Chunjing Hu

Admission and power control for Device-to-Device links with quality of service protection in spectrum sharing hybrid network

In this paper, we propose admission control and power optimization algorithms for Device-to-Device (D2D) communication which share uplink resources with cellular system in underlay mode. In the presence of quality of service (QoS) protection (both for D2D and cellular users) and transmit power limit, we propose a set-based admission control (SAC) algorithm in which some novel strategies of sets adjustment are presented to handle the violations of constraints. By using SAC, we get a set which contains admitted D2D links (DLs) as many as possible, and other DLs are rejected, but the D2D system capacity is not optimal. Then, we propose a distributed power optimization (DPO) algorithm to maximize the D2D system capacity after implementation of SAC. In DPO, QoS protection is also provided, each DL uses gradient descent method to update its transmit power, and DLs only need to exchange “price” messages which indicate the effect of interference. Through the numerical results, we demonstrate the feasibility and satisfying performance of proposed algorithms compared to traditional algorithms.

Cluster-Based Multicast Transmission for Device-to-Device (D2D) Communication

The development of the storage capacity and electricity of handheld devices makes it possible to share multimedia data (e.g. video) through Device-to-Device (D2D) communication. In this paper, we present a cluster-based multicast transmission method for D2D communication with the target of decreasing the data distribution time. We consider the users who store multimedia data serve as cluster head (CH) and distribute data to the users around them in D2D multicast mode. The CHs and users in the system compose different clusters and there is only one CH in a cluster. We derive out outage capacity of a single cluster and establish a clustering selection model for the whole system. Since the clustering selection is a non-convex problem, we propose a practical clustering strategy with theoretical analysis using game theory. Simulation results show that the clustering strategy lowers the average transmission time effectively and applies to the situation of high user density.

A resource allocation scheme for D2D multicast with QoS protection in OFDMA-based systems

A resource allocation scheme involving several Device-to-Device (D2D) multicast groups underlaying OFDMA-based systems is proposed in this paper. In order to guarantee Quality of service (QoS) of cellular users (CeUEs), we define the signal-to-interference-pulse noise ratio (SINR) threshold value for CeUEs. Meanwhile, the minimum throughput of a single D2D multicast group is predetermined to maintain fairness of each D2D multicast group. Since the subcarrier assignment and power allocation are a joint optimization problem with complex calculation, the proposed scheme is divided into two steps. First, subcarriers are allocated under the premise of ensuring the minimum throughput of the D2D multicast groups and the QoS of CeUEs. Then by applying an approximation form of Shannon capacity, the power allocation problem can be solved by convex optimization. Simulation results show that the proposed scheme improves sum-throughput significantly as well as fairness. Besides, the sum-throughput increases with the increasing number of D2D multicast groups.

Distributed cooperative admission and power control for Device-to-Device links with QoS protection in cognitive heterogeneous network

In this paper, we propose two distributed admission and power control (DAPC) algorithms for Device-to-Device (D2D) communication which share uplink resources with cellular system through cognitive radio techniques in underlay mode. In the presence of quality of service (QoS) protection (both for admitted D2D link (DL) and cellular link (CL)) and transmit power limit, the DAPC-Capacity algorithm is proposed with preference for the D2D system capacity, and the DAPC-Number algorithm is proposed with preference for the number of admitted DLs. Both the two algorithms try to determine if the new coming DL can be admitted in the system, and optimize powers of DLs to maximize the D2D system capacity. In proposed algorithms, each DL uses gradient descent method to update its transmit power, and the new coming DL and admitted DLs use different formulas to calculate gradients. In calculation of gradients, DLs cooperate with each other and only need to exchange “price” messages which indicate the effect of interference. Through the numerical results, we demonstrate the satisfying performance of proposed algorithms compared to traditional centralized algorithm.

