30results about How to "Increase search time" patented technology

The embodiment of the present invention discloses a money search circuit, the circuit includes t sub-modules, t×(P-1) shift registers, P adders and P-1 dividers, among the t sub-modules Each submodule includes a multiplier, a two-input selector (MUX) and a D flip-flop, and the t submodules and an adder form the P channel; every t shift registers and an adder and a divider The device constitutes a channel, forming the first to P-1 channels; based on the money search circuit, the present invention also proposes an ECC decoding structure and an ECC decoding method using a two-stage pipeline. The money search circuit of the present invention saves the chip area by adjusting the number of channels, and greatly shortens the error correction time of the money search, thereby reducing the ECC decoding to two-stage pipelines and accelerating the ECC decoding speed.

Disclosed is a WCDMA mobile terminal TIS test method. Since it uses the method of quickly locating the sensitivity of any location under radiation by a CPICH RSCP value corresponding to the sensitivity conducted by a WCDMA terminal, the WCDMA mobile terminal TIS test method provided in the present invention can improve the sensitivity search time of each angle of the WCDMA TIS test, thus improving the entire measurement speed of the WCDMA TIS. Furthermore, using the method of quickly locating the sensitivity of any location under radiation by a CPICH RSCP value corresponding to the sensitivity of any point in a WCDMA terminal radiation mode can improve the sensitivity search time of each angle of the TIS test, thus improving the TIS test efficiency of the WCDMA mobile terminal.

The invention discloses an optimal selection method of a cloud data center application-aware distributed multi-resource combination path, comprising: a cloud data center application-aware cloud resource manager is responsible for collecting resource status information such as network and CPU among basic resources; Reinforcement learning, value network and policy network are combined to form an unsupervised deep hybrid architecture model, which evaluates model training and node movement positions of various request streams; a new tree search algorithm, namely parallel Monte Carlo tree search, is adopted (Parallel Monte Carlo Tree Search, PMCTS) algorithm to accelerate the search for a suitable resource path for each type of request flow, and combine the value and strategy network to give the optimal resource path selection result. By adopting the technical scheme of the present invention, it can ensure that the total response time delay of various intensive request flows in the cloud data center is minimal.

The invention discloses a multi-sink-based dynamic multi-routing source node location privacy protection method in a WSN. The method comprises the following steps: firstly, a phantom source node P<h>is randomly selected; then, a real data packet is segmented into M parts by the phantom source node P<h>, and M forwarding nodes are randomly selected in a candidate region; a shortest routing strategy is adopted to transmit M real data segments to the forwarding nodes; afterwards, the M real data segments are randomly transmitted to corresponding sink nodes along a grid by the forwarding nodes, respectively, multiple paths simultaneously exist in the network at the same moment to protect the source node location privacy; and the sink node generates a new pseudo data packet and continues to transmit along the clockwise loop to interfere with the judgement of an attacker. According to factors of network scale, the location of the source node, the number of attackers and the like, the segmented parts of the data packet are controlled, and multipath is manufactured. According to the method in the invention, the difficulty of cracking the location of the source data packet and the source node is enhanced, a large number of data packets need to be simultaneously intercepted so as to obtain the location of the source node, and the dynamic multipath improves the use ratio of the multi-sink.

The invention provides a doubly-fed wind generator high-voltage traversing rotor control method based on a follow-up algorithm aiming at the defects of an existing control method for a doubly-fed wind generator. The high-voltage traversing rotor control method includes the steps that (1), the three-phase current, the three-phase voltage, the rotating speed of the wind driven generator and the swell amplitude of power grid voltage on the rotor side are collected; (2), the collected three-phase current and the three-phase voltage are converted into numerical values in a two-phase synchronous rotating coordinate system; (3), a self-adaptive differential negative feedback value L 'alpha corrected based on the follow-up algorithm is introduced into a current feedback channel on the rotor side; (4), the q-axis voltage component V'rq on the rotor side is obtained through inversion according to the corrected differential negative feedback value L 'alpha; (5), a result of the step (4) is converted into numerical values in the two-phase static coordinate system, and the numerical values are input into a power tube in an inverter module for control. Self-adaptive differential negative feedback values of the doubly-fed wind generator at different rotating speeds and with different swell amplitudes can be optimized, so that the prediction result is more accurate, and the time consumed for processing is shorter.