot symbols for channel estimation, the capacity for OFDM over time-variant frequency selective channels is maximized. The authors show that the maximum capacity is achieved by placing pilots with maximum equidistant spacing given by the sampling theorem, if perfect interpolation is assumed and pilots are appropriately boosted. Allowing for realizable and possibly sub-optimum estimators where interpolation is not perfect, a semi-analytical best pilot allocation strategy is presented.

In this paper, we address the problem of channel estimation in the uplink of TDD MC-CDMA with combined-equalization. Channel responses are estimated based on least-squares approach assuming that training blocks are periodically inserted into the uplink data stream. Compared to channel estimation approaches that are designed for conventional uplink MC-CDMA with post-equalization, the proposed scheme provides better performance in terms of mean-square channel estimation error and bit-error rate. Moreover, we investigate the impacts of channel variations and of energy boosting of the training blocks on the system performance

Combined equalization has recently been proposed to enhance the error rate performance of conventional multicarrier code-division multiple-access (MC-CDMA) systems. This technique applies pre-equalization at the transmitter in conjunction with post-equalization at the receiver, thereby splitting the overall equalization process into two separate parts. In this way, efficient power allocation over the available subcarriers is possible at the transmitter, while leaving the interference cancellation task at the receiver. In this paper, we consider the uplink of an MC-CDMA system employing combined equalization. As the users transmit from different locations, the uplink signals arrive at the base station after passing through different multipath channels and the goal is to estimate the pre-equalized channel frequency response of each user. This is pursued following two different approaches. The first operates in the frequency-domain and treats the channel gains over adjacent subcarriers as independent unknown parameters. The second operates in the time-domain and achieves better performance by reducing the number of unknown parameters. Both schemes are based on maximum-likelihood reasoning and require knowledge of the transmitted symbols. Numerical examples are given to highlight the effectiveness of the proposed methods.

In this paper, we consider wireless multicarrier communications systems applying different transmit-antenna diversity techniques. Whereas the transmitter has available multiple antennas, we assume a single antenna at the receiver. Thus, we consider multiple-input single-output multicarrier transmission systems which are of special interest for downlink mobile radio applications. For these systems, we investigate the matched-filter bounds (MFBs) which represent the performance in an environment free of any interference and, thus, are lower bounds on the achievable performance. Especially, we focus on the analytical determination of the MFBs for selection diversity (SD), since this transmit-antenna diversity technique requires the least amount of preprocessing at the transmitter. Based on the derived MFBs, SD is investigated in detail, and compared with other transmit-antenna diversity techniques which are based on the maximum-ratio and the equal-gain transmission principles, respectively

In this paper, a new technique for sidelobe suppression in orthogonal frequency-division multiplexing (OFDM) systems is proposed and investigated. This technique can be applied to both pure OFDM and OFDM based multiple-access systems to improve spectral efficiency. Moreover, it can be utilized within OFDM based overlay systems to avoid interference towards the legacy systems operating in the same frequency band as the overlay system. The proposed sidelobe suppression technique is based on subcarrier weighting. The real-valued subcarrier weights are determined in such a way that the sidelobes of the transmission signal are minimized using an optimization algorithm which is capable to take several optimization constraints into account. Results show that sidelobes can be easily suppressed by more than 10 dB with the proposed technique. These results are achieved by allowing only a moderate loss in bit-error rate performance.

We consider the downlink of a multi-carrier code-division multiple-access (MC-CDMA) system operating in a time division duplexing (TDD) mode and propose a non-linear pre-filtering scheme based on Tomlinson-Harashima pre-coding. In designing the pre-filtering matrices we adopt a minimum mean square error (MMSE) approach under a constraint on the overall transmit power since multiple access interference and multipath distortions are pre-compensated at the base station, low-complex receivers can be employed at the mobile units. Compared to other existing solutions, the proposed scheme provides better error rate performance

In this letter, we propose the concept of combined equalization for uplink multicarrier code-division multiple access (MC-CDMA) and perform a theoretical analysis which shows that better single-user bounds than the classical matched-filter bounds are achieved with this new concept. Moreover, we illustrate how to properly design an uplink MC-CDMA transmitter and receiver for combined equalization, and show by Monte Carlo simulations that the improved single-user bounds are closely approached, even in the case of a fully loaded system.

In this paper, uplink time division duplex (TDD) multi-carrier code-division multiple-access (MC-CDMA) systems applying pre-equalization to combat channel impairments are considered. Especially, the performance degradation caused by the Doppler induced frequency offsets among the different mobile users is analyzed in detail. Based on this analysis different frequency interleaving schemes are proposed and their potential to reduce the performance degradation is investigated. It turns out, that a frequency interleaver which is allowed to skip a few, well-defined subcarriers is capable of reducing the influence of Doppler induced frequency offsets significantly causing a very little loss in bandwidth efficiency.