DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-11 and 19 are rejected under 35 U.S.C. 101 because the claimed invention is directed to abstract idea without significantly more. The claim(s) recite(s) a method carried by a terminal and a terminal respectively determining a condition and using an AI parameter that corresponds to the condition. This judicial exception is not integrated into a practical application because this amounts to collecting data without an inventive concept that transform it into anything practical. The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the terminal or the terminal inclusive of a processor, a memory and the program embedded on the memory does not add anything significantly more.
Claims 14-18 and 20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to abstract idea without significantly more. The claim(s) recite(s) a method carried out by a network device and a network device respectively sending a configuration to the terminal and for configuring based on the configuration AI models parameters in different conditions. This judicial exception is not integrated into a practical application because this amounts to generic application of AI without specific technical steps. The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the terminal or the terminal inclusive of a processor, a memory and the program embedded on the memory does not add anything significantly more.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-11 and 19 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
In regards to the claims, the phrases “a first condition that the terminal meet” and an “an artificial intelligence (AI) parameter” are broad and indefinite. The condition and the AI parameters are not defined.
Claims 14-18 and 20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
In regards to the claims, the phrases “conditions” and “an artificial intelligence (AI) parameters” are broad and indefinite. For example, The condition and the AI parameters are not defined.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 1-20 are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Wang et al. (US Publication 2024/0080788 A1).
In regards to claims 1 and 19, Wang et al. (US Publication 2024/0080788 A1) teaches, a parameter selection method, comprising: determining, by a terminal, a first condition that the terminal meets (see paragraph 203 and figure 6 step 201; S201: A terminal device sends a first notification message to a network device, where the first notification message indicates that the terminal device has completed training on a target model, and the first notification message includes a model name corresponding to the target model); and using, by the terminal, an artificial intelligence (AI) model parameter corresponding to the first condition (see paragraph 212; if the first notification message includes the model parameter information of the target model, and the model parameter information includes only the parameter file information corresponding to a default parameter file type, the policy information does not need to include the lightweight object).
In regards to claim 2, Wang teaches, wherein the method further comprises: receiving, by the terminal, first configuration information from a network side device, wherein the first configuration information is for configuring AI model parameters in different conditions for the terminal; and the using an AI model parameter corresponding to the first condition comprises: using, by the terminal based on the first configuration information, the AI model parameter corresponding to the first condition (see paragraph 208 and step s202 in figure 6; S202: The terminal device receives a lightweight indication message sent by the network device, where the lightweight indication message includes the model name corresponding to the target model and policy information, the policy information is determined based on uplink channel state information of the terminal device, and the lightweight indication message indicates to the terminal device to perform lightweight processing on model parameter information of the target model).
In regards to claim 3, Wang teaches, wherein the first configuration information is for indicating, configuring, or activating an AI model parameter corresponding to each condition (see paragraph 208; where the lightweight indication message includes the model name corresponding to the target model and policy information, the policy information is determined based on uplink channel state information of the terminal device, and the lightweight indication message indicates to the terminal device to perform lightweight processing on model parameter information of the target model).
In regards to claim 4, Wang teaches, wherein the first configuration information comprises at least one of the following: a correspondence between the AI model parameter and the condition (see paragraph 209; the network device obtains the uplink channel state information of the terminal device through measurement, allocates an uplink transmission resource to the terminal device, and sends the lightweight indication message to the terminal device after determining the policy information based on the uplink channel state information, where the lightweight indication message includes the model name corresponding to the target model and the policy information); a correspondence between the AI model parameter and an event; or a correspondence between the AI model parameter and a cell.
In regards to claim 5, Wang teaches, wherein the first configuration information is for indicating, configuring, or activating an AI model parameter set corresponding to each condition (see paragraph 209; when the first notification message further includes the model parameter information, the network device may determine the policy information based on the uplink channel state information of the terminal device and the model parameter information of the target model. Because the policy information may be the size of the parameter file, the sparseness of the parameter file, or the information that helps the network device determine the lightweight method and the lightweight policy configuration, the policy information is further determined based on the parameter file information).
In regards to claim 6, Wang teaches, wherein the using, based on the first configuration information, the AI model parameter corresponding to the first condition comprises: receiving, by the terminal, first indication information from the network side device, and using, based on the first configuration information and the first indication information, the AI model parameter corresponding to the first condition, wherein the first indication information is for indicating the AI model parameter, in the AI model parameter set, that corresponds to the first condition; or the first indication information is for indicating the terminal to use at least one of the following when the condition is met: an AI model parameter used by default (see paragraph 221; if the policy information includes the lightweight object, the terminal device performs lightweight processing on the specified lightweight object; or if the policy information does not include the lightweight object, the terminal device performs lightweight processing on a default parameter file), an initially activated AI model parameter, or a preferentially used AI model parameter.
