Heterogeneous network (HetNet) plays an important role in the upcoming 5G communications. A HetNet in futuristic communications is envisaged to provide full access data services, such that users will always have the best service with their available radio resources. To realize a full access HetNet, statistical signal transmission (SST) can be an important enabling technique. SST exploits statistical features of ordinary signals as a data transmission scheme to carry information data. Due to its inherent unique properties, SST signal is more robust than ordinary signals in conventional quasi-static channels. However, detection of SST signals in fast time-varying channels will face challenges. Motivated by this fact, this paper aims to analyze the impact of fast time-varying channels to SST, and reveals negative effects of burst of deep fading channel-state (BDFC) and feature dissipation with out-of-phase (FDOP). Accordingly, we design an automatic re-transmission scheme and a fragmental feature capturing method to mitigate the aforementioned issues, respectively. Based on these studies, we propose a channel condition aware detection scheme for SST, aiming to enhance SST detection performance in fast time-varying channels. Numerical results show that with the proposed scheme, a maximum gain of more than 10 dB can be obtained under either BDFC or FDOP condition. Furthermore, it is also demonstrated that the proposed scheme is robust with imperfect channel coherence time information, which is attractive for practical applications.

Wireless caching at users' devices in mobile social network is considered to be a promising solution to alleviate backhaul overload in future wireless networks. However, most of the current works propose caching schemes based on heuristic reasoning and intuition with poor performance or high complexity which are impractical due to individual devices' computing capacity restriction. In this paper, we design a cooperative caching scheme aimed at maximizing hit ratio, incorporating probabilistic modeling of mobility and user interests patterns from mobile social networks. Furthermore, we reformulate this optimization problem into a submodular function maximization and propose a semigradient-based cooperative caching scheme, while this scheme's efficiency is shown to significantly outperform the greedy caching by 99.6%.

Based on orthogonal frequency division multiplexing (OFDM) systems, an additional individual data transmission can be established in statistical spectrum domain, which can be adopted for specific applications. However, the performance of existing detection methods for this kind of transmission will be severely restricted when the channel coefficients are dynamic within each observation window. In this paper, we propose an adaptive detection method for statistical signal transmission. This method has adequate adaptability for different channel varying rates, while maintaining low complexity. Simulation results show that the proposed method achieves significant performance gain compared to the original method under time varying channels.

Network Coding based Epidemic Routing (NECR) has been proposed to improve the data delivery efficiency in Delay Tolerant Networks (DTNs). With NCER, data packets are not only replicated but also encoded while being forwarded to their destination. To better understand how NCER facilitates the transmission, in this paper, we present an analytical framework for modeling the data transmission delay performance of NCER in DTNs. Unlike previous work, our framework can accommodate the inaccuracy of a widely adopted assumption that each node carrying one or more encoded packets is able to transmit an innovative encoded packet to the node it encounters. Numerical results are presented to validate the accuracy of our analytical framework.

The focus of this paper is to study the performance of the quantize-and-forward (QF) scheme over a half-duplex relay channel that is slowly fading, with the assumption that the channel state information (CSI) is available only at the receiver side. In order to do so, three steps are taken. The first step is to characterize the achievable rate of the QF scheme over a discrete memoryless half-duplex relay channel. Then, the achievable rate over a corresponding additive white Gaussian noise channel is obtained (the specific assumption regarding the CSI in this paper makes this step nontrivial). With the results from the first two steps, performance measures such as outage probability, expected rate, and diversity-multiplexing tradeoff (DMT) over slow fading channels are evaluated. It is shown that the QF scheme can significantly outperform the compress-and-forward scheme at finite signal-to-noise ratio (SNR) and it can achieve the optimal DMT at asymptotically high SNR. Moreover, it is shown that simple feedback from the destination node to the relay node can further improve the performance of the QF scheme.

The original compress-and-forward relaying scheme uses the technique of random binning at the relay node and successive decoding at the destination node. Recently, a scheme (termed the quantize-and-forward scheme in this paper) without binning and using joint decoding at the destination node has been proposed, which has been shown to achieve the same rate as the original compress-and-forward scheme. Since the previous work focuses on the so-called full duplex relay network, in this paper, an adaption of it for relay networks with a half-duplex relay is provided. Coding schemes and achievable rate results are presented for discrete memoryless half-duplex relay channels and half-duplex additive white Gaussian noise (AWGN) relay channels. Moreover, slow fading channels are considered, for which outage-related performance measures are evaluated. Specifically, the outage probability and the expected rate of the quantize-and-forward scheme are derived and compared with other well-known schemes. Furthermore, the diversity-multiplexing tradeoff is derived. It is shown that the quantize-and-forward scheme is a more suitable scheme than the compress-and-forward scheme over slow fading channels and it achieves the optimal diversity-multiplexing trade-off of a half-duplex relay channel.

