Uplink capacity of a code division multiple access (CDMA) system is limited by the mutual interference among users. It is well known that interference cancellation can significantly increase the CDMA uplink capacity. In this paper, we present a new interference cancellation scheme based on non-parametric channel estimation and waveform reconstruction. Compared with the traditional parametric scheme, the new scheme can effectively solve the model mismatch problem caused by a multipath channel. We also propose channel estimation cleaning methods to further improve the performance.

Wireless MIMO relay networks have received significant interests in recent years. However, most MIMO relay techniques require high complexity equalizers and error free forwarding at the relay to collect diversity. In this paper, we propose high-rate and low-complexity decode-and-forward (DF) MIMO relay designs that achieve cooperative diversity, receive diversity or joint diversity. We introduce a power-scaling strategy for MIMO relays to enable cooperative diversity. Zero-forcing equalizers are employed at both relay and destination to maintain low-complexity. We also develop a channel-controlled ARQ scheme to collect receive diversity. Through mathematical analysis, we prove the diversity order achieved by each proposed design and numerical simulations confirm the theoretical claims.

Relay networks with amplify-and-forward (AF) protocol have attracted attention due to its high performance and low implementation complexity at the relay. To collect full cooperative diversity, the AF relay networks are facing two main issues. One is that the channel state information (CSI) of the source to relay link (i.e., two-hop information) is needed at the destination. The other concern is that the power scaling factor (PSF) and output signals at the relay are unbounded. These two issues make AF less practical in resource constrained networks, e.g., blind and peak power constrained relays. In this paper, we reveal the sufficient and necessary conditions on the PSF for the maximum ratio combining (MRC) receiver with two-hop CSI to achieve full cooperative diversity. Furthermore, we also provide necessary conditions on the PSF design so that MRC with one-hop CSI still collects full cooperative diversity. Based on these design criteria, we show a practical peak power constrained AF strategy so that neither the relay nor the destination node needs the source-relay CSI, yet full cooperative diversity can still be achieved.

Wireless relay networks have caught a lot of attention recently due to their potential to enhance performance by combating fading effects. Numerous strategies have been proposed in the literature to enable cooperative diversity. Among them, decode-and-forward protocol has been well adopted because of its low implementation complexity. However, most decode-and-forward based schemes assume that the relay node has perfect error detection or has feedback information from the destination. In this paper, we design power profiles at the relay which enable diversity at the destination without any feedback or coding assumptions at the relay and source. Furthermore, the diversity orders enabled by any general power profile are quantified. We then summarize necessary and sufficient conditions on the power profiles to guarantee full diversity at the destination. The theoretical analysis is corroborated with numerical simulations.

In this paper, we propose a novel amplify-and-forward relaying strategy for a two-hop single-relay network by incorporating an intentional peak power limit (IPPL). We show that the proposed strategy is not only practical for peak power constrained relay networks, but also helps to simplify the system at both relay and destination. With the proposed IPPL strategy, neither the relay nor the destination node needs the channel state information for the source-relay link, and full cooperative diversity can still be achieved.

Cooperative networks allow the nodes relaying each other's messages to enhance the transmission reliability over wireless fading channels by achieving cooperative diversity. Among the various relaying protocols, the amplify-and-forward (AF) strategy is well studied for its simplicity. However, to collect the cooperative diversity, there are two main issues that the AF protocol is facing. One is that the channel state information (CSI) of the source-to-relay link (i.e., two-hop CSI) is needed at the destination. The other concern is that the power scaling factor (PSF) and output signals at the relay are unbounded. These two issues make AF less practical in resource constrained networks, e.g., blind and peak power constrained relays. In this paper, we reveal the necessary and sufficient conditions on designing the PSF for the maximum ratio combining (MRC) receiver at the destination with two-hop (TH) CSI to achieve full cooperative diversity. Furthermore, we also provide the necessary conditions on the PSF design so that MRC with only one-hop (OH) CSI still collects full cooperative diversity. These designs make AF strategies more general and practical. The theoretical analysis is corroborated by numerical simulations.

Interference cancelation (IC) has been well studied in literature as a potential capacity approaching scheme for uplink code-division multiple-access (CDMA) systems. With the advance of silicon technology, it is becoming implementationally feasible. In this paper, we study a soft IC algorithm which could be iteratively concatenated with Turbo decoding. The algorithm only relies on a conventional single user Rake receiver. We study the IC efficiency as well as optimal soft decision methods. We compare different types of add-back operations between iterations. Finally we provide both link level as well as system level simulation results.

Lattice reduction (LR) techniques have been introduced to enhance the performance of linear equalizers by collecting diversity with low complexity for many transmission systems. Although LR-aided linear detectors may collect the same diversity as that collected by the maximum-likelihood (ML) detector, there still exists a performance gap between LR-aided and ML equalizers. One approach to fill this gap is to adopt soft-output detectors. Soft-output detectors enable iterative decoding when error-control codes are employed for a system. In this paper, we propose three LR-aided soft-output detectors with different methods to generate the candidate lists. We compare the performance and complexity of our algorithms with the existing alternatives and show that our methods achieve better performance with lower complexity. The performance-complexity tradeoffs of our proposed algorithms are also studied. Iterative decoders are performed based on these soft-output detectors in the simulations to validate the effectiveness of our algorithms.

