In OFDM transmission system, phase noise (PN) caused by non-ideal oscillator induces inter carrier interference (ICI), which is problematic for operation of high order modulation, especially in mm-wave frequency bands. This paper develops two low-complexity and effective ICI compensation algorithms: (i) Direct de-ICI filtering approach which is agnostic to the exact locations of phase tracking reference signal (PTRS), and hence it can readily operate with existing 3GPP Rel-15 PTRS and (ii) Circulant ICI filter approximation which exploits the circulant matrix constructed by a PTRS sequence with cyclic structure to reduce the complexity of ICI compensation algorithm. Our results reveal that both algorithms significantly reduce the negative impact of phase noise on the link performance.

We study a multihop “virtual” full-duplex relay channel as a special case of a general multiple multicast relay network. For such a channel, quantize-map-and-forward (QMF) [and its generalization of noisy network coding (NNC) and short message NNC] achieves the cut-set upper bound within a constant additive gap, where the gap grows linearly with the number of relay stages K. This gap, however, may not be acceptable for practical communication systems with multihop transmissions (e.g., a wireless backhaul operating at high frequencies). Recently, we improved the capacity scaling by using a forward sliding-window (SW) decoding and by optimizing the quantization level at each relay, obtaining the gap that grows logarithmically as log K. Furthermore, the improved scheme has lower decoding complexity and delay than the general QMF and NNC approaches. In this paper, we further improve the performance by presenting a mixed scheme in which each relay can perform either decode-and-forward (DF) or the improved QMF (with SW decoding) and can choose to perform rate-splitting to enable partial interference cancellation. In general, the optimization of the relay DF/QMF configuration is combinatorial. Nevertheless, we provide that a simple greedy algorithm finds an optimal configuration under some practically reasonable assumptions. We derive an achievable rate that is easily computable and show that the proposed mixed scheme outperforms the QMF-only schemes. We demonstrate that the performance improvement increases with K, which indicates that the mixed scheme is indeed beneficial for multihop transmission.

A method of distributing CRC bits from a single CRC polynomial over a block of information bits for Polar codes via an interleaver to facilitate early termination and enhanced error correction is proposed and investigated. The interleaver is designed such that each CRC bit is placed in a decoding position as early as possible while keeping all the information bits that the CRC bit depends on in front of it in the successive decoding order. This allows the CRC bits to be used as dynamic frozen bits to assist surviving path selection during SC/SCL decoding to improve error performance. Furthermore, it allows the distributed CRC bits to be used for early error detection in the list decoding process and thus reduces the average decoding latency. Simulation results show that interleaved CRC can provide moderate benefits in improving error correction and reducing decoding latency through early error detection.

A method of constructing rate-compatible polar codes that are capacity achieving at multiple code rates with low-complexity sequential decoders is presented. The underlying idea of the construction exploits certain common characteristics of polar codes that are optimized for a sequence of successively degraded channels. The proposed code consists of parallel concatenation of multiple polar codes with information-bit divider at the input of each polar encoder. Thus, it is referred to as parallel concatenated polar (PCP) codes. A lower-rate PCP code is simply constructed by adding more constituent polar codes, which enables incremental retransmissions at different rates in order to adapt to channel conditions. Due to the length limitation of polar codes, the PCP code can only support a restricted set of rates that is characterized by the size of the kernel when conventional polar codes are used. To overcome this limitation, punctured polar codes, which provide more flexibility on blocklength by controlling a puncturing fraction, are considered as constituent codes. The existence of capacity-achieving punctured polar codes for any given puncturing fraction is proven. Using such punctured polar codes as constituent codes, it is shown that the proposed PCP code is capacity achieving for an arbitrary sequence of rates and for any class of degraded channels.