Network coding in cooperative relay networks

Network coding is seen as a promising technique to improve network throughput. In this paper, network coding in cooperative relay networks and user cooperative networks is investigated and an optimized power allocation scheme is provided. By comparing with traditional relay forwarding in a simple two-user scenario, network coding shows a better system performance with a lower hardware cost and higher spectral efficiency, and this scenario can be extended to general multi-user environment without much cost.

An overload indicator & high interference indicator hybrid scheme for inter-cell interference coordination in LTE system

Overload indicator (OI) and high interference indicator (HII) are two main approaches to mitigate inter-cell interference in long term evolution (LTE) system. Using OI or HII separately will incur inevitable shortcomings, therefore, in this paper, an OI&HII hybrid scheme is proposed for inter-cell interference coordination (ICIC) through signaling interaction between the two schemes. Instead of broadcasting OI message to all the neighboring cells, in the proposed scheme, the serving base stations (BS) just sends it to selected neighbors to inform them to adjust uplink transmission power. Simulation results show that this new scheme can effectively enhance the uplink throughput while reducing signaling overhead on the X2 interface at the same time.

Observing correlation aware (OCA) routing metric in cognitive radio networks

The end-to-end throughput of a route in Cognitive Radio Network (CRN) is related to three factors: rate of links which consist of the route, channel diversity in the route and the primary usage pattern. In this paper, we considered all these factors together. Firstly, we analyzed the capacity of the individual CR link after considering the primary usage pattern. Then, we carefully investigated the observing correlation among CR links and its influence on the end-to-end throughput. After that, we proposed a novel routing metric named Observing Correlation Aware (OCA) routing metric to capture the end-to-end throughput. Finally, simulations were conducted and results indicate that the proposed routing metric performs much better in capturing the end-to-end throughput than the routing metric which dose not consider the observing correlation.

A low complexity channel estimation approach of LTE downlink system

Channel estimation is critical to receiver performance in the long term evolution (LTE) system. Reference signals are interspersed with data signals among the subcarriers to aid channel estimation. Linear minimum mean square error (LMMSE) channel estimation is an efficient approach to estimate the knowledge of channel with high complexity. This paper proposed a channel estimation approach based on singular value decomposition (SVD) for downlink channel in LTE system. Compared to the conventional algorithm, the proposed approach investigates the channel autocorrelation matrix respectively and substitutes it by small submatrices to reduce the complexity. The simulation result demonstrates that the complexity is significantly reduced with negligible performance degradation, so it could be a candidate of the channel estimation algorithm for downlink channel estimation in LTE system.

Poster: An optimization based on semi-blind detection for LTE PUCCH format 2

Physical uplink control channel (PUCCH) in Long Term Evolution (LTE) consists of two basic transmission formats: format 1/la/lb and format 2/2a/2b. The semi-blind algorithm can improve channel estimation accuracy by using both data energy and RS energy for PUCCH format 2 in low-speed environment with a cost of higher computation complexity. This paper analyzes the semi-blind algorithm, and then we simplify the equations by further derivation. We find that the computation complex can be decreased by deducing the semi-blind algorithm formulas further. The times of complex multiplication after derivation is decreased to 2.61% to 8.67% than the non-simplified.

A compression scheme for LTE baseband signal in C-RAN

In C-RAN, the baseband units (BBU) and the remote radio heads (RRH) are separated, using the Common Public Radio Interface (CPRI) to transport data to each other. It is based on centralized processing and Real-time Cloud Infrastructure, which requires significant transport network resources. A low latency compression scheme of LTE downlink baseband signal based on clustering algorithm is presented in this paper to reduce the CPRI transmission bandwidth in C-RAN. The main procedure of algorithm is removing redundant bandwidth VQ data in frequency domain, clustering and quantifying the points on the constellation, and finally selecting an appropriate coding scheme adaptively. The Matlab simulation and FPGA-based performance analysis are presented in this paper to verify the compression algorithm. Under the premise of that EVM is less than 0.025%, especially in low throughput scenarios, the data compression ratio can reach about 4%. Moreover, the compression scheme has a strong independence and versatility, and can be used as a standalone module directly added to the current LTE system.