In regards to claim 7, Wang teaches, wherein the using, based on the first configuration information, the AI model parameter corresponding to the first condition comprises: using, by the terminal based on the first configuration information and a protocol agreement, the AI model parameter corresponding to the first condition, wherein the protocol agreement is that the terminal uses at least one of the following when the condition is met: an AI model parameter used by default (see paragraph 221; if the policy information includes the lightweight object, the terminal device performs lightweight processing on the specified lightweight object; or if the policy information does not include the lightweight object, the terminal device performs lightweight processing on a default parameter file), an initially activated AI model parameter, or a preferentially used AI model parameter.
In regards to claim 8, Wang teaches, wherein any one of the AI model parameter used by default, the initially activated AI model parameter, and the preferentially used AI model parameter comprises at least one of the following: an AI model parameter with a minimum identifier; an AI model parameter with a maximum identifier; an AI model parameter with a maximum data amount; an AI model parameter with a minimum data amount; an AI model parameter with a most complex model structure; an AI model parameter with a simplest model structure; an AI model parameter with a largest quantity of model layers; an AI model parameter with a smallest quantity of model layers; an AI model parameter with a highest quantization level; an AI model parameter with a lowest quantization level (see paragraph 179; To help understand model quantization, FIG. 4 is a schematic diagram of the model quantization; The quantization policy includes but is not limited to a quantity of quantized bits, a quantization manner, a quantization step, and a zero-value offset. A model processed by using the model quantization method needs to be restored to some extent before being used for model retraining or model inference); an AI model parameter with a fully-connected neural network structure; or an AI model parameter with a convolutional neural network structure.
In regards to claim 9, Wang teaches, the using an AI model parameter corresponding to the first condition comprises: using, by the terminal according to a first preset rule, the AI model parameter corresponding to the first condition (see paragraph 186; a UE may select, according to a rule from lightweight methods supported by the UE, a lightweight method and a lightweight policy configuration to perform lightweight processing on the local model, and then upload model parameter information), wherein the first preset rule comprises at least one of the following: the AI model parameter of the terminal is used by default, initially activated, or preferentially used in each condition (see paragraph 186; The selected lightweight method is a method that may be directly applied without restoring the lightweight model parameter information); the terminal uses any AI model parameter; or a common AI model parameter is used by default, initially activated, or preferentially used in each condition; or, wherein the method further comprises: skipping using, by the terminal according to a second preset rule, the AI model parameter corresponding to the first condition, wherein the second preset rule comprises: a non-AI model parameter is used by default, initially activated, or preferentially used in each condition (see paragraph 186; However, when an uplink transmission rate is high, if the UE uses a lightweight method and a lightweight policy configuration that have a high compression ratio, model precision is greatly affected. As a result, precision after aggregation is reduced. Alternatively, when an uplink transmission rate is low, if the UE uses a lightweight method and a lightweight policy configuration that have a low compression ratio, an excessively large amount of model parameter information is transferred. As a result, a waste of transmission resources can be caused, and transmission efficiency is reduced).
In regards to claim 10, Wang teaches, wherein the receiving first configuration information from a network side device comprises: receiving, by the terminal, the first configuration information from the network side device by using at least one of the following: radio resource control (RRC) signaling (see paragraph 159; For example, the CU implements functions of a radio resource control (RRC) layer and a packet data convergence protocol (PDCP) layer. The DU implements functions of a radio link control (RLC) layer, a media access control (MAC) layer, and a physical layer (PHY). Information at the RRC layer eventually becomes information at the PHY layer, or is converted from the information at the PHY layer. Therefore, in the architecture, higher layer signaling such as RRC layer signaling or PHCP layer signaling may also be considered as being sent by the DU or sent by the DU and the RU; the receiving or sending via the RRC is implicit due to the teachings of figure 6), a medium access control unit (MAC CE), or downlink control information (DCI).
In regards to claim 11, Wang teaches, wherein the method further comprises: receiving, by the terminal, second configuration information from a network side device, wherein the second configuration information comprises an updated AI model parameter (see paragraph 253 and step s313; The network device updates the model parameter information of the target model; it is implicit that the update will be sent to the terminals).