We present a method of constructing rate-compatible polar codes that are capacity-achieving with low-complexity sequential decoders. The proposed code construction allows for incremental retransmissions at different rates in order to adapt to channel conditions. The main idea of the construction exploits certain common characteristics of polar codes that are optimized for a sequence of degraded channels. The proposed approach allows for an optimized polar code to be used at every transmission thereby achieving capacity. Due to the length limitation of conventional polar codes, the proposed construction can only support a restricted set of rates that is characterized by the size of the kernel when conventional polar codes are used. We thus consider punctured polar codes which provide more flexibility on block length by controlling a puncturing fraction. We show the existence of capacity-achieving punctured polar codes for any given puncturing fraction. Using punctured polar codes as constituent codes, we show that the proposed rate-compatible polar code is capacity-achieving for an arbitrary sequence of rates and for any class of degraded channels.

In this paper, we present a cooperative relaying strategy for half-duplex multihop relay networks. This scheme consists of three parts: 1) relay selection to yield a layered relay network; 2) group successive relaying that establishes a relay schedule to efficiently exploit half-duplex relays; and 3) a cooperative relaying scheme named short message noisy network coding with sliding-window decoding (SNNC-SW) that outperforms other state-of-the-art information theoretical schemes with lower decoding complexity and delay. We derive an achievable rate region of the proposed SNNC-SW scheme and attain a closed-form rate expression in the asymptotic case for several network models of interests. We then focus on the first part of our relaying strategy regarding efficient relay selection. We develop interference-harnessing routing that exploits the fact that in SNNC-SW, interference is treated as a useful signal. We show that, due to the efficient treatment of interference, this scheme can outperform routing schemes that deploy store-and-forward, a solution previously proposed for practical wireless multihop networks. Finally, we develop a low-complexity successive decoder of our scheme (implemented by a conventional MIMO decoder), which is a solution that can readily be implemented in practice. It is shown that also this practical scheme provides a significant gain over routing (based on store-and-forward) and the performance gap increases as the network becomes denser.

In this paper we present a novel transmission scheme for wireless multihop backhaul networks. The scheme consists of group successive relaying that efficiently exploits half-duplex relays and a coding scheme that improves quantize-map-and- forward (QMF). We derive an achievable rate region of the proposed scheme and attain a closed-form expression in the asymptotic case for several network models of interests. It is shown that the proposed scheme outperforms the multihop routing, which is a solution currently proposed for wireless multihop backhaul networks. Furthermore, the performance gap increases as a network becomes denser. Based on the proposed scheme, we present energy-efficient routing referred to as energy- harvesting in which each node requires a lower transmission power to achieve a desired performance compared to other schemes.

We study a multihop “virtual” full-duplex relay channel as a special case of a general multiple multicast relay network. For such channel, quantize-map-and-forward (QMF) (or noisy network coding (NNC)) achieves the cut-set upper bound within a constant gap where the gap grows linearly with the number of relay stages K. However, this gap may not be negligible for the systems with multihop transmissions (i.e., a wireless backhaul operating at higher frequencies). We have recently attained an improved result to the capacity scaling where the gap grows logarithmically as logK, by using an optimal quantization at relays and by exploiting relays' messages (decoded in the previous time slot) as side-information. In this paper, we further improve the performance of this network by presenting a mixed scheme where each relay can perform either decode-and-forward (DF) or QMF with possibly rate-splitting. We derive an achievable rate and show that the proposed scheme outperforms the optimized QMF. Furthermore, we demonstrate that this performance improvement increases with K.

We introduce a multihop “virtual” full-duplex relay channel as a special case of a general multiple multicast relay network. For such network, quantize-map-and-forward (QMF) (or noisy network coding (NNC)) can achieve the cut-set upper bound within a constant gap where the gap grows linearly with the number of relay stages K. However, this gap may not be negligible for the systems with multihop transmissions (e.g., a power-limited wireless backhaul system operating at high frequencies). In this paper, we obtain an improved result to the capacity scaling where the gap grows logarithmically as log (K). This is achieved by using an optimal quantization at relays and by exploiting relays' messages (decoded in the previous time slot) as side-information at the destination. We further improve the performance of this network by presenting a mixed strategy where each relay can perform either decode-and-forward (DF) or QMF with possibly rate-splitting.

