# US 10,873,371 B1 · Antenna for massive multiple input and multiple output (mMIMO) > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,873,371 B1 - Document ID: US10873371B1 - Status: Granted - Kind: B1 - Application number: 16/544,427 - Filed: 2019-08-19 - Granted: 2020-12-22 - Earliest claimed: 2019-08-19 - Assignees: Cisco Technology, Inc. - Applicants: Cisco Technology, Inc. - Inventors: Satish Kumar Mandavilly; Amitesh Shukla; Manish Jhanji; Rakesh Mishra; Subhash Kodnad - Web page: https://patentgrep.com/patent/US10873371B1 ## CPC - H04B 7/0417: Feedback systems - H04B 7/0617: {for beam forming} - G06N 20/00: Machine learning - H04B 7/0426: Power distribution - H04B 7/0452: Multi-user MIMO systems - G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound - G06N 5/025: Extracting rules from data - G06N 20/10: using kernel methods, e.g. support vector machines [SVM] - G06N 20/20: Ensemble learning - H04B: TRANSMISSION - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract An antenna may be provided. First, a Machine Learning (ML) model may be used, at a predetermined time, to predict a probable location of at least one of a plurality of client devices. Next, spatial coordinates may be obtained for the probable location from the ML model. Then an antenna sub-array of an antenna array may be steered toward the spatial coordinates of the probable location. The antenna sub-array may be unallocated. ## Claims ### Claim 1 A method comprising: using, by a computing device at a predetermined time, a Machine Learning (ML) model to predict a probable location of one of a plurality of client devices in a cell comprising a beam area of an antenna array operative to send and receive signals from the plurality of client devices; obtaining spatial coordinates with respect to the antenna array for the probable location of the one of the plurality of client devices from the ML model; and steering an antenna sub-array of the antenna array toward the spatial coordinates of the probable location of the one of the plurality of client devices, the antenna sub-array being unallocated. ### Claim 2 (depends on 1) The method of claim 1, wherein obtaining spatial coordinates comprises obtaining spatial coordinates comprising an angular rotation value, a phase shift value, and a power value associated with the probable location of the one of the plurality of client devices. ### Claim 3 (depends on 2) The method of claim 2, wherein steering the antenna sub-array toward the spatial coordinates of the probable location of the one of the plurality of client devices comprises focusing the antenna sub-array on the angular rotation value and the phase shift value and operating the antenna sub-array at the power value. ### Claim 4 (depends on 1) The method of claim 1, further comprising moving the antenna sub-array to a free pool when a client device does not latch on to the antenna sub-array. ### Claim 5 (depends on 1) The method of claim 1, further comprising latching a client device on to the antenna sub-array. ### Claim 6 (depends on 1) The method of claim 1, further comprising maintaining the ML model associated with probable locations of the plurality of client devices within a cell serviced by the antenna array. ### Claim 7 (depends on 6) The method of claim 6, wherein maintaining the ML model comprises training the ML model with data provided by the antenna array to create an input feature vector. ### Claim 8 (depends on 6) The method of claim 6, wherein maintaining the ML model comprises training the ML model with data external to the antenna array to create an input feature vector. ### Claim 9 (depends on 6) The method of claim 6, wherein maintaining the ML model comprises receiving feedback with respect to the probable location of the one of the plurality of client devices when an actual location at which a client device latched on the antenna sub-array is different from the probable location of the one of the plurality of client devices. ### Claim 10 (depends on 9) The method of claim 9, wherein maintaining the ML model comprises making an adjustment to the probable location of the one of the plurality of client devices in the ML model based on the feedback. ### Claim 11 A system comprising: a memory storage; and a processing unit coupled to the memory storage, wherein the processing unit is operative to: use a Machine Learning (ML) model to predict a probable location of one of a plurality of client devices in a cell comprising a beam area of an antenna array operative to send and receive signals from the plurality of client devices; obtain spatial coordinates with respect to the antenna array for the probable location of the one of the plurality of client devices from the ML model; and steer an antenna sub-array of the antenna array toward the spatial coordinates of the probable location of the one of the plurality of client devices, the antenna sub-array being unallocated. ### Claim 12 (depends on 11) The system of claim 11, wherein the spatial coordinates comprise an angular rotation value, a phase shift value, and a power value associated with the probable location of the one of the plurality of client devices. ### Claim 13 (depends on 12) The system of claim 12, wherein the processing unit being operative to steer the antenna sub-array toward the spatial coordinates of the probable location of the one of the plurality of client devices comprises the processing unit being operative to focus the antenna sub-array on the angular rotation value and the phase shift value and operating the antenna sub-array at the power value. ### Claim 14 (depends on 11) The system of claim 11, wherein the processing unit is further operative to move the antenna sub-array to a free pool when a client device does not latch on to the antenna sub-array. ### Claim 15 (depends on 11) The system of claim 11, wherein the processing unit is further operative to latch a client device on to the antenna sub-array. ### Claim 16 A non-transitory computer-readable medium that stores a set of instructions which when executed perform a method comprising: using, by a computing device at a predetermined time, a Machine Learning (ML) model to predict a probable location of one of a plurality of client devices in a cell comprising a beam area of an antenna array operative to send and receive signals from the plurality of client devices; obtaining spatial coordinates with respect to the antenna array for the probable location of the one of the plurality of client devices from the ML model; and steering an antenna sub-array of the antenna array toward the spatial coordinates of the probable location of the of the plurality of client devices, the antenna sub-array being unallocated. ### Claim 17 (depends on 16) The non-transitory computer-readable medium of claim 16, wherein obtaining spatial coordinates comprises obtaining spatial coordinates comprising an angular rotation value, a phase shift value, and a power value associated with the probable location of the one of the plurality of client devices. ### Claim 18 (depends on 17) The non-transitory computer-readable medium of claim 17, wherein steering the antenna sub-array toward the spatial coordinates of the probable location of the one of the plurality of client devices comprises focusing the antenna sub-array on the angular rotation value and the phase shift value and operating the antenna sub-array at the power value. ### Claim 19 (depends on 16) The non-transitory computer-readable medium of claim 16, further comprising moving the antenna sub-array to a free pool when a client device does not latch on to the antenna sub-array. ### Claim 20 (depends on 16) The non-transitory computer-readable medium of claim 16, further comprising latching a client device on to the antenna sub-array.