Due to the open nature of wireless communication, malicious electromagnetic jamming has long been a severe threat to the establishment and stability of communication links. To address this anti-jamming problem, a Markov decision process (MDP) with a two-dimensional action space consisting of transmit frequency and power is proposed in this paper, modeling the interaction between a normal communication link and the presence of malicious jammers in a frequency hopping (FH) communication system. Furthermore, we also prove the existence of the deterministic optimal policy of the proposed model theoretically. To obtain a policy for the communication link to avoid being jammed, the Dual Action Network-Based Deep Reinforcement Learning Algorithm, and Action Feedback Mechanism are proposed. The energy consumption and frequency switching overhead are considered and evaluated in both the proposed model and the algorithm. Finally, the proposed model and algorithm are verified not only in a virtual simulation environment but also in the field testing environment. The result suggests that the proposed algorithm is of great practical value for solving anti-jamming problems.

The upcoming 5G communication is committed to providing ultra-high throughput and ultra-low delay service. However, digital signal processing technologies will be a critical challenge with the increasing bandwidth and transmitting streams. Especially, traditional fast Fourier transform (FFT) architectures are hard to meet the high-speed, high-performance, and low-overhead requirements. To overcome this issue, this article proposes a reconfigurable hyper-parallel bit-serial (HPBS) FFT processor. A bit-serial paradigm implements the datapath to reduce the hardware cost due to high parallelism as high as 64 to support FFT size from 64 to 2048. The HPBS design brings several significant advantages. Firstly, the hyper-parallel architecture supports large FTT radix with lower computation complexity, e.g., fewer processing stages and constant multipliers than conventional multipliers. Secondly, the bit-serial design enables the hyper-parallel FFT implemented by an acceptable hardware cost. Thirdly, the HPBS design supports the flexible and dynamic adjusting between latency and precision, which provides an additional optimization dimension. Based on the 55-nm process, the proposed 2048-point FFT can provide 32.128 Gbps throughput at 502MHz frequency with an average 12-bit word length. Besides, the normalized hardware efficiency can be up to 46.630 Gbps/mm ² , which is at least 2X that of the traditional design.

The pedestrian trajectory prediction is critical for autonomous driving, intelligent navigation, and abnormal behavior detection. With the booming of artificial intelligence (AI), many researchers have employed deep learning technologies to solve the pedestrian trajectory prediction problem and obtained relatively better performance in the short-term trajectory prediction. However, long-term trajectory prediction is still challenging to achieve high prediction accuracy. In this work, we propose a space-time tree search (STTS) method for long-term pedestrian trajectory prediction. Compared with existing methods only considering the problem from the space dimension, the proposed method formulates the trajectory prediction problem as a joint space-time tree search process by mapping the environment to a grid map. Since the human’s trajectory is relative to space and time dimensions, the trajectory prediction accuracy can be improved by the two dimensions. Then, a space-time reward trained neural network is employed to extract the pedestrian’s intent with both the scene image and the historical trajectory as input and outputs the prior search probabilities. Finally, the tree search can obtain the optimal predicted trajectory according to the prior probabilities, significantly improving the tree search efficiency. After testing, our proposed method can perform better than existing methods.

In two-tier LEO/MEO satellite networks, reconstruction of any inter-layer inter-satellite links (ILISLs) may cause the reconfiguration of the whole network topology. Hence, the time variability of the two-tier satellite network topology will be aggravated. In this paper, two novel convergent-reconstruction-based ILISL establishment strategies are proposed. The proposed strategies can reduce the number of network topology reconfigurations and improve the topological stability of the LEO/MEO two-tier satellite networks by forcing all ILISLs to reconstruct at the same time instant. Simulation results prove the validity of the proposed strategies and show that the topological stability of the two-tier LEO/MEO satellite networks can be improved with acceptable cost

Depth estimation from a single image is a challenging task, yet this field has a promising prospect in automatic driving and augmented reality. However, the prediction accuracy is degraded significantly when the trained network is transferred from the training dataset to real scenarios. To solve this issue, we propose MonoMeMa, a novel deep architecture based on the human monocular cue, which means humans can perceive depth information with one eye through the relative size of objects, light and shadow, etc. based on previous visual experience. Our method simulates the process of the formation and utilization of human monocular visual memory, including three steps: Firstly, MonoMeMa perceives and extracts real-world objects feature vectors (encoding). Then, it maintains and replaces the extracted feature vector over time (storing). Finally, MonoMeMa combines query objects feature vectors and memory to inference depth information (retrieving). According to the simulation results, our model shows the state-of-the-art results on the KITTI driving dataset. Moreover, MonoMema exhibits remarkable generalization performance when our model is migrated to other driving datasets without any finetune.