Methods and systems for handling thermal mitigation in a user equipment(UE)

Abstract

Embodiments herein disclose methods and systems for handling thermal mitigation in a User Equipment (UE) using a Machine Learning (ML) module. Embodiments herein disclose the method comprising: estimating, by a Machine Learning (ML) module, a time required by at least one component of a Radio Resource Configuration (RRC) to operate on a wireless communication network before initiating the thermal mitigation; and predicting, by the ML module, a cumulative time till the at least one component of the RRC can operate on the wireless communication network based on the estimated time of each component of the RRC. Embodiments herein disclose methods and systems for identifying a configuration required for the at least one component of the RRC, to extend time for using at least one service on the wireless communication network.

Description

[Rectified under Rule 91, 04.09.2025]METHODS AND SYSTEMS FOR HANDLING THERMAL MITIGATION IN A USER EQUIPMENT(UE)

[0001] Embodiments as disclosed herein relate to handling thermal mitigation in a User Equipment (UE), and more particularly to predicting a temperature of the UE using at least one service on a wireless communication network, based on at least one component of a Radio Resource Configuration (RRC).

[0002] Currently, when using Fifth Generation (5G) networks, the amount of heat generated by the user devices are higher compared to the heat generated when using previous technologies. The heat generated when using the 5G network may include, but are not limited to, higher frequencies of data transmission, higher data rates that require faster processing, larger current drain, higher order carrier aggregation (CA), performing multiple tasks, availing multiple services, Multiple-Input, Multiple-Output (MIMO) on non-standalone (NSA) and standalone 5G (SA) network, and so on.

[0003] In existing mechanisms, to mitigate the thermal issues in the 5G network, a modem can be employed which can be configured to use the thermal management application. The thermal management application may use inputs provided by a thermistor that are configured on the chipset. When the temperature reaches a pre-set threshold, the user equipment (UE) / user devices may be configured to provide necessary actions to cap the performance of the UE to mitigate the increase in the temperature. The modem can be configured to fake the network (about its capability to not support the high-performance components, which may include, but are not limited to MIMO, CA, and so on).

[0004] In another procedure, the third Generation Partnership Project (3GPP) has introduced UE Assistance Information (UAI) procedure. In the procedure, the UAI can request the network to cut down the performance parameters, wherein the UAI can provide the performance preferences to the network for reducing the usage of radio configuration to mitigate the thermal overheating. Also, the network may re-configure the UE with the requested reduced radio configuration.

[0005] Also, the corrective actions performed by the thermal application may be late, i.e., the thermistor may raise the interruption after reaching the threshold breach level. Therefore, the time required to reach the final threshold level may be quick, and the 5G network may be disabled. For an instance, the first threshold breach may occur at forty degrees and at the forty-two to forty-three degrees, the 5G network may be disabled. The radio configuration that causes heating may have a steady temperature ramp, which can be modelled and known to the UE before the occurrences, but the thermal application may wait for the threshold breach before being identified by the thermistor.

[0006] Further, the thermal application may not consider the ambient temperature or the rate at which the heat dissipates in the device. The heat dissipation may vary, which may include, but are not limited to, the configuration of the components, the hardware, material, cover / casing, ambient temperature, and so on. Although the current 3GPP has introduced UAI for the thermal mitigation, currently there does not exist procedure in the UE by which the UAI configuration can be formulated and negotiated with the network.

[0007] Existing applications does not provide any indication of the thermal state of the device to a user of the device. On estimating the amount of time that the UE can hold on 5G before the Long-Term Evolution (LTE) fall back, the application can optimize the Transmission Control Protocol (TCP) windows, application performances, provide a better User Experience (UX). Also, some components of the radio configuration are very expensive in terms of the thermal cost which are higher order MIMO, CA combinations, high bandwidth configuration, and so on.

