Caching popular files in small-cell base stations (SBSs) is considered as a promising technique to meet the demand of ever growing mobile data traffic in ultra dense networks (UDNs). Considering the limited cache capacity and dense deployment of SBSs, how to support uninterrupted and successful caching downloading for moving users is still a challenging problem. In this paper, a graph-coloring-based caching (GCC) algorithm in UDN for moving user under limited SBS storage capacities is proposed. Firstly, considering there may be downloading interruption or even failure due to the random moving of users and small coverage of SBSs, graph coloring algorithm （GCA） is employed for grouping the SBS to cache fragments of several files. Then, the problem of how to conduct caching placement on SBSs is formulated aiming to maximize the amount of data downloaded from SBSs. Finally, an efficient heuristic solution is proposed to get an optimal result. Simulation results show that the algorithm performs better than other caching strategies in prior work, in terms of reducing both backhaul traffic and user download delay.

A novel adaptively iterative list decoding (ILD) approach using for Reed-Solomon (RS) codes was investigated. The proposed scheme is exploited to reduce the complexity of RS Chase algorithm （CA）via an iterative decoding attempt mode. In each decoding attempt process, a test pattern is generated by flipping the bits of least reliable positions (LRPs) within the received hard-decision (HD) vector. The ILD algorithm continues until a test pattern is successfully decoded by the underlying Berlekamp-Massey algorithm (BMA) of RS codes. Flipping within the same bits, the ILD algorithm provides the same test pattern set as the conventional RS CA, thus there is no degradation in error-rate performance. Without decoding all test patterns, the ILD algorithm can simplify the decoding complexity by its early termination. Simulation results show that the average complexity of the ILD algorithm is much lower than that of the conventional RS CA (and is similar to that of BMA decoding) at high signal-to-noise ratio (SNR)region with no less to the RS Chase decoding error-rate performance.

Mobility prediction is one of the promising technologies for improving quality of service (QoS) and network resource utilization. In future heterogeneous networks (HetNets), the network topology will become extremely complicated due to the widespread deployment of different types of small-cell base stations (SBSs). For this complex network topology, traditional mobility prediction methods may cost unacceptable overhead to maintain high prediction accuracy. This problem is studied in this paper, and the hierarchical mobility prediction scheme (HMPS) is proposed for the future HetNets. By dividing the entire process into two prediction stages with different granularity, the tradeoff between prediction accuracy and computation complexity is investigated. Before performing prediction of user mobility, some frequently visited locations are identified from the user’s trajectory, and each location represents an important geographic area (IGA). In the coarse-grained prediction phase, the next most possible location to be visited is predicted at the level of the possible geographic areas by using a second-order Markov chain with fallback. Then, the fine-grained prediction of user position is performed based on hidden Markov model (HMM) from temporal and spacial dimensions. Simulation results demonstrate that, compared with the existing prediction methods, the proposed HMPS can achieve a good compromise between prediction accuracy and complexity.

Low earth orbit (LEO) satellite network provides global coverage and supports a wide range of services. However, due to the rapid changes and energy-limitation of satellites, how to meet the demand of the quality of service (QoS) from ground traffic and prolong the lifetime of LEO satellite network is the research emphasis of the investigator. Hence, a routing algorithm which takes into account the multi-QoS requirements and satellite energy consumption (QER) of LEO satellite network is proposed. Firstly, the satellite intimacy degree (SID) and the path health degree (PHD) are introduced to obtain the path evaluation function according to the energy consumption and queue state of the satellite. Then, the distributed routing QER is established through the path evaluation function and the idea of genetic algorithm (GA), which enables each satellite to adjust traffic and realizes the network load balancing. Simulation results show that QER performs well in terms of end-to-end delay, delay jitter, and system throughput.

Using relay in the wireless communication network is an efficient way to ensure the data transmission to the distant receiver. In this paper, a dynamic power control approach (DPC) is proposed for the amplify-and-forward (AF) relay-aided downlink transmission scenario based on deep reinforcement learning (DRL) to reduce the co-channel interference caused by spectrum sharing among different nodes. The relay works in a two-way half-duplex (HD) mode. Specifically, the power control of the relay is modeled as a Markov decision process (MDP) and the sum rate maximization of the network is formulated as a DRL problem. Simulation results indicate that the proposed method can significantly improve the system sum rate.

Hybrid analog-digital beamforming is recognized as a promising solution for a practical implementation of massive multiple-input multiple-output (MIMO) systems based on millimeter-wave （mmWave） technology. In view of the overwhelming hardware cost and excessive power consumption and the imperfection of thechannel state information （CSI）, a robust hybrid beamforming design is proposed for the mmWave massive MIMO systems, where the robustness is defined with respect to imperfect knowledge or error of the CSI) at the transmitter due to limited feedback and/or imperfect channel estimation. Assuming the errors of the CSI are bounded, the optimal hybrid beamforming design with robustness is formulated to a mean squared error (MSE) minimization problem. An iterative semidefinite programming (SDP) based algorithm is proposed to obtain the beamforming matrices. Simulation results show that the proposed robust design can provide more than 4 dB performance gain compared to that of non-robust design.