We consider the multiple-access relay channel with two source nodes, one relay node and one destination node. For practical simplicity, we consider orthogonal transmission of the source messages and half-duplex relaying. We also assume that the relay is memoryless and is implemented based on a two-to-one deterministic mapping. Our focus is on proposing and investigating such mappings. Essentially, the proposed relaying functions combine the two incoming analog signals and forward them to the destination, thus we term them as analog network coding mappings. Both linear and non-linear relaying are investigated for the multiple-access relay channel and the special case of the multiple-access two-hop channel. In particular, we suggest to use mappings based on the Archimedean spiral for analog non-linear combining. In addition, we propose to couple spiral mappings with sawtooth-like mappings to exploit the potential side information provided by the direct links of the multiple-access relay channel. In the case of symmetric topology, our proposed scheme can be seen as an extension to the amplify-and-forward scheme, where the asymmetric input/output dimensionality in the relay node is handled before amplifying. We investigate the resulting achievable rate regions and sum rates, and demonstrate significant gains over conventional relaying schemes.

We study the problem of optimal design of relay mappings for the Gaussian relay channel in order to maximize the reliable transmission rate. We consider both Gaussian and modulation constrained signaling at the source. To optimize the relay mapping, we use an iterative integral equation as a necessary condition for optimality. The optimized relay mappings demonstrate significant rate improvement over conventional linear relaying (amplify-and-forward). The optimized mappings allow an efficient utilization of the side information received via the source-destination link at the destination. Hence, the proposed mappings can be considered as an analog, memoryless approach to implementing compress-and-forward relaying with Wyner-Ziv compression in the relay.

We consider the multiple-access relay channel with two source nodes, one relay node and one destination node. For practical simplicity, we consider orthogonal transmission of the source messages and half-duplex relaying. Further, we assume that the relay is memoryless and is implemented based on a two-to-one deterministic mapping. Essentially the proposed mappings combine the two incoming analog signals and forward them to the destination, thus we term them analog network coding mappings. We investigate both linear and non-linear mappings. In particular, we propose to use mappings based on the Archimedean spiral for analog non-linear combining. In addition, we also propose to couple spiral mappings with sawtooth-like mappings to exploit the potential side information provided by the direct links. We investigate the resulting achievable rate regions and sum rates, and demonstrate significant gains.

We consider the multiple-access two-hop channel where two source nodes transmit to a destination node via a relay node. The relaying function is memoryless, in contrast to the conventional schemes based on coding with long codewords. That is, we model the operation of the relay as a two-to-one deterministic mapping, which combines the two received analog signals from the sources. This procedure resembles the concept of network coding where information combining is applied in the intermediate nodes. However, as our mapping directly combines the received analog signals without decoding, we coin the term (memoryless) analog network coding mapping. In this paper, both linear and non-linear mappings are studied. In particular, the Archimedean spiral is used for the non-linear 2:1 mapping, inspired by similar work in the context of joint source-channel coding. We discuss both the achievable rate regions and sum rates and demonstrate significant gains of applying the proposed analog mappings in the relay.

Hybrid digital-analog coding schemes have been proposed in source-channel coding to increase the robustness toward channel mismatch, in the absence of transmitter channel state information (CSIT). Recognizing that the same kind of robustness is needed at the relay in a three-node relay network, we propose several novel relaying protocols based on hybrid digital-analog transmission. We compare the performance of the new schemes with traditional digital-only (decode-and-forward or compress-and-forward) or analog-only (amplify-and-forward) relaying, as well as to performance bounds corresponding to genie-aided compress-and-forward relaying. Our new protocols achieve significant gains in terms of achievable expected rates, and they are able to close in on the performance bounds. In particular, we conclude that the best overall performance is obtained by an adaptive combination of decode-and-forward and hybrid digital-analog relaying.

Hybrid digital-analog schemes for bandwidth compression/expansion have been proposed in source-channel coding to increase the robustness in terms of end-to-end distortion, and to combat threshold and leveling-off effects in the absence of transmitter channel state information. In the scenario that the dimension of the incoming codeword is compressed by the relay, we address a similar problem in a three-node half-duplex orthogonal relay network over slow fading channels, and propose two hybrid digital-analog relaying protocols that show significant improvements over digital-only (compress-and-forward and decode-and-forward) and analog-only (amplify-and- forward) schemes, in terms of the maximum achieved expected rate from source to destination.