To cope with the exponential growth in mobile data traffic, substantially denser deployments of base stations are anticipated in the future. The feasibility of such dense deployments is predicated on the existence of a high-data-rate transport network. Wireless self-backhaul, where the same radio spectrum is used for both access and transport, is an attractive solution due to its potentials of substantially reducing the deployment cost. With the limited range of radio signals in the high frequency bands where some future wireless systems are likely to be deployed, a wireless route of multiple hops may need to be established between each access node and an aggregation node where data traffic originates and terminates. To maximize the throughput of each of these routes, the route selection process should take into account the mutual interference among wireless links. In this paper, we investigate the feasibility of the wireless self-backhaul solution for an ultra-dense wireless network in an indoor environment with inter-node distance in the order of ten meters or less. A joint design approach is adopted here where the routes for all active access nodes, along with the radio resources associated with each link of these routes, are simultaneously selected. Simulation results based on a ray-tracing indoor channel model indicate that multi-hop self-backhaul is a viable backhaul solution. However, compared to wired backhaul, self-backhaul incurs significant limitation on total system throughput, especially at high load.

A frequently encountered problem in a cellular network is the selection for each cell in the network a set of parameter values that is uniquely distinguishable from or compatible with those selected by its neighboring cells. Examples include reuse frequency, reference signal offset, beamforming weights and transmit power, just to name a few. It may also be advantageous for the parameter selection to be carried out in a distributed manner so that no centralized controller is required. In this paper, we study a general framework for coordinating parameters in a distributed manner with the goal of applying such a framework to solving most of the above-mentioned problems that have well-defined cost functions. The main components of the framework are the exchange of the impact functions among neighboring nodes and the parameter update based on the weighted sum of these impact functions. The impact function is essentially an indication to the recipient of the information the impact of the recipient's parameter setting on the cost function of the node that transmits the information. Simulations using frequency reuse planning as an example show good convergence behavior and performance comparing with ideal assignment.

Coordinated multi-point (CoMP) transmission is a general method for mitigating the inter-cell interference in a cellular network through the use of multi-user precoding across neighboring base stations. In this paper, we investigate the system performance of MMSE-based multi-user precoding in a cellular network with CoMP transmission. An iterative algorithm that exploits the duality between the downlink broadcast channel and the uplink multi-access channel is used to compute the multi-user precoding matrices. The algorithm uses the knowledge of the channel error covariance during the precoder computation to alleviate the impact of channel uncertainty at the network. We simulate the system performance of various MMSE-based precoders under the CoMP setting using a realistic MIMO channel model and compare them with those of the zero-forcing precoders. We show that the error-compensated MMSE precoder is more robust towards channel uncertainty and can significantly outperform zero-forcing precoding at high loads.

Coordinated multi-point (CoMP) transmission has recently received much attention due to its promise in effectively mitigating the inter-cell interference in a cellular network. In this paper, we compare two different architectures for CoMP transmission. The first architecture partitions all basestations in the network into disjoint clusters with signal transmissions within each cluster being coordinated independently of one another at a centralized location of each cluster. In the second architecture, the processing for coordinated transmission is distributed among overlapping clusters, each centered around one cell. For this architecture, we propose a simple method for computing linear multi-user precoders for all clusters in a distributed manner, where the precoders are chosen not only to maximize throughput to the served users but also to reduce interference with other users in overlapping clusters. A low-complexity iterative algorithm is proposed for negotiating transmit power levels among adjacent clusters to ensure the power constraints at all transmission points are satisfied. Through system simulations using practical channel models, we show that the distributed architecture substantially outperforms the cluster-wise centralized architecture in maximizing the user data throughputs in the network.

This paper addresses the channel state information (CSI) feedback problem for downlink transmission in a distributed antenna system (DAS). One feedback-and-transmission scheme is proposed, where channel response ratios of the nondominant paths (e.g. paths other than that from the nearest site) to the dominant path (e.g. that from the nearest site) are quantized and fed back, from each user, to the coordination center. Vector quantization with log squared error as a distortion measure is introduced to design codebook for quantizing these ratios. Zero-forcing transmitter or zero-forcing dirty-paper-coding transmitter can be designed based on the feedback information and be used for downlink joint transmission. Performance of the proposed scheme in a system with three single-antenna sites and three single-antenna users is evaluated. Simulation results indicate that to achieve the same performance, the proposed scheme requires feedback amount less than half of that needed with a straightforward scheme, where all channel elements are fed back directly.