Recently, physical layer security in wireless communication system attracts much attention, and the reconciliation protocol plays an important role in the final secure key distillation, since the secret keys extracted from the realistic characteristics of the wireless channel may not be the same between the transmitter and legitimate receiver. A high efficiency Polar coding key reconciliation scheme is proposed in the paper to correct these errors. In the scheme, the transmitter generates a random stream with the known frozen bits and positions. After that, the transmitter encodes the random bit stream to a code stream by Polar encoding and sends the corrupted version of the code stream and the secret keys to the legitimate receiver. The receiver decodes the received stream with Polar successive cancellation decoding algorithm. With Polar encoding, the receiver obtains a random bit stream, and achieves the final secure key by XOR operator. The results show that the proposed scheme has a higher efficiency and a lower computational complexity, along with a high success rate. The consistency of the keys is very good after the reconciliation.

Updatable block-level message-locked encryption（MLE)） can efficiently update encrypted data, and public auditing can verify the integrity of cloud storage data by utilizing a third party auditor (TPA). However, there are seldom schemes supporting both updatable block-level deduplication and public auditing. In this paper, an updatable block-level deduplication scheme with efficient auditing is proposed based on a tree-based authenticated structure. In the proposed scheme, the cloud server (CS)can perform block-level deduplication, and the TPA achieves integrity auditing tasks. When a data block is updated, the ciphertext and auditing tags could be updated efficiently. The security analysis demonstrates that the proposed scheme can achieve privacy under chosen distribution attacks in secure deduplication and resist uncheatable chosen distribution attacks (UNC-CDA) in proof of ownership (PoW). Furthermore, the integrity auditing process is proven secure under adaptive chosen-message attacks. Compared with previous relevant schemes, the proposed scheme achieves better functionality and higher efficiency.

Arbitrated quantum signature (AQS) is an important branch in quantum cryptography to authenticate quantum information, and cryptanalysis on AQS protocols helps to evaluate and improve security of AQS. Recently, it is discovered that an AQS protocol base on chained controlled-NOT (CNOT) algorithm is vulnerable to a novel attack because a transformation from binary keys into permutations and the chained CNOT algorithm have special properties, which enables a malicious receiver to forge signatures with probability 1/2. Moreover, a malicious signer can also deny his signatures with probability 1/4. Then, two possible improved methods are presented to resist these attacks: one is padding constants to reduce probability of the successful attacks, and the other is a circular chained CNOT algorithm to make the attack strategy invalid. And the security analysis shows that both the two improve methods could well resist these attacks.

Real-time graphics processing is all along a crucial task of mobile device, and it is conventionally supported by programmable graphics processing unit (GPU). These GPUs are designed to flexibly support vertex and pixel processing with classic techniques such as on-chip cache and dynamic programmable pipelining. However, it is difficult for the vertex shader and pixel shader to achieve high utilization of hardware resources, even though there is a certain balance by reasonable processor quantity ratio. In this paper, a unified render shader with a very long instruction word (VLIW) processor was designed. The viewport transformation algorithm and the mipmap mapping algorithm are respectively mapped on the shader, with the purpose of providing an energy-efficient and flexible hardware platform for graphics processing in mobile device. The implemented operating frequency is up to 134 MHz on Xilinx XC7Z045-2-FFG900 field programmable gate array (FPGA), and unified architecture shader has a performance of 134 Mpixels/s in pixel fill rate, 546Mtexels/s in Texel Fill rate.

A novel low profile multiband rectenna was proposed for harvesting the 2nd generation(2G)\ the 3rd generation (3G)\ the 4th generation (4G)\wireless local area networks (WLAN) electromagnetic wave energy. The proposed rectenna consists of a novel multiband antenna and a rectifier. The multiband antenna includes a radiating element on one side of a single layer dielectric substrate and a feeding spiral balun on the other side of the substrate. A conductive via is connected between the balun and the radiating element. In the radiating element, a deformed dipole structure is connected with an equiangular spiral slot structure and is used to generate a low frequency radiation around 900 MHz. The multiband antenna can work simultaneously at 0.869 GHz–0.948 GHz, 1.432 GHz–2.173 GHz, and 2.273 GHz–2.465 GHz with its peak gains of 7.1 dBi at 903 MHz, 4.1 dBi at 1800 MHz, 5.2 dBi at 2430 MHz. The radio frequency to direct current (RF-to-DC) conversion efficiencies of the rectifier are 58 %–62 % at these three frequencies for an input power of 0 dBm. The overall measurement results validate that the rectenna suits for energy harvesting and exhibits approximate maximum efficiencies of 58 % at 0.9 GHz, 56 % at 1.8 GHz, and 55 % at 2.4 GHz with a low incident power density of 8 μW/cm2.

The independent hypothesis between frames in vocal effect (VE) recognition makes it difficult for frame based spectral features to describe the intrinsic temporal correlation and dynamic change information in speech phenomena. A novel VE detection method based on echo state network (ESN) is presented. The input sequences are mapped into a fixed-dimensionality vector in high dimensional coding space by reservoir of the ESN. Then, radial basis function (RBF) networks are employed to fit the probability density function (pdf) of each VE mode by using the vectors in the high dimensional coding space. Finally, the minimum error rate Bayesian decision is employed to judge the VE mode. The experiments which are conducted on isolated words test set achieve 79.5% average recognition accuracy, and the results show that the proposed method can overcome the defect of the independent hypothesis between frames effectively.