[0008] In various networks, slices may have different thermal costs. Some slices such as Enhanced Mobile Broadband (eMBB) may use very high bandwidths leading to a high thermal cost compared to a slice in Mobile Internet of Things (M-IoT) which is a low-capacity slice. Hence, profiling the slices based on the thermal costs provides in reducing the thermal cost. Also, the high-cost slices can be avoided in the thermal mitigation scenarios. The existing thermal application may rely on the thermistor values which may not be accurate due to hardware limitations.

[0009] Hence, there is a need in the art for solutions which will overcome the above-mentioned drawback(s), among others.

[0010] These and other aspects of the embodiments herein will be better appreciated and understood when considered in conjunction with the following description and the accompanying drawings. It should be understood, however, that the following descriptions, while indicating at least one embodiment and numerous specific details thereof, are given by way of illustration and not of limitation. Many changes and modifications may be made within the scope of the embodiments herein without departing from the spirit thereof, and the embodiments herein include all such modifications.

[0011] The embodiment discloses a method for handling thermal mitigation in a User Equipment (UE), the method comprising: estimating, by a Machine Learning (ML) module, a time required by at least one component of a Radio Resource Configuration (RRC) to operate on a wireless communication network before initiating the thermal mitigation; and predicting, by the ML module, a cumulative time till the at least one component of the RRC can operate on the wireless communication network based on the estimated time of each component of the RRC.

[0012] The embodiment discloses a Machine Learning (ML) module for handling thermal mitigation in a User Equipment (UE), the ML module comprising: a processor; and a memory; wherein the processor is configured to: estimate a time required by at least one component of a Radio Resource Configuration (RRC) to operate on a wireless communication network before initiating the thermal mitigation; and predict a cumulative time till the at least one component of the RRC can operate on the wireless communication network based on the estimated time of each component of the RRC.

[0013] The embodiments of the present invention provide the following advantages:

[0014] By utilizing a Machine Learning (ML) module, thermal mitigation in the User Equipment (UE) can be handled efficiently, enabling effective thermal control without degrading device performance.

[0015] The time required for at least one component of the Radio Resource Configuration (RRC) to operate on a wireless communication network can be accurately estimated, improving resource allocation and network responsiveness.

[0016] The cumulative time until the RRC component becomes operational on the wireless network can be predicted, facilitating optimized scheduling and system resource management.

[0017] The configuration required by the RRC component to extend the usage duration of at least one service on the wireless network can be identified, thereby enhancing user experience and ensuring service continuity.

[0018] Embodiments herein are illustrated in the accompanying drawings, throughout which like reference letters indicate corresponding parts in the various figures. The embodiments herein will be better understood from the following description with reference to the following illustratory drawings. Embodiments herein are illustrated by way of examples in the accompanying drawings, and in which:

[0019] FIG.1 is an example block diagram illustrating various units of a User Equipment (UE), for handling thermal mitigation in the wireless communication network, according to embodiments as disclosed herein;

[0020] FIG. 2 is an example block diagram illustrating various units of a processor of the UE, for handling thermal mitigation in the wireless communication network, according to embodiments as disclosed herein;

[0021] FIG. 3 is an example flow diagram illustrating the processes involved in training the Machine Learning (ML) module of the UE, according to embodiments as disclosed herein;

[0022] FIG. 4 is an example flow diagram illustrating the training the ML module and profiling at least one component of a RRC configuration of the wireless communication network, according to embodiments as disclosed herein;

[0023] FIG. 5 is an example table illustrating the sample feature set for training the ML module by predicting the change in temperature and the cumulative time required by the RRC to operate on the wireless communication network, according to embodiments as disclosed herein;

[0024] FIGs. 6A and 6B are example diagrams illustrating the training involved in the ML module to predict the change in temperature of the UE operating in the wireless communication network, according to embodiments as disclosed herein;

[0025] FIGs. 7A and 7B are example diagrams illustrating the training involved in the ML module to predict the time duration of at least one component of the RRC configuration to reach a certain temperature-based ion the current configuration, according to embodiments as disclosed herein;

[0026] FIGs. 8A and 8B are example flow diagrams illustrating a Deep Neural Network (DNN) involved in estimating the thermal cost of the radio configuration in the wireless communication network, according to embodiments as disclosed herein;

[0027] FIGs. 9A, 9B and 9C are example chart diagrams for obtaining various thermal costs for at least one component of the RRC configuration, according to embodiments as disclosed herein; and

[0028] FIGs. 10, 11 and 12 are example scenarios wherein the ML modules are trained to provide extension of time in the Fifth Generation (5G) network by optimizing the applications, and optimizing the modem protocols in the communication network, according to embodiments as disclosed herein.

