Aiming at the problem of low generalization of the license plate detection model， which makes it difficult to reuse in different application scenes of smart transportation， a license plate detection algorithm in unrestricted scenes based on adaptive confidence threshold was proposed. Firstly， a multi-prediction head network model was constructed， in it， the segmentation prediction head was used to reduce the model reuse pre-processing work， the adaptive confidence threshold prediction head was used to improve the model detection ability， and the multi-scale fusion mechanism and bounding box regression prediction head were used to improve the model generalization ability. Secondly， a differentiable binary network training method was adopted to learn model parameters through differentiable binary transformation combined with the training of classification confidence and confidence threshold. Finally， the Connectivity Aware Non-Maximum Suppression （CANMS） method was used to improve the post-processing speed of license plate detection， and the lightweight network ResNet18 was introduced as the backbone network of feature extraction to reduce the model parameters and further improve the detection speed. Experimental results show that in 6 scenes with different constraints in Chinese City Parking Dataset （CCPD）， the proposed algorithm can achieve the average precision of 99.5% and the recall of 99.8%， and achieves the efficient detection rate of 70 frames per second， which are better than the performance of anchor-based algorithms such as Faster Region-Conventional Neural Network （Faster R-CNN） and Single Shot MultiBox Detector （SSD）. On the three supplementary scene test sets， the license plate detection accuracy of the proposed algorithm is higher than 90% in unrestricted scenes with different resolutions， different shooting distances， and different shooting angles of pitch. Therefore， the proposed algorithm has good detection performance and generalization ability in unrestricted scenes， and can meet the requirements of model reuse.

a novel Dynamic multi-Species Particle Swarm Optimization (DSPSO) algorithm based on food chain mechanism was proposed aiming at the problem that the basic Particle Swarm Optimization (PSO) algorithm is easy fall into local optimal solution when solving multimodal problems. Inspired by the natural ecosystem, a food chain mechanism and a reproduction mechanism were employed to keep the swarm diversity and good performance. In food chain mechanism, the swarm was divided into several sub-swarms, and each sub-swarm could prey on the others. The memory leader swarm was evolved and the less contributed particle was eliminated through predation, and then the new particle was generated through reproduction mechanism. The diversity was kept through the evaluation of the swarm, and the efficiency of the algorithm was enhanced through eliminating the misleading effect of the less contributed particles. In order to verify the effectiveness of the algorithm, ten benchmark problems including shifted problems and rotated problems were chose to test the performance of DSPSO. The experimental results show that DSPSO has a well optimizing performance. Compared with PSO algorithm, Local version Particle Swarm Optimization (LPSO) algorithm, Dynamic Multi-Swarm Particle Swarm Optimization (DMS-PSO) algorithm and Comprehensive Learning Particle Swarm Optimization (CLPSO) algorithm, DSPSO algorithm not only obtains more accurate solutions, but also has higher reliability.

For solving the problem of decision-tree algorithms' too much memory usage when coping with packet classification under the circumstance of high rate network and large volume rule set, a multiple decision-tree algorithm based on rule set partitioning was proposed in this paper. On the condition of controlling the number of subsets, heuristics were used to partition the rule set into limited number of subsets, in which the overlapping rules had been separated. Cascading decision-tree structure was proposed to lower the depth and reduce search time. The theoretical analysis shows that space complexity has been reduced greatly compared to traditional single decision-tree algorithm. The simulation results demonstrate that the algorithm reduces memory usage about 30% and has better dimension scalability when being compared with EffiCuts, which has the best performance for memory usage so far.

The conventional Partial Transmit Sequence (PTS) approaches get high computational complexity and need to transmit side information, which is difficult for hardware implementation. Concerning these problems, this paper proposed an algorithm of using m sequences as phase rotation factors and transferring them by pilot information. The m sequence can reduce the complexity of Field Programmable Gate Array (FPGA) implementation and the pilot transferring phase rotation factor need no side information. The Matlab simulation proves the algorithm is effective. Meanwhile, a Peak-to-Average Power Ratio (PAPR) suppression module was designed to be implemented on FPGA, and the results show that this module not only reduces the complexity of OFDM systems, but also works well in PAPR suppression.