In this paper, we explore a method for interference suppression in EGPRS (also commonly known as EDGE) systems, wherein 8PSK is used as the modulation for the desired signal and the interferer is modulated using GMSK. Previous methods for single antenna interference cancellation (SAIC) have concentrated on the case when both the desired and interfering signals are GMSK modulated, or have relied on highly complex methods such as joint demodulation. We derive a simple method that exploits the redundancy inherent in the interferer signal to suppress it. The interference suppression method causes a well- modeled nonlinear distortion to the 8PSK signal. This non-linear distortion is handled using a modified equalizer that deals with a time-varying input symbol constellation applied to a matrix channel. Simulation results using the method show great promise.

User-specific, linear precoding is used extensively by almost all existing and emerging wireless MIMO standards [1], [2], [3], [4]. With user-specific, linear precoding, the data symbols to be transmitted to each user are passed through a linear transformation before being sent to the transmit antennas, and a different precoder is used for each user depending on his/her channel. For example, with 4 transmit antennas at the BS, 2 receive antennas at the mobile, and instant channel quality indicator (CQI), the cell capacity of a 4times2 2-clustered system with user-specific, linear precoding is more than 80% higher than the cell capacity of a system with one transmit antenna [10]. But such schemes are vulnerable to delayed CQI due to fast variations of interference, leading in some cases to performance that is worse than a SISO system. In this paper, we will present a method that mitigates the degradation due to fast-varying interference in MIMO systems that use linear precoding. We will show that the temporal variation of other-cell interference is almost eliminated with this method. Our results indicate that for a 4times2 2-Clustered transmit array with the proposed method, the performance loss due to delayed CQI is reduced from 35% to 5%.

For a cellular system with a fixed number of transmit antennas at each base station, we investigate how the system-level performance varies as a function of the geometry of the transmit array used at each base station. Our results show that the system-level performance of a cellular system can be significantly improved by using transmit antenna geometries other than the uniform linear arrays that have been investigated extensively in the literature so far. For example with 4 transmit antennas at each base station, we show that the spectral efficiency for achieving a 5 percentile user data rate of 2 Mbps (in a 5 MHz bandwidth) is improved by 58% when a non-uniform, linear array is used instead of a traditional uniform linear array.

In an Orthogonal Frequency Division Multiplexing (OFDM) system, the channel is often modeled as a two-dimensional zero-mean Gaussian random process in time and frequency with known second order statistics. With pilot symbols placed across the time-frequency plane, the channel's response at a particular time-frequency point can be derived from its correlation with the nearby pilot locations. However, this a priori channel statistics may not always be available or the channel may not even be stationary to be adequately characterized by a correlation function. In this paper, we introduce a model of time-varying channel based on its delay-Doppler response. We show that a widely adopted time-frequency correlation model, upon which popular channel estimators for OFDM are often based, can be viewed as a special case of such a model with separable statistical profiles imposed on the delay-Doppler plane. Without using any statistical assumption on the channel, we derive a classical estimator based on the two-dimensional delay- Doppler correlator, which can be implemented using Discrete Fourier Transforms (DFT) of the pilot symbol measurements. Simulation results show promising performance even when the pilot symbols' insertion rate is at the channel's maximum delay-Doppler spread.

In this paper, we present an effective method for suppressing spatially and temporally correlated interference. The method treats adjacent temporal samples of the received signal as signal branches coming from additional antennas and uses conventional methods for cancelling spatially correlated interference to decorrelate blocks of spatio-temporal signal samples. Based on a maximum-likelihood formulation, two algorithms for jointly estimating the channel response, the synchronization position, and the spatio-temporal block decorrelation matrix are derived. A multiple-input-single-output prefilter is used to combine the multiple signal branches generated by the proposed method into a single branch so that conventional single-branch equalizers can be employed. In the case of single-antenna interference cancellation (SAIC) in GSM terminals, the proposed technique achieves substantial performance gains over conventional interference-rejection techniques in various interference scenarios.

