# US 12,593,240 B2 · Dictionary-based AI components in wireless systems > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,593,240 B2 - Document ID: US12593240B2 - Status: Granted - Kind: B2 - Application number: 18/034,504 - Filed: 2021-10-25 - Published: 2023-12-14 - Granted: 2026-03-31 - Earliest claimed: 2020-10-29 - Assignees: InterDigital Patent Holdings, Inc. - Applicants: InterDigital Patent Holdings, Inc. - Inventors: Yugeswar Deenoo Narayanan Thangaraj; Ghyslain Pelletier - Web page: https://patentgrep.com/patent/US12593240B2 ## CPC - H04W 28/0273: {adapting protocols for flow control or congestion control to wireless environment, e.g. adapting transmission control protocol [TCP] (wireless network protocols or protocol adaptations to wireless operation, e.g. wireless application protocol H04W80/00)} - G06N 3/042: Knowledge-based neural networks; Logical representations of neural networks - G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] - G06N 3/0455: Auto-encoder networks; Encoder-decoder networks - G06N 3/0464: Convolutional networks [CNN, ConvNet] - G06N 3/084: Backpropagation, e.g. using gradient descent - G06N 3/09: Supervised learning - H04L 5/0053: {Allocation of signalling, i.e. of overhead other than pilot signals} - H04L 45/08: {Learning-based routing, e.g. using neural networks or artificial intelligence} - G06N 3/044: Recurrent networks, e.g. Hopfield networks - G06N 3/045: Combinations of networks - H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) ## Abstract A wireless transmit-receive unit (WTRU) may be configured to perform one or more actions. A first transmission of a first type and a second transmission of a second type may be received. The first transmission and the second transmission may be determined to be linked based on a time frequency relationship or configuration information. First control information may be determined based on the first transmission and pre-defined processing rules. The second transmission may be provided as an input to an artificial intelligence (AI) component and may cause the AI component to generate second control information as an output. A loss value may be determined based on a difference between the first control information and the second control information. The loss value may be used to train the AI component. The training of the AI component may be determined to satisfy a condition. An indication of the determination may be transmitted. ## Claims ### Claim 1 A wireless transmit-receive unit (WTRU), the WTRU comprising: a processor, the processor configured to: receive a first transmission and a second transmission, wherein the first transmission is a first type of transmission and the second transmission is a second type of transmission different from the first type of transmission; determine that the first transmission and the second transmission are linked; determine first control information based on the first transmission and a pre-defined processing rule; provide the second transmission as a first input to an artificial intelligence (AI) component, wherein the second transmission being provided as the first input to the AI component causes the AI component to generate second control information as a first output; determine a loss value based on at least a first difference associated with the first control information and the second control information; train the AI component based on the loss value, wherein the AI component being trained causes the AI component to generate third control information as a second output, and wherein a second difference associated with the first control information and the third control information is less than the first difference; determine that the training of the AI component satisfies a condition; and transmit, to a network, an indication of the determination that the training of the AI component satisfies the condition. ### Claim 2 (depends on 1) The WTRU of claim 1, wherein the first type of transmission is rule-based and the second type of transmission is AI component-based. ### Claim 3 (depends on 1) The WTRU of claim 1, wherein the first transmission and the second transmission differ in one or more of waveform, modulation, bit arrangement, precoding, or compression. ### Claim 4 (depends on 1) The WTRU of claim 1, wherein the first transmission and the second transmission are linked based on a time frequency relationship, and wherein the time frequency relationship is based on control resource set (CORESET) configuration information or search space configuration information. ### Claim 5 (depends on 1) The WTRU of claim 1, wherein the pre-defined processing rule is associated with a rule-based component. ### Claim 6 (depends on 1) The WTRU of claim 1, wherein the determination that the training of the AI component satisfies the condition is based on the loss value being below a threshold. ### Claim 7 (depends on 1) The WTRU of claim 1, wherein the first difference associated with the first control information and the second control information comprises a difference between a first part of the first control information and a first part of the second control information. ### Claim 8 (depends on 1) The WTRU of claim 1, wherein the processor is further configured to receive a command indicating to activate the trained AI component. ### Claim 9 (depends on 1) The WTRU of claim 1, wherein the processor is further configured to receive a third transmission, wherein the third transmission is of the second type of transmission, and apply the third transmission as input to the trained AI component to obtain a third output corresponding to a control message. ### Claim 10 (depends on 9) The WTRU of claim 9, wherein the processor is further configured to perform one or more actions based on the control message. ### Claim 11 A method performed by a wireless transmit-receive unit (WTRU), the method comprising: receiving a first transmission and a second transmission, wherein the first transmission is a first type of transmission and the second transmission is a second type of transmission different from the first type of transmission; determining that the first transmission and the second transmission are linked; determining first control information based on the first transmission and a pre-defined processing rule; providing the second transmission as a first input to an artificial intelligence (AI) component, wherein the second transmission being provided as the first input to the AI component causes the AI component to generate second control information as a first output; determining a loss value based on at least a first difference associated with the first control information and the second control information; training the AI component based on the loss value, wherein the AI component being trained causes the AI component to generate third control information as a second output, and wherein a second difference associated with the first control information and the third control information is less than the first difference; determining that the training of the AI component satisfies a condition; and transmitting, to a network, an indication of the determination that the training of the AI component satisfies the condition. ### Claim 12 (depends on 11) The method of claim 11, wherein the first type of transmission is rule-based and the second type of transmission is AI component-based. ### Claim 13 (depends on 11) The method of claim 11, wherein the first transmission and the second transmission differ in one or more of waveform, modulation, bit arrangement, precoding, or compression. ### Claim 14 (depends on 11) The method of claim 11, wherein the first transmission and the second transmission are linked based on a time frequency relationship, and wherein the time frequency relationship is based on control resource set (CORESET) configuration information or search space configuration information. ### Claim 15 (depends on 11) The method of claim 11, wherein the pre-defined processing rule is associated with a rule-based component. ### Claim 16 (depends on 11) The method of claim 11, wherein the determination that the training of the AI component satisfies the condition is based on the loss value being below a threshold. ### Claim 17 (depends on 11) The method of claim 11, wherein the first difference associated with the first control information and the second control information comprises a difference between a first part of the first control information and a first part of the second control information. ### Claim 18 (depends on 11) The method of claim 11, further comprising receiving a command indicating to activate the trained AI component. ### Claim 19 (depends on 11) The method of claim 11, further comprising receiving a third transmission, wherein the third transmission is of the second type of transmission, and applying the third transmission as input to the trained AI component to obtain a third output corresponding to a control message. ### Claim 20 (depends on 19) The method of claim 19, further comprising performing one or more actions based on the control message.