Machine-Learning-Based Collision Detection for Retransmissions

    公开(公告)号:US20250062853A1

    公开(公告)日:2025-02-20

    申请号:US18452435

    申请日:2023-08-18

    Abstract: Described herein are devices, systems, methods, and processes for improving retransmissions in wireless communication networks by distinguishing between temporal interference and longer-term radio frequency (RF) condition issues. The fact that the access point (AP) does not usually move may be leveraged, and a machine learning process can be utilized to learn and adapt to the RF conditions in the cell. The AP records various parameters for each frame received from client devices and uses this data to build a pairwise temporal matrix. Machine learning models are trained using these parameters, enabling the AP to compute the likely efficient set of modulation and coding schemes (MCSs) at each static position and along moving positions. The AP can then adapt its MCS accordingly for the downlink traffic and provide the client device with recommended MCSs for upcoming uplink transmissions. Accordingly, the retry count at the client devices can be reduced.

    SYSTEMS AND METHODS FOR STATIC CONTEXT HEADER COMPRESSION INSTANTIATION

    公开(公告)号:US20250030783A1

    公开(公告)日:2025-01-23

    申请号:US18354045

    申请日:2023-07-18

    Abstract: Systems, methods and computer-readable storage media are provided for encoding, within the SRH-6LoRH field within a data packet, an IPV6 address that can be used to decompress the SCHC information in the data packet. A rule is generated that indicates that the first network address in the SRH-6LoRH field of the data packet is usable to decompress the SCHC information from the data packet as opposed to the compression residue. When the data packet is received at the destination node, the destination node, through a SCHC decompressor, uses the first network address in the SRH-6LoRH field according to the rule to decompress the SCHC information.

    Generating Floorplans via Machine Learning in Network Devices

    公开(公告)号:US20240380672A1

    公开(公告)日:2024-11-14

    申请号:US18451662

    申请日:2023-08-17

    Abstract: Devices, systems, methods, and processes for network device power management at the floor level are described herein. Network devices can communicate with various devices in a wireless network. However, environmental changes, barriers, and other hazards exist in each deployment environment. To better understand each deployment environment, a floorplan can be generated by sending out a plurality of test frames. These test frames can be formatted with power data that is associated with the transmission power level that each test frame is transmitted at. The receiving network device can compare this power data with an actual measurement of the received signals. This data can all be packaged into a client report that can be sent back to the transmitting device which can then process this client report from multiple network devices such that a floorplan can be generated which can be utilized to make more efficient network decisions in the future.

    HOT STANDBY FOR BORDER ROUTERS IN LOW-POWER AND LOSSY NETWORKS

    公开(公告)号:US20250023811A1

    公开(公告)日:2025-01-16

    申请号:US18352858

    申请日:2023-07-14

    Abstract: A process can include determining respective link state information corresponding to a plurality of links between two or more border routers and a plurality of child nodes of the two or more border routers, the border routers and the child nodes included in a Destination Oriented Directed Acyclic Graph (DODAG) of a Low-Power Lossy Network (LLN). Consensus information indicative of a current status of each border router of the two or more border routers can be determined based on the respective link state information. The consensus information can be used to update an election of one or more active border routers from the two or more border routers to utilize as a virtual DODAG root for the LLN. Traffic directed to the virtual DODAG root can be routed to an active border router of the two or more border routers based on the updated election.

Patent Agency Ranking