For a cellular system with a fixed number of transmit antennas at each base station, we investigate how the data rate on the downlink varies as a joint function of the geometry of the transmit array and the linear precoding that precedes the transmit array. The maximum average data rate that can be reliably transmitted over the downlink channel (i.e. the ergodic capacity of the downlink channel) is used in this paper to compare various configurations of transmit antennas at the base stations. Using the 3GPP Spatial Channel Models [4] for a macro-cellular environment, we find the optimum configuration of transmit antennas with a total of 2 or a total of 4 transmit antennas at various SNRs. We offer guidelines for choosing the best transmit antenna geometry for different SNR regions.

A method of estimating DC offset based solely on the received signal, without using any training information, is presented. This method takes into account the non-Gaussian statistical distribution of the modulated desired signal in order to produce a DC estimate that is more accurate than the conventional sample average of the received signal. Specifically, two iterative algorithms are derived based on the generalized expectation-maximization algorithms for jointly estimating the DC offset, the noise variance, and the parameters that characterize the signal distribution. As an application, we consider the special case where the modulated signal has a constant-envelope characteristic. Using the GSM/EDGE air interface, we demonstrate that the proposed method yields significant reduction in the DC-offset estimation error, which translates into performance gains in receiver sensitivity when used in direct-conversion receivers.

Traditional designs of mobile communication receivers commonly rely on a (colored or white) Gaussian model of the signal disturbance. Such a model does not accurately reflect the statistics of co-channel interference, a dominant source of signal disturbance that limits the capacity of most cellular networks. Consequently, the performance of receivers designed based on the Gaussian model is often far from optimum in a heavily loaded network. In this paper, we explore the potential of suppressing interference in channel estimation by using a non-Gaussian noise model of the signal disturbance. An EM-based, iterative algorithm is derived for jointly estimating the channel response and the parameters that characterize the non-Gaussian noise model. As an example, we also consider a special case where the non-Gaussian model has a constant envelope characteristic. Using the GSM/EDGE air interface, we demonstrate that significant performance gains can be obtained with the proposed channel estimation algorithm when a dominant GMSK-modulated interference exists. The proposed algorithm can also be used in combination with equalizers designed for non-Gaussian noise models to provide further performance gains.

In this paper, reduced state sequence estimation (RSSE) for EDGE system is described. A significant issue associated with soft information generation for the RSSE receiver is discussed in detail. Several practical algorithms that calculate soft values efficiently are given. The performance of these algorithms is evaluated through computer simulation and compared with the performance of a more complex decision feedback sequence estimation (DFSE) receiver.

A receiver that adapts to the second-order statistics of the disturbance (interference or noise) is presented. By adapting to the statistics of the disturbance, this receiver produces the maximum-likelihood (ML) estimate of the transmitted sequence for any type of disturbance, in contrast to a conventional receiver that only produces the ML estimate when the discrete-time disturbance is white. This disturbance signal could result from (1) cochannel interference, (2) adjacent-channel interference or (3) thermal noise. The proposed receiver consists of a Nyquist analog receive filter and an adaptive maximum-likelihood sequence estimator for colored noise. We show that the samples of the output of this analog receive filter, at one sample per symbol, are sufficient statistics for finding the most likely transmitted sequence regardless of the type of disturbance present at the input to the receiver. An important feature of this adaptive receiver is that the analog part of the this receiver is fixed, i.e. adaption to the statistics of the disturbance is accomplished by processing in the digital domain only (at one sample per symbol). Using the EGPRS air interface as an example, we show that this adaptive receiver performs substantially better than a conventional non-adaptive receiver: (1) cochannel dominated case: 4.0 dB of gain, (2) adjacent-channel dominated case: 3.5 dB of gain, (3) thermal-noise dominated case: 1.2 dB of gain.

