Chaozhu Zhang

Wireless body area networks (WBAN) require long life links and energy efficient system. Besides the increasing commercialization of WBAN, health monitoring applications calls for enhanced quality of service (QoS). The establishment of the reliable and energy efficient link is crucial to support the improvement of the WBAN performance parameters. In this article, we propose a cross-layer routing mechanism for WBAN quality of service enhancement. The protocol uses a cost function, which linearly combines node energy ratio, link reliability, and specific absorption rate functions. The proposed algorithm initially maximizes network lifetime longevity by reducing node energy consumption with nearly reasonable throughput and the packet delivery ratio whereas the enhancement of the QoS focused on improving network throughput and the packet delivery success rate for WBAN applications. The algorithm is implemented in two stages, firstly by designing the energy efficient and reliable link routing policy in the network layer and secondly the adjustment of the contention window for QoS performance enhancement in the data link layer using IEEE 802.11 medium access control (MAC) protocol. We conduct parametric modeling of the cost function to analyze network performance in different parametric combinations and contention window adjustments. Simulation results show the proposed protocol improves network performance indicators such as energy efficiency, lifetime longevity maximization, throughput, packet success delivery ratio, reduced end to end delay, and routing overhead.

This paper proposes a coupled compressed samples online robust learning (C2ORL) algorithm for constructing an adaptive analog-to-information conversion (AIC) structure. The C2ORL optimizes both the dictionary and the observation matrix in each mini-batch learning process. The compressed samples are used for learning to resolve the issue that both the analog received signals and the unstable reconstructed signals affected by reconstruction error are not suitable for direct training. Furthermore, the online mechanism and the robust objective functions are adopted in order to make the learning algorithm better matched for the requirements of the analog signal compressed sensing (CS). Based on the C2ORL, we construct an adaptive AIC structure, which can considerably improve the performance in the sense of minimizing the reconstruction error. Also, the structure is more tolerant to the outliers and the noise in the receive process and has denoising ability to improve the signal-to-noise ratio (SNR) under the noise background. The mathematical expression shows that the proposed structure clearly reflects the optimization brought by the coupled learning algorithm. Finally, the experiments demonstrate that the proposed scheme considerably outperforms any existing approaches.

In the development of analog signal compressive sensing (CS), the degradation of reconstruction performance under noise is the main bottleneck, because the CS framework is very sensitive to noise. This paper proposes an adaptive channelization-based orthogonal matching pursuit algorithm (C-OMP) combining the channelization and the adaptive iteration methods. The proposed C-OMP has two steps: channel screening and global iteration. Based on the proposed method, the original signal can be recovered adaptively in high probability of success with fewer observations under the noise background. Simultaneously, the noise can be reduced as much as possible to enhance the output signal-to-noise ratio (SNR) by excluding the noise channel during the channel screening and separating noise atoms during the global iteration. The relationship between the probability of successful reconstruction and the number of observations is mathematically analyzed. Furthermore, the parameter settings, computational complexity and output SNR are analytically evaluated. The simulation results confirm the analytical results and further demonstrate the effectiveness and advantages of the C-OMP in the noise environment. Overall, the proposed algorithm considerably improves the performance of the analog signal CS in the practical noisy environment.

This paper studies joint beamforming optimization and time allocation for a wireless powered two-way hybrid-relay network. As a general relaying architecture accommodating the existing relaying architectures in the literature as two special cases, we consider the two-way communication system where both energy-harvesting (EH) relays and non-EH relays exist. We aim to maximize the minimum rate by jointly optimizing the distributed beamforming vector at the relays, the energy beamforming matrix at the non-EH relays, and the EH time allocation with the energy constraints at the relays. The formulated problem is intractable to solve because the objective function and the energy constraints are highly non-convex. We solve the optimization problem using the successive convex approximation (SCA). Simulations demonstrate that the proposed iterative algorithm considerably outperforms any existing approaches.

This paper studies the joint beamforming design and time allocation for a two-way energy harvesting (EH) relay network, where a single-antenna base station and a single-antenna user communicate with each other through multiple-single antenna EH relay nodes, and an external multiantenna power beacon wirelessly power the relays. Unlike the existing studies that assumed the symmetric two-way relaying scenario, we assume a more practical scenario of asymmetric two-way relaying considering different traffic demands on the downlink and uplink. In this paper, we design the optimal beamforming vectors at the relays and at the power beacon jointly with the time allocation between information and EH transmissions, in the sense of maximizing the sum rate of the asymmetric information transmission under the EH constraints at the relays, and the power constraint at the power beacon. The formulated optimization problem is nonconvex, and thus, it is difficult to solve. To solve this challenging problem, we propose an iterative algorithm based on the semidefinite relaxation and the successive convex approximation techniques. Also, to reduce the computational complexity, we propose an efficient noniterative scheme based on the maximum-ratio transmission. The performance of the proposed schemes is demonstrated through simulations.

In this paper, we address the problem of detecting and tracking targets with a low signal-to-noise ratio (SNR) by exploiting hybrid differential evolution (HDE) in the particle filter track-before-detect (PF-TBD) context. Firstly, we introduce the Bayesian PF-TBD method and its weaknesses. Secondly, the HDE algorithm is regarded as a novel particle updating strategy, which is proposed to optimize the performance of the PF-TBD algorithm. Thirdly, we combine the systematic resampling approach to enhance the performance of the proposed algorithm. Then, an improved PF-TBD algorithm based on the HDE method is proposed. Experiment results indicate that the proposed method has better performance in detecting and tracking than previous algorithms when the targets have a low SNR.