Lattice reduction (LR) has been applied to linear equalizers and decision feedback equalizers to improve their performance. Recently, Seysen's algorithm (SA) has been proposed as an alternative LR method to the well-documented LLL algorithm. In this paper, we first provide a tree-search implementation method of SA which enables flexible performance-complexity trade-offs. Based on this tree structure, SA-aided hard-output and soft-output detectors are proposed to enhance performance. The diversity and complexity of the proposed methods are also studied. Thorough comparisons of SA and other LR algorithms are provided. Our analysis of complexity and numerical simulations provide insights that are useful to system designers.

Lattice reduction-aided equalization techniques have emerged as a low-complexity method to achieve the same diversity as maximum likelihood detectors. We address the VLSI implementation of these LR-aided equalizers by modifying the CLLL algorithm from a fixed-point hardware perspective. We then apply the modified algorithm together with additional micro-architecture and operation scheduling enhancements to create an updated CLLL processor. Finally, through BER simulations and FPGA synthesis results we demonstrate the suitability of our CLLL processor for integration into a 64-QAM MIMO system.

Orthogonal frequency division multiplexing (OFDM) is an effective technique to deal with frequency-selective channels since it facilitates low complexity equalization and decoding. Many existing OFDM designs successfully exploit the multipath diversity offered by frequency-selective channels. However, most of them require maximum likelihood (ML) or near-ML detection at the receiver, which is of high complexity. On the other hand, empirical results have shown that linear detectors have low complexity but offer inferior performance. In this paper, we analytically quantify the diversity orders of linear equalizers for linear precoded OFDM systems, and prove that they are unable to collect full diversity. To improve the performance of linear equalizers, we further propose to use a lattice reduction (LR) technique to help collect diversity. The LR-aided linear equalizers are shown to achieve maximum diversity order (i.e., the one collected by the ML detector), but with low complexity that is comparable to that of conventional linear equalizers. The theoretical findings are corroborated by simulation results.

Diversity order is widely adopted as a critical metric on the reliability of wireless transmissions over fading channels. Numerous designs have been proposed to collect the diversity from different types of channels. However, since the simulators and practical systems can only afford finite bits to represent the real/complex numbers, the definition of diversity needs to be revisited. In this paper, we thoroughly investigate the effects of finite-bit representation on the statistical properties of the channel and the diversity with different transmitters/receivers. We show that finite-bit representation may cause diversity loss relative to the infinite precision representation. For different transmission systems, the diversity loss follows different patterns as the number of bits decreases. Furthermore, we find that the number of non-vanishing eigenvalues plays an important role in quantizing the diversity for infinite lattice/constellation. Numerical examples verify our theoretical findings.

Multi-input multi-output (MIMO) systems equipped with multiple antennas have well documented merits in combating fading and enhancing data rates. MIMO V-BLAST transmission is a widely adopted method to achieve high spectral efficiency and low-complexity implementation. When the maximum likelihood (ML) or near-ML detector is employed, receive diversity is collected for MIMO V-BLAST systems to enhance the performance. However, because of its exponential complexity, ML detector may be infeasible for practical systems when the number of antennas and/or the constellation size is large. On the other hand, linear equalizers have much lower complexity but come with inferior performance. In this paper, we analytically quantify the diversity order of linear detectors for MIMO V-BLAST systems. Then, we adopt low-complexity complex lattice-reduction (LR) aided linear equalizers for V-BLAST systems to improve the performance and prove that LR-aided linear equalizers collect the same diversity order as that exploited by the ML detector but with much lower complexity. Relative to the existing real LR-aided equalizers, we illustrate that the complex LR further reduces the complexity while keeping the same performance. Simulation results corroborate our theoretical claims.

Linear equalizers (LEs) have been widely adopted for practical systems due to their low computational complexity. However, it is also well known that LEs provide inferior performance relative to a maximum-likelihood equalizer (MLE) or other near-MLEs, because LEs usually cannot collect the diversity order enabled by the transmitter and at the same time they lose mutual information. More importantly, unlike MLE or near-MLEs, the performance of LEs has not been well quantified. This hinders more general applications of LEs in wireless systems. In this paper, we reveal a fundamental link between a channel parameter—orthogonality deficiency$(od)$of the channel matrix—and the diversity and capacity of LEs. We identify that when the$od$of channel matrix has an upper bound strictly less than$1$, the same diversity order as that of MLEs is collected by LEs and the outage capacity loss relative to MLEs is also a constant over signal-to-noise ratio (SNR). These results can be applied to designing a framework for hybrid equalizers. Furthermore, by studying the statistical properties of the$od$and comparing the complexity of different equalizers, we show that hybrid equalizers can trade off between performance and complexity by tuning the channel matrix$od$. The theoretical analysis is corroborated by computer simulations.

Diversity techniques are important in combating the deleterious effects of channel fading. To enjoy diversity, both transmitter and receiver have to be designed appropriately. Many existing designs have exploited different flavors of diversity by employing maximum likelihood (ML) or near-ML detectors at the receiver, which is of high complexity. On the other hand, empirical results have shown that linear equalizers (LEs) offer inferior performance but come with low complexity. Unfortunately, the diversity collected by general LEs has not been quantified, which reduces the attraction of LEs in theoretical and practical aspects. In this paper, we reveal a fundamental yet simple condition that determines whether LEs can achieve the same diversity order as an ML equalizer does. The condition is based on the orthogonal deficiency (od) of channel matrix. Based on the distribution of the channel od, we propose a framework of hybrid equalizers which guarantees maximum diversity while trading off complexity for coding gains. The theoretical findings are verified by simulation results.