Aiming at the deep structure and high computational complexity of current network models based on deep learning for super-resolution image reconstruction, as well as the problem that the networks can not operate effectively on resource-constrained devices caused by the high storage space requirement for the network models, a super-resolution convolutional neural network compression method based on knowledge distillation was proposed. This method utilizes a teacher network with large parameters and good reconstruction effect as well as a student network with few parameters and poor reconstruction effect. Firstly the teacher network was trained; then knowledge distillation method was used to transfer knowledge from teacher network to student network; finally the reconstruction effect of the student network was improved without changing the network structure and the parameters of the student network. The Peak Signal-to-Noise Ratio (PSNR) was used to evaluate the quality of reconstruction in the experiments. Compared to the student network without knowledge distillation method, the student network using the knowledge distillation method has the PSNR increased by 0.53 dB, 0.37 dB, 0.24 dB and 0.45 dB respectively on four public test sets when the magnification times is 3. Without changing the structure of student network, the proposed method significantly improves the super-resolution reconstruction effect of the student network.

Aiming at the lag consensus problem of first-order and second-order hybrid heterogeneous multi-agent systems, a distributed lag consensus control protocol based on pinning control was proposed. Firstly, the lag consensus analysis was transformed into stability verification. Then, the stability of closed loop system was analyzed by using graph theory and Lyapunov stability theory. Finally, the sufficient conditions for solvability of lag consensus based on Linear Matrix Inequality (LMI) were given under fixed and switching topologies respectively, so that leader-follower lag consensus of heterogeneous multi-agent system was achieved. The numerical simulation results show that, the proposed lag consensus control method can make the heterogeneous mult-agent systems achieve leader-follower lag consensus.

High-speed packet inspection can be achieved through storing attack signatures on the high-speed on-chip memory. Concerning the limited on-chip memory, this paper proposed a new trie structure with non-collision hash functions based on middle-point partition. The algorithm evenly partitioned attack signatures into multiple groups at each layer in trie tree to achieve the effective control of memory. The trie-tree structure can be implemented on a single chip and perform query operations by pipelining and parallelism, thus achieving higher throughput. The space complex of storing middle-point is O(n) and the construction time of hash table is linearly growing with the number of attack signatures. The experimental results show that the new structure decreases the demand of on-chip memory and can facilitate access to the attack signature on the on-chip memory while allowing to perform the signatures matching operations only once.

The traditional Orthogonal Frequency Division Multiplexing (OFDM) scheduling does not consider the proportional fairness among users' rates while allocating resource. To solve this problem, a new proportional fair scheduling scheme was proposed in multi-user OFDM system for heterogeneous classes of traffic in this paper; its users' queues carry heterogeneous classes of traffic. The scheme maximizes the sum of system's weight capacity under the constraint of the proportion among users' rates; provides different types of packets in users' queues different weight factor, and calculates the user's weight by the packet weight factor; not only defines the channel priority factor, but also allocates users sub-carriers by the factor under the constraint of the proportion among users' rates while allocating sub-carriers; finally derives a linear power allocation pattern. The simulation results and analysis demonstrate that the proposed scheme better satisfies users' requirements of throughput and strictly guarantees the fairness among users' capacity on the basis of improving system's throughput.