Lindskog and Paulraj (see ICC-2000, June, 2000) proposed a transmit diversity scheme for dispersive channels which by utilizing two transmit antennas and only one receive antenna achieves the same diversity benefit (except for the 3 dB array gain) as can be achieved by using one transmit antenna and two receive antennas. This scheme generalizes Alamouti's (see IEEE JSAC, vol.16, no.8, p.1451-58, 1998) transmit diversity scheme (which works best for flat channels) to dispersive channels. We show how the transmit diversity (TXD) scheme of Lindskog et al. can be deployed in GPRS/FGPRS networks. A low-complexity receiver structure for demodulating Lindskog and Paulraj (LP) encoded signals is presented based on a combination of a novel pre-filter and a delayed decision feedback sequence estimation (DFSE) equalizer. This pre-filter is designed to simultaneously achieve three desired effects: (1) the effective baseband channel seen by the equalizer has half the number of taps of the original baseband channel; (2) the effective response of the baseband channel becomes minimum-phase; and (3) the effective baseband noise is whitened. An information-theoretic analysis is also presented to demonstrate the advantage of the LP scheme over other TXD schemes (e.g. over delay diversity). A burst format for transmitting LP encoded signals which is compatible with the existing GPRS/EGPRS burst format is also presented. Finally, simulations results are presented to verify the benefit of using the LP scheme in the typical urban channel of GSM/EDGE.

Studies the asymptotic characteristics of uniform scalar quantizers that are optimal with respect to mean-squared error (MSE). When a symmetric source density with infinite support is sufficiently well behaved, the optimal step size /spl Delta//sub N/ for symmetric uniform scalar quantization decreases as 2/spl sigma/N/sup -1/V~/sup -1/(1/6N/sup 2/), where N is the number of quantization levels, /spl sigma//sup 2/ is the source variance and V~/sup -1/(/spl middot/) is the inverse of V~(y)=y/sup -1/ /spl int//sub y//sup /spl infin// P(/spl sigma//sup -1/X>x) dx. Equivalently, the optimal support length N/spl Delta//sub N/ increases as 2/spl sigma/V~/sup -1/(1/6N/sup 2/). Granular distortion is asymptotically well approximated by /spl Delta//sub N//sup 2//12, and the ratio of overload to granular distortion converges to a function of the limit /spl tau//spl equiv/lim/sub y/spl rarr//spl infin//y/sup -1/E[X|X>y], provided, as usually happens, that /spl tau/ exists. When it does, its value is related to the number of finite moments of the source density, an asymptotic formula for the overall distortion D/sub N/ is obtained, and /spl tau/=1 is both necessary and sufficient for the overall distortion to be asymptotically well approximated by /spl Delta//sub N//sup 2//12. Applying these results to the class of two-sided densities of the form b|x|/sup /spl beta//e(-/spl alpha/|x|/sup /spl alpha//), which includes Gaussian, Laplacian, Gamma, and generalized Gaussian, it is found that /spl tau/=1, that /spl Delta//sub N/ decreases as (ln N)/sup 1//spl alpha///N, that D/sub N/ is asymptotically well approximated by /spl Delta//sub N//sup 2//12 and decreases as (ln N)/sup 2//spl alpha///N/sup 2/, and that more accurate approximations to /spl Delta//sub N/ are possible. The results also apply to densities with one-sided infinite support, such as Rayleigh and Weibull, and to densities whose tails are asymptotically similar to those previously mentioned.

This paper introduces methods for reducing the table storage required for encoding and decoding with unstructured vector quantization (UVQ) or tree-structured vector quantization (TSVQ). Specifically, a low-storage secondary quantizer is used to compress the code vectors (and test vectors) of the primary quantizer. The relative advantages of uniform and nonuniform secondary quantization are investigated. A Linde-Buzo-Gray (LBG) like algorithm that optimizes the primary UVQ codebook for a given secondary codebook and another that jointly optimizes both primary and secondary codebooks are presented. In comparison to conventional methods, it is found that significant storage reduction is possible (typically a factor of two to three) with little loss of signal-to-noise ratio (SNR). Moreover, when reducing dimension is considered as another method of reducing storage, it is found that the best strategy is a combination of both. The method of secondary quantization is also applied to TSVQ to reduce the table storage required for both encoding and decoding. It is shown that by exploiting the correlation among the test vectors in the tree, both encoder and decoder storage can be significantly reduced with little loss of SNR--by a factor of about four (or two) relative to the conventional method of storing test vectors (or test hyperplanes).