[0029] The embodiments herein and the various features and advantageous details thereof are explained more fully with reference to the non-limiting embodiments that are illustrated in the accompanying drawings and detailed in the following description. Descriptions of well-known components and processing techniques are omitted so as to not unnecessarily obscure the embodiments herein. The examples used herein are intended merely to facilitate an understanding of ways in which the embodiments herein can be practiced and to further enable those of skill in the art to practice the embodiments herein. Accordingly, the examples should not be construed as limiting the scope of the embodiments herein.

[0030] For the purposes of interpreting this specification, the definitions (as defined herein) will apply and whenever appropriate the terms used in singular will also include the plural and vice versa. It is to be understood that the terminology used herein is for the purposes of describing particular embodiments only and is not intended to be limiting. The terms "comprising", "having" and "including" are to be construed as open-ended terms unless otherwise noted.

[0031] The words / phrases "exemplary", "example", "illustration", "in an instance", "and the like", "and so on", "etc.", "etcetera", "e.g.," , "i.e.," are merely used herein to mean "serving as an example, instance, or illustration." Any embodiment or implementation of the present subject matter described herein using the words / phrases "exemplary", "example", "illustration", "in an instance", "and the like", "and so on", "etc.", "etcetera", "e.g.,", "i.e.," is not necessarily to be construed as preferred or advantageous over other embodiments.

[0032] Embodiments herein may be described and illustrated in terms of blocks which carry out a described function or functions.　These blocks, which may be referred to herein as managers, units, modules, hardware components or the like, are physically implemented by analog and / or digital circuits such as logic gates, integrated circuits, microprocessors, microcontrollers, memory circuits, passive electronic components, active electronic components, optical components, hardwired circuits and the like, and may optionally be driven by a firmware. The circuits may, for example, be embodied in one or more semiconductor chips, or on substrate supports such as printed circuit boards and the like.　The circuits constituting a block may be implemented by dedicated hardware, or by a processor (e.g., one or more programmed microprocessors and associated circuitry), or by a combination of dedicated hardware to perform some functions of the block and a processor to perform other functions of the block.　Each block of the embodiments may be physically separated into two or more interacting and discrete blocks without departing from the scope of the disclosure.　Likewise, the blocks of the embodiments may be physically combined into more complex blocks without departing from the scope of the disclosure.

[0033] It should be noted that elements in the drawings are illustrated for the purposes of this description and ease of understanding and may not have necessarily been drawn to scale. For example, the flowcharts / sequence diagrams illustrate the method in terms of the steps required for understanding of aspects of the embodiments as disclosed herein. Furthermore, in terms of the construction of the device, one or more components of the device may have been represented in the drawings by conventional symbols, and the drawings may show only those specific details that are pertinent to understanding the present embodiments so as not to obscure the drawings with details that will be readily apparent to those of ordinary skill in the art having the benefit of the description herein. Furthermore, in terms of the system, one or more components / modules which comprise the system may have been represented in the drawings by conventional symbols, and the drawings may show only those specific details that are pertinent to understanding the present embodiments so as not to obscure the drawings with details that will be readily apparent to those of ordinary skill in the art having the benefit of the description herein.

[0034] The accompanying drawings are used to help easily understand various technical features and it should be understood that the embodiments presented herein are not limited by the accompanying drawings. As such, the present disclosure should be construed to extend to any modifications, equivalents, and substitutes in addition to those which are particularly set out in the accompanying drawings and the corresponding description. Usage of words such as first, second, third etc., to describe components / elements / steps is for the purposes of this description and should not be construed as sequential ordering / placement / occurrence unless specified otherwise.

