88 results about "Network performance degradation" patented technology

The invention discloses a wireless network deployment method and a wireless access point, and the method comprises the steps of using one wireless access point (AP) and more than one wireless controller (AC) for respectively establishing tunnel connection, leading the AP to independently download and maintain a set of AP configuration for each tunnel, constructing the corresponding basic service setting (BSS) for each tunnel, providing wireless network services through the BSS, determining a forwarding tunnel according to an ID of the BSS of a wireless terminal when the AP receives uplink dataof the wireless terminal and sending the forwarding tunnel determined through the uplink data to the corresponding AC. The method and the wireless access point lead one AP to be capable of establishing the connection with a plurality of AC, thereby avoiding the network performance degradation due to that a plurality of operators repeatedly deploy the AP in the same region.

The embodiment of the invention discloses a method and device for network adjustment. The method includes the steps that the corresponding relation between the number of occupied network resources and the service volumes of various types of services of the network resources is obtained; according to the corresponding relation and predicated service volumes of the various types of services, the predicated occupied quantity of the network resources is determined; according to the predicated occupied quantity of the network resources, the available quantity of the network resources is adjusted. According to the embodiment, a reasonable network adjustment plan can be performed according to predicated demand conditions of future resources, so that a network operator obtains optimal return with reasonable investment, and the problems that due to constraining of the network resources, the network performance is reduced, and due to idling of the network resources, the rate of return on investment of the operator is reduced are avoided.

The present disclosure describes systems and methods that provide a hybrid framework for augmenting statistical anomaly detection with contextual features, machine learning and human Subject Matter Expert (SME) input to learn significant characteristics of true anomalies for which alerts should be generated. The framework presented herein is domain agnostic and independent of the underlying statistical anomaly detection technique or the machine learning algorithm. The framework described herein is therefore applicable and adaptable to a number of real world service provider systems and applications, such as, for example, detecting network performance degradation in a service provider network or detecting anomalous conditions from data received from a sensor while filtering out false positives.

A traffic analysis and flow-based dynamic access control system and method. The flow-based dynamic access control system for controlling a user's access to an internal communication network through an external communication network includes an access control unit operating in an access control mode in which traffic received from a user is basically blocked, generating state management information of a flow, which is received from the user, based on a specified packet of the flow, and verifying whether access of the flow to the internal communication network is a normal access. As a proactive defense concept of allowing only normal users to access an internal network, a method of blocking attacks from a system contaminated by a worm virus, detecting a cyber attack on a certain system in advance and automatically avoiding the cyber attack, and guaranteeing the quality of normal traffic even under cyber attacks without performance degradation of the internal network is provided.

The invention discloses a method and a device for ZESR and STP hybrid networking to realize service interconnection; wherein, the method comprises the steps of: a configuration step, configuring all nodes of ZESR domains to cause the nodes to pertain to the same ZESR domain and enabling the ZESR protocol; an establishing step, enabling the STP protocol for all nodes in a ZESR and STP hybrid network and establishing ZESR protocol protection examples by utilizing the STP protocol; an abandoning step, the STP protocol abandoning the management of ZESR ports under the ZESR protocol protection examples and leaving the management to the ZESR protocol. The method and the device of the invention effectively fulfill the service interconnection in the ZESR and STP hybrid network, efficiently prevent problems happening to closed loops, therefore, radically avoiding the reduction of network performance or the paralysis of network as a result of closed loops, the invention also enlarges the application scope of ZESR networking, guarantees reliable service and stable operation and improves the quick recovery ability of ZESR under the situation of complex network topology.

A system and method is disclosed to dynamically alter service offerings to a user based on the available network resources. As network resources are diminished, the end user will see less available services so as to avoid diminished network performance. If network performance improvements, the system detects the improved network recourse availability and provides more offerings to the end user.

The invention discloses a method for dynamically setting a minimum node contention window value according to retransmission frequency, which mainly solves the problem that the network performance is lowered because the minimum contention window value is statically set in the prior art. The method comprises the following implementation steps: (1) transmitting a current data packet through a node, and recording the retransmission frequency with transmission success; (2) comparing the recorded retransmission frequency with two set threshold values, when the retransmission frequency is smaller than the first threshold value, setting the minimum contention window value as a default minimum contention window value in a communication standard; when the retransmission frequency is larger than the first threshold value and smaller than the second threshold value, setting the minimum contention window value to be alpha times that of the default minimum contention window value in the communication standard; when the retransmission frequency is larger than the second threshold value, setting the minimum contention window value to be beta times that of the default minimum contention window value in the communication standard; and (3) continuing to contend channels through the nodes after the minimum contention window value is set. According to the mode, the time delay of node transmission data is reduced, and the performance of the communication network is improved.

A bus-based optical network system comprising optical fibers, optical line terminals (OLTs), and a plurality of optical network units (ONUs) is revealed. The OLT consists of the near part (OLTN) and the far part (OLTF). Optical fibers respectively form the transmitting and receiver buses (T Bus and R Bus) with the technique of time-division multiple access (TDMA). The slot flows on both buses are in opposite directions. The OLTN performing traffic control is situated in or near a central office (CO) while the OLTF is far from the CO. ONUs, whose MAC performs traffic control, connect with two buses and share their bandwidth. The MAC with traffic control keeps the performance of the network from degradation. The required numbers of COs and optical fibers for establishing the bus-based network are so smaller that the cost of the network can be significantly reduced.

The invention provides a three-dimensional U-Net brain tumor segmentation method fusing conditional randomness and residual errors. The three-dimensional U-Net brain tumor segmentation method comprises the steps: carrying out the three-layer cascade network architecture training of a training set, and obtaining a model of a convolutional neural network; testing the test set through a convolutionalneural network model to obtain a probability matrix corresponding to each category of the brain tumor; and carrying out post-processing on the probability matrix, and updating the probability to obtain a final brain tumor segmentation result. The three-dimensional U-Net brain tumor segmentation method simplifies the problem step by step, and obtains a better effect. compared with a traditional method, the three-dimensional U-Net brain tumor segmentation method has more advantages that the adopted network has better learning ability, and a residual block is introduced to reduce the influence of gradient explosion, gradient disappearance and network performance degradation caused by deepened network layers; and information of a three-dimensional space is used to the greatest extent.