In regards to claim 12, Wang teaches, wherein the first condition comprises at least one of the following: initial access; multi-cells; cell switching; a condition determined based on a cell identifier; a condition determined based on a location area; a condition determined based on at least one of the following: a signal-to-noise ratio (SNR), a reference signal received power (RSRP), a signal-to-interference-plus-noise ratio (SINR), a reference signal received quality (RSRQ), a layer 1 SNR, a layer 1 RSRP, a layer 1 SINR, or a layer 1 RSRQ; a condition determined based on a bandwidth part (BWP); a condition determined based on a tracking area (TA) and/or a radio access network notification area (RNA); a condition determined based on an operating frequency; a condition determined based on a public land mobile network (PLMN); a condition determined based on a terminal state; a condition determined based on a quality of service flow (QoS flow) (see paragraph 237; the first notification messages sent by the two terminal devices, the network device triggers an uplink channel measurement procedure for the terminal device A and the terminal device B, to obtain the uplink channel state information of the terminal device A and the terminal device B); a condition determined based on a radio link failure (RLF) event; a condition determined based on a radio resource management (RRM) event; a condition determined based on a beam failure (BF) event and/or a beam failure recovery (BFR) event; a condition determined based on a timing measurement result and/or a timing advance measurement result; a condition determined based on a round-trip time (RTT) measurement result; or a condition determined based on an observed time difference of arrival (OTDOA) measurement result.
In regards to claim 13, Wang teaches, wherein the AI model parameter comprises at least one of the following: structure information of an AI model; or a parameter of each neuron in the AI model (see paragraph 183; or a model pruning lightweight method, a parameter configuration template may include a pruning ratio, a pruning manner, and the like of each layer in a neural network; see paragraph 178; the pruning manner in some embodiments of this application includes but is not limited to weight pruning, neuron pruning, channel pruning, and core pruning); or, wherein an AI model corresponding to the AI model parameter is used for at least one of the following: signal processing; signal transmission; signal demodulation; obtaining of channel state information; beam management; channel prediction; interference suppression; positioning; prediction of a higher layer service and a higher layer parameter; management of the higher layer service and the higher layer parameter; or parsing of control signaling.
In regards to claims 14 and 20, Wang teaches, a parameter configuration method, comprising: sending, by a network side device, first configuration information to a terminal, wherein the first configuration information is for configuring AI model parameters in different conditions for the terminal (see paragraph 208 and step s202 in figure 6; S202: The terminal device receives a lightweight indication message sent by the network device, where the lightweight indication message includes the model name corresponding to the target model and policy information, the policy information is determined based on uplink channel state information of the terminal device, and the lightweight indication message indicates to the terminal device to perform lightweight processing on model parameter information of the target model).
In regards to claim 15, Wang teaches, wherein the first configuration information is for indicating, configuring, or activating an AI model parameter corresponding to each condition (see paragraph 208; where the lightweight indication message includes the model name corresponding to the target model and policy information, the policy information is determined based on uplink channel state information of the terminal device, and the lightweight indication message indicates to the terminal device to perform lightweight processing on model parameter information of the target model); or the first configuration information is for indicating, configuring, or activating an AI model parameter set corresponding to each condition.
In regards to claim 16, Wang teaches, wherein the sending first configuration information to a terminal comprises: sending, by the network side device, the first configuration information to the terminal by using at least one of the following: RRC signaling, a MAC CE, or DCI (see paragraph 159; For example, the CU implements functions of a radio resource control (RRC) layer and a packet data convergence protocol (PDCP) layer. The DU implements functions of a radio link control (RLC) layer, a media access control (MAC) layer, and a physical layer (PHY). Information at the RRC layer eventually becomes information at the PHY layer, or is converted from the information at the PHY layer. Therefore, in the architecture, higher layer signaling such as RRC layer signaling or PHCP layer signaling may also be considered as being sent by the DU or sent by the DU and the RU; the receiving or sending via the RRC is implicit due to the teachings of figure 6).
In regards to claim 17, Wang teaches, wherein when the first configuration information is for indicating, configuring, or activating the AI model parameter set corresponding to each condition, the method further comprises: sending, by the network side device, first indication information to the terminal, wherein the first indication information is for indicating an AI model parameter, in the AI model parameter set, that corresponds to a current condition of the terminal; or the first indication information is for indicating the terminal to use at least one of the following when the condition is met: an AI model parameter used by default (see paragraph 221; if the policy information includes the lightweight object, the terminal device performs lightweight processing on the specified lightweight object; or if the policy information does not include the lightweight object, the terminal device performs lightweight processing on a default parameter file), an initially activated AI model parameter, or a preferentially used AI model parameter.
In regards to claim 18, Wang teaches, wherein the method further comprises: sending, by the network side device, second configuration information to the terminal, wherein the second configuration information comprises an updated AI model parameter (see paragraph 253 and step s313; The network device updates the model parameter information of the target model; it is implicit that the update will be sent to the terminals).
Relevant Prior Art
Prior art Yan et al. (US Publication 2023/0351207 A1) teaches in figure 2 a schematic diagram of model quantization.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JAY P PATEL whose telephone number is (571)272-3086. The examiner can normally be reached M-F 9:30-6.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Faruk Hamza can be reached at 571-272-7969. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/JAY P PATEL/Primary Examiner, Art Unit 2466
Prosecution Insights