[0035] Embodiments herein disclose methods and systems for handling thermal mitigation in a User Equipment (UE) Referring now to the drawings, and more particularly to FIGs. 1 through 12, where similar reference characters denote corresponding features consistently throughout the figures, there are shown at least one embodiment.

[0036] Embodiments herein disclose methods and systems for handling thermal mitigation in a User Equipment (UE). The ML module can estimate a time required by at least one component of a Radio Resource Configuration (RRC) to operate on a wireless communication network before initiating the thermal mitigation. The ML module can predict a cumulative time till the at least one component of the RRC can operate on the wireless communication network based on the estimated time of each component of the RRC.

[0037] FIG.1 is an example block diagram illustrating various units of a User Equipment (UE), for handling thermal mitigation in the wireless communication network. As illustrated in FIG. 1, the UE 100 comprises a processor 102, a communication interface unit 104, and a memory 106.

[0038] The UE 100 referred to herein may be an electronic device / user device that is being used by the user to connect, and / or interact, and / or control the operations of the plurality of other devices using a 3GPP network. Examples of the UE 100 may include, but are not limited to, a smartphone, a mobile phone, a video phone, a computer, a tablet personal computer (PC), a laptop, a wearable device, a personal digital assistant (PDA), an IoT device, or any other device that may use a 3GPP network.

[0039] The processor 102 may include one or a plurality of processors. The one or a plurality of processors may be a general-purpose processor, such as a central processing unit (CPU), an application processor (AP), or the like, a graphics-only processing unit such as a graphics processing unit (GPU), a visual processing unit (VPU), and / or an Artificial Intelligence (AI)-dedicated processor such as a neural processing unit (NPU).

[0040] The communication interface unit 104 may include one or more components, which enable the UE 100 to communicate with another device (for example, the IoT devices, the IoT server (not shown)) using the communication methods that have been supported by the communication network. The communication interface 104 may include the components such as a wired communicator, a short-range communicator, a mobile / wireless communicator, and a broadcasting receiver.

[0041] The memory 106 referred herein include at least one type of storage medium, from among a flash memory type storage medium, a hard disk type storage medium, a multi-media card micro type storage medium, a card type memory (for example, an SD or an XD memory), random-access memory (RAM), static RAM (SRAM), read-only memory (ROM), electrically erasable programmable ROM (EEPROM), programmable ROM (PROM), a magnetic memory, a magnetic disk, or an optical disk. Operations of the UE 100 described below may be performed as instructions stored in the memory 106 are executed by the processor 102 individually and / or collectively.

[0042] FIG. 2 is an example block diagram illustrating various units of a processor of the UE, for handling thermal mitigation in the wireless communication network. As illustrated in FIG. 2, the processor 102 of the UE 100 comprises a Machine Learning module 202, and a predicting module 204.

[0043] The ML module 202 can be configured to estimate a time required by at least one component of a Radio Resource Configuration (RRC) to operate on a wireless communication network before initiating the thermal mitigation. The ML module 202 can be configured to predict a cumulative time till the at least one component of the RRC can operate on the wireless communication network based on the estimated time of each component of the RRC. Based on the pre-training, the ML module 202 can predict the temperature ramp of at least one component of the RRC configuration. The components of the RRC configuration comprises a Multiple-Input, Multiple-Output (MIMO), a Reference Signal Received Power (RSRP), Transmit power (TX power), a Carrier Aggregations (CA), frequency, a bandwidth, one or more slices, and a scheduling rate. The ML module 202 may be a software-based component or hardware-based component implemented on the processor 102.

