In this paper, we study compression and identifica- tion algorithms for the identification systems using polar codes. High dimensional feature vectors representing users are first compressed and th ...
In the multiple-input multiple-output (MIMO) broadcast channel with a restriction to transmit strategies that treat interference as noise (TIN), improper signals in combination with widely linear transceivers can bring performance gains compared to proper signals and linear transceivers. However, it has only been shown that there exist scenarios in which these gains can be observed, and it is not yet understood under which circumstances this is the case. As a step towards solving this problem, we present an analysis at high signal-to-noise ratio (SNR) in this paper, and we show that two cases need to be distinguished. In systems with a sufficiently high number of transmit antennas, employing improper signaling can only enlarge the rate region of pure strategies without time-sharing. By contrast, gains due to improper signaling can occur in overloaded systems even if time-sharing is allowed. In addition to the analytical study at high SNR, we provide numerical simulations that demonstrate the implications of our results at finite SNR.
In multi-user massive MIMO uplink scenarios, noncoherent detection schemes are an attractive low-complexity alternative to channel-estimation-based equalization. Since phase information is not available at the receiver, the spatial separation of the induced power is used to resolve the different users. In case of no spatial separation, noncoherent detection is not possible. In this paper, we employ antenna arrays where an electromagnetic lens is placed in front of the antenna elements to focus the induced power to a smaller footprint. This results in a spatial separation required for noncoherent detection. The performance of the EM-lens-enabled system is assessed by means of numerical simulations.
Future autonomous systems require wireless connectivity able to support extremely stringent requirements on both latency and reliability. In this paper, we leverage recent developments in the field of finite-blocklength information theory to illustrate how to optimally design wireless systems in the presence of such stringent constraints. Focusing on a multi-antenna Rayleigh block-fading channel, we obtain bounds on the maximum number of bits that can be transmitted within given bandwidth, latency, and reliability constraints, using an orthogonal frequency-division multiplexing system similar to LTE. These bounds unveil the fundamental interplay between latency, bandwidth, rate, and reliability. Furthermore, they suggest how to optimally use the available spatial and frequency diversity. Finally, we use our bounds to benchmark the performance of an actual coding scheme involving the transmission of short packets.
A superposition modulation method, able to increase the peak data rate of intensity-modulation/direct-detection communication systems by the number of individually switchable light sources, is presented. This is possible by systematically delaying the binary-modulated light sources such that the sum signal is separable at the receiver. In this way the data rate can be increased by the number of available light sources.
The distribution theory serves as an important theoretical foundation for some approaches arose from the engineering intuition. Particular examples are approaches based on the delta-"function". We show that for the Shannon sampling/ interpolation series (SSS/SIS) of continuous signals "vanishing" at infinity, the classical notion of convergence given in complex analysis is equivalent with the modern notion given by the distribution theory, in the sense that the SSS converges at a point on the real line, different from the sampling/interpolation point, if and only if it converges distributionally. This result is in spirit of Weyl's Lemma on the Laplace equation. As an extension, we give those results also for the sampling/interpolation series based on the sine-type function.
The main purpose of this paper is to make the study of spatially coupled turbo-like codes (SC-TCs) more complete by investigating the impact of spatial coupling on the thresholds of hybrid concatenated codes (HCCs). In our previous studies, we introduced some classes of SC-TCs and considered their density evolution (DE) analysis. The obtained results indicated that for a fixed coupling memory, braided convolutional codes (BCCs) have the best belief propagation (BP) thresholds among the considered classes. Besides having excellent BP thresholds, BCCs have good distance properties and their minimum distance grows linearly with block length. Similarities between BCCs and HCCs make HCCs good competitors for BCCs. This has motivated us to investigate the impact of spatial coupling on HCCs. In this paper, we introduce two spatially coupled ensembles of HCCs, referred to as Type-I SC-HCCs and Type-II SC-HCCs. Then, we derive the exact density evolution (DE) equations for the uncoupled and the coupled ensembles for the binary erasure channel (BEC). Finally, considering different component encoders, we compute the thresholds of the SC-HCC ensembles and compare them with the thresholds of BCCs for a range of different rates.
In classical biometric authentication systems, the authenticity of each user is determined by comparing two secret-keys which are generated based on the user's biometric enrollment and authentication sequences. These sequences are generated based on the correlated components of a common source whose statistics are assumed to be known. In the compound biometric authentication model in contrast, the statistics of the source are unknown. In this case, the protocol should guarantee the security and reliability of authentication procedure, simultaneously for all realizations of the compound source. In compound biometric authentication setups which have been investigated until now, only the weak secrecy criterion has been guaranteed. In this work, a single-letter secret-key rate versus privacy leakage rate capacity region of the compound biometric authentication system is given for the case where strong secrecy is also guaranteed. It is shown that the capacity regions under both strong and weak secrecy conditions are equivalent.
A syndrome decoding algorithm for lifted interleaved Gabidulin codes of order L is proposed. The algorithm corrects L times more deviations (packet insertions) than known syndrome decoding methods with probability at least \(1-8q^{-n}\), where n is the length of the (interleaved) Gabidulin code. For \(n<L\), the proposed scheme has L times less computational complexity than known interpolation-factorization based decoders which attain the same decoding region. Upper bounds on the decoding failure probability are derived. Up to our knowledge this is the first syndrome-based scheme for interleaved subspace codes that can correct deviations beyond the unique decoding radius.
The design of low-density parity-check (LDPC) code ensembles optimized for a finite number of decoder iterations is investigated.Our approach employs EXIT chart analysis and differential evolution to design such ensembles for the binary erasure channel and additive white Gaussian noise channel.The error rates of codes optimized for various numbers of decoder iterations are compared and it is seen that in the cases considered, the best performance for a given number of decoder iterations is achieved by codes which are optimized for this particular number.The design of generalized LDPC (GLDPC) codes is also considered, showing that these structures can offer better performance than LDPC codes for low-iteration-number designs.Finally, it is illustrated that LDPC codes which are optimized for a small number of iterations exhibit significant deviations in terms of degree distribution and weight enumerators with respect to LDPC codes returned by more conventional design tools.
This paper analyzes the performance of maximum ratio transmission (MRT) precoding in a massive MIMO interference channel operating in a time-division-duplex mode, where receivers have no channel state information (CSI) except for the average of their effective channel gains. The CSI error vector, inter-user interference vector resulting from MRT precoding, and Gaussian noise vector represent the total noise vector at each receiver whose covariance matrix can be computed. By noting that the total noise vector is neither Gaussian nor uncorrelated with the signal of interest, lower bounds can be obtained based on a worst-case uncorrelated Gaussian noise argument. The bounds hold for an arbitrary number of receivers and receive antennas with full-spatial multiplexing. Simulation results verify the validity of the obtained bounds.
Codes over quotient rings of Lipschitz integers have recently attracted some attention. This work investigates the performance of Lipschitz integer constellations for transmission over the AWGN channel by means of the constellation figure of merit. A construction of sets of Lipschitz integers is presented that leads to a better constellation figure of merit compared to ordinary Lipschitz integer constellations. In particular, it is demonstrated that the concept of set partitioning can be applied to quotient rings of Lipschitz integers where the number of elements is not a prime number. It is shown that it is always possible to partition such quotient rings into additive subgroups in a manner that the minimum Euclidean distance of each subgroup is strictly larger than in the original set. The resulting signal constellations have a better performance for transmission over an additive white Gaussian noise channel compared to Gaussian integer constellations and to ordinary Lipschitz integer constellations.
We assess the power saving potential of cooperative communication for a two-dimensional user distribution scenario. In particular, we compare the average total transmit power of cooperative transmission (i.e., the sum of the user and the relay transmit power, averaged over the distribution of the users) with the average transmit power of conventional (direct) communication. The comparison is performed on the basis of achieving the same signal-to-noise ratio (SNR) at the base station in both cases. In this regard, a circular shaped radio cell with circularly uniformly distributed users is considered. A theoretical investigation of the savings in transmit power offered by cooperation is conducted for the case where the path-loss exponent equals four. The theoretical analysis is verified by simulations, which also consider a wider range of path loss exponent values for the evaluation of the power savings achieved by cooperation. Our results demonstrate that the power savings per user increase with increasing path loss exponent and with increasing number of users within the circular cell.
Filter bank multicarrier transmission with offset-QAM (FBMC/OQAM) is a promising candidate waveform for the next mobile communication systems as it is well suited for many new scenarios and challenges like improved spectral efficiency, spectrum sharing approaches or high mobility scenarios. It does not require a cyclic prefix (CP) leading to a higher spectral efficiency than orthogonal frequency division multiplexing with CP (CP-OFDM) and the flexibility of the transmit and receive filters enables higher throughput in spectral sharing and high mobility scenarios. One aspect to be considered is efficient channel estimation that is needed in order to realize these gains. As the classical channel estimation used for CP-OFDM cannot be applied directly to FBMC/OQAM, new competitive solutions are needed. One promising solution for the pilot design suited for FBMC introduces an auxiliary pilot (precoding symbol) that nullifies the intrinsic interference at the pilot position, but this leads to increased power on these auxiliary pilots. In this paper, a new non-linear scattered pilot design and channel estimation technique is proposed that addresses the problem of increased power of the auxiliary pilot. In addition to the reduced transmit power, we also observe that the receive power at the pilot position is increased with this new proposal which leads to an improved channel estimation performance.
Differential linear network coding (DLNC) is a precoding scheme for information transmission over random linear networks. By using differential encoding and decoding, the conventional approach of lifting, required for inherent channel sounding, can be omitted and in turn higher transmission rates are supported. However, the scheme is sensitive to variations in the network topology. In this paper, we derive an extended DLNC channel model which includes slow network changes. Based on this, we propose and analyze a suitable channel coding scheme matched to the situation at hand using rank-metric convolutional codes.
Physical Unclonable Functions evaluate manufacturing variations to generate secure cryptographic keys for embedded systems without secure key storage. It is explained how methods from coding theory are applied in order to ensure reliable key reproduction. We show how better results can be obtained using code classes and decoding principles not used for this scenario before. These methods are exemplified by specific code constructions which improve existing codes with respect to error probability, decoding complexity and codeword length.
An ultra-wideband system design is presented which supports wireless internet access and similar short-range applications with data rates of the order of 100 Gbps. Unlike concurrent work exploring the 60 GHz regime and beyond for this purpose, our focus is on the 6.0-8.5 GHz frequency band. Hence, a bandwidth efficiency of about 50 bps/Hz is necessary. This sophisticated goal is targeted by employing two key enabling techniques: massive MIMO communications in conjunction with multi-mode antennas. This concept is suitable both for small-scale terminals like smartphones, as well as for powerful access points. Compared to millimeter wave and THz band communications, the 6.0-8.5 GHz frequency band offers more robustness in NLOS scenarios and is more mature with respect to system components.
The paper addresses a comparison between binary and multilevel continuous phase modulation (CPM) with a serially concatenated convolutional code. In order to achieve a high spectral efficiency, the binary approach uses an increased baud rate and a small modulation index. Such signals are currently considered for standardization as narrowband tactical VHF waveforms. These waveforms provide user bit rates from 10 kbit/s to 82.285 kbit/s on a 25 kHz channel. As an alternative, multilevel CPM signals utilizing 4 up to 16 levels are proposed in this paper, allowing for baud rates approximately equivalent to the channel bandwidth. In order to take into account the harsh environment for tactical radio systems, resulting in increased phase noise and oscillator instabilities, noncoherent detection is used. From a power efficiency point of view, both binary and multilevel signals perform similar and close to coherent detection on a Gaussian channel. In presence of adjacent channel interference, the proposed multilevel signals perform clearly better than the binary signals. Especially the binary signals providing high bit rates are degraded drastically. We show that the receiver complexity of the proposed multilevel schemes is affordable with modern FPGAs.
In a dense small-cell (SC) network with several users to be served, a multi-user detection (MUD) can be employed across SCs, and distributed estimation is a promising technique for such a scenario. Nevertheless, large communication overhead due to frequently exchange of variables among SCs will cause high energy consumption and processing latency. This paper is focused on the reduction of communication overhead for the distributed processing. To this end, two algorithms, Augmented Lagrangian based Cooperative Estimation (ALCE) and Priorityaided ALCE (PALCE) will be presented. In ALCE a new efficient approach is adopted to achieve parallel processing among all SCs, which needs fewer variables to be exchanged. Thus, a considerable amount of overhead will be saved. However, the ALCE algorithm is not robust when applied to a network with erroneous backhaul (BH) links, therefore a variant of this approach termed PALCE is proposed using a priority oriented principle to enhance the robustness and maintain low amount of information exchange. The proposed algorithms are investigated by means of error rate and communication overhead showing significant improvement in estimation performance compared to state of the art algorithms.
A new approach for the analysis of cyclostationary noise is presented based on the theory of cyclic autocorrelation and on the cyclic spectrum. Usually, the cyclic spectrum is mainly applied in the field of spectrum sensing for the analysis of cyclostationary signals such as common digitally modulated signals as phase shift keying (PSK) or quadrature amplitude modulation (QAM). However, it is stated in this paper that the cyclic spectrum can also be employed to analyze cyclostationary noise in various scenarios. It turns out to be superior to conventional means. This is made clear showing its ability to separate two cyclostationary noise sources with different cycle frequencies. Furthermore, the analysis of amplitude noise and phase noise is discussed. Because of the results, this new approach for the examination of various kinds of cyclostationary noise processes is considered to be quite promising.