[0044] In the RRC configuration, some components (such as, bandwidth, MIMO, and the CA) significantly contribute to the heating of UE. The signal parameters of the UE 100, such as TX power, and the RSRP provides input for a thermal equation. For the provided RRC configuration, the ML module 202 can be configured to predict the temperature ramp of the UE (i.e., a linear rise in temperature and gradual cool down) until the device reaches the mitigation temperature for the Long-Term Evolution (LTE) to fall back.

[0045] The ML module 202 can be configured to learn the existing RRC configuration, which heats the UE until the mitigation temperature is reached. The ML modules 202 uses an "on-device" algorithm for the provided RRC configuration, wherein the ramp is either more or less linear. The "on-device" learning algorithm considers additional parameters such as, but not limited to, ambient temperature, device construction material and casing of the UE.

[0046] The ML module 202 can be trained with the temperatures, wherein the thermal mitigation has been reached. Also, the pre-trained model can use reinforcement learning. The ML module 202 of a neural network can be processed by the processor to estimate the time required by at least one component of the RRC configuration to operate on the wireless communication network before initiating the thermal mitigation.

[0047] The prediction module 204 can be configured to predict the threshold temperature / temperature ramp of the existing RRC configuration operating on the wireless communication network. The prediction module 204 can be configured to predict one or more parameters, which may include, but are not limited to, the amount of time left for the UE to operate on the 5G wireless communication network, the temperature of the UE after a particular period of time, the configuration required by the UE to operate on the 5G network for the particular period of time, the network slice with the highest thermal cost, CA combination, and other radio components with the highest thermal cost. Also, the prediction module can be provided with the reinforced data from a thermistor / thermal gun to check the correctness of the prediction. On reaching the maturity level (accuracy of prediction being substantially high), the prediction module 204 does not require reinforcement and can predict by itself.

[0048] The predictions provided by the prediction module 204 can be used by the UE 100, wherein the processor 102 can decide whether to release / reduce / de-activate components of the RRC configuration.

[0049] If (predicted time on 5G < imminent_thermal_threshold), the UE can take one or more corrective actions.

[0050] Hence, on determining that the UE 100 is out of imminent risk of thermal mitigation, the processor 102 can activate the original device configuration. The predictions provided by the prediction module 204 can be provided to the higher-level application optimizations.

[0051] Examples of the neural network, the ML module 202 may be, but are not limited to, an Artificial Intelligence (AI) model, a multi-class Support Vector Machine (SVM) model, a Convolutional Neural Network (CNN) model, a deep neural network (DNN), a recurrent neural network (RNN), a restricted Boltzmann Machine (RBM), a deep belief network (DBN), a bidirectional recurrent deep neural network (BRDNN), generative adversarial networks (GAN), a regression-based neural network, a deep reinforcement model (with ReLU activation), a deep Q-network, and so on. The neural network may include a plurality of nodes, which may be arranged in layers. Examples of the layers may be but are not limited to, a convolutional layer, an activation layer, an average pool layer, a max pool layer, a concatenated layer, a dropout layer, a fully connected layer, a SoftMax layer, and so on. Each layer has a plurality of weight values and performs a layer operation through calculation of a previous layer and an operation of a plurality of weights / coefficients. A topology of the layers of the neural network may vary based on the type of the respective network. In an example, the neural network may include an input layer, an output layer, and a hidden layer. The input layer receives a layer input and forwards the received layer input to the hidden layer. The hidden layer transforms the layer input received from the input layer into a representation, which may be used for generating the output in the output layer. The hidden layers extract useful / low-level features from the input, introduce non-linearity in the network and reduce a feature dimension to make the features equivalent to scale and translation. The nodes of the layers may be fully connected via edges to the nodes in adjacent layers. The input received at the nodes of the input layer may be propagated to the nodes of the output layer via an activation function that calculates the states of the nodes of each successive layer in the network based on coefficients / weights respectively associated with each of the edges connecting the layers.

[0052] The ML module 202 can be trained using at least one learning method. Examples of the learning method may be, but are not limited to, supervised learning, unsupervised learning, semi-supervised learning, reinforcement learning, regression-based learning, and so on. The ML module 202 may be neural network models in which several layers, a sequence for processing the layers, and parameters related to each layer may be known and fixed for performing the intended functions. Examples of the parameters related to each layer may be, but are not limited to, activation functions, biases, input weights, output weights, and so on, related to the layers. A function associated with the learning method may be performed through the non-volatile memory, the volatile memory, and / or the processor.

[0053] Here, being provided through learning means that, by applying the learning method to a plurality of learning data, a predefined operating rule, or the neural network, the ML module 202 of the desired characteristic is made. Functions of the neural network, the ML module 202 may be performed in the UE 100 itself in which the learning according to an embodiment is performed, and / or maybe implemented through a separate server / system.

[0054] FIG. 3 is an example flow diagram illustrating the processes involved in training the Machine Learning (ML) module of the UE. As illustrated in FIG. 3, in a first step, the RRC can be connected to the radio configuration for a particular connection of the UE on the network. The heat generated in the UE 100 can be caused due to one or more RRC connected sessions running on the high-cost configuration. The ML module can be configured to track the start temperature of the UE operating on the wireless communication network.

[0055] As illustrated in FIG. 3, in a second step, the RRC configuration is identified, wherein the RRC configuration comprises one or more RRC components, which comprise various components such as, but not limited to, a Multiple-Input, Multiple-Output (MIMO), a Reference Signal Received Power (RSRP), Transmit power (TX power), a Carrier Aggregations (CA), frequency, a bandwidth, one or more slices, and a scheduling rate. The ML module may be configured to analyze the time available to operate on the 5G network, before the mitigation temperature is reached.

[0056] As illustrated in FIG. 3, in a third step, the ML module can be configured to receive temperature input from the thermistor, existing RRC configuration, and the ambient temperature of the UE. Further, the prediction module can be configured to predict the best configuration of the RRC components to be used to avoid thermal mitigation, which is been provided to the modem. Also, the prediction module can be configured to predict the time left for the UE to operate on the 5G network.

[0057] Therefore, as illustrated in FIG. 3, the modem can be configured on the ML module, wherein the modem can optimize the RRC configuration based on the prediction provided by the prediction module, to extend the operating time of the UE on the 5G network. On continuously receiving input from the modem, the ML module can provide the best configuration of the RRC components to extend the operating time of the UE on the network. Hence, the thermal mitigation can be handled using the best configuration of the RRC components.

[0058] FIG. 4 is an example flow diagram illustrating the training the ML module and profiling at least one component of a RRC configuration of the wireless communication network. As illustrated in FIG. 4, training the ML module 202 can be initiated by training with existing data from the log database. Also, the different components of the RRC configuration and time taken to reach the mitigation can be profiled.

[0059] As illustrated, training the data can be performed from the existing field logs, the UE can be configured to profile various RRC configurations. The temperature ramp for each of the RRC configurations can be tracked. The UE can be configured to learn the temperature predictions after a pre-defined time "ΔT". Also, the training data can be obtained from lab simulation.

[0060] Further, the pre-trained model may be pushed to the UE, wherein the UE can use the pre-trained model to perform reinforcement learning. Also, the reinforcement learning can be performed based on the inputs provided by the thermistor. In this phase, the ML module 202 continues to perform "on-device" learning. The ML module 202 can predict the temperature based on the training and verifies accuracy with the reinforcement input from the thermistor. The ML module can be trained, till the ML module reaches the higher level of accuracy.

[0061] As illustrated in FIG. 4, on reaching the higher accuracy (99%), the ML module 202 shall start the predict without reinforcements from the input provided by the thermistor. Hence, on attaining the maturity, the ML module 202 can predict the time remaining on the 5G network for the UE, based on the given radio configuration. The ML module 202 can determine the RRC configuration to be used for a given time, so the UE may remain on the 5G network for the estimated period of time. Finally, the ML module 202 can estimate the time required by the RRC to operate on the wireless communication network. The ML module 202 can predict the RRC configuration to be used for a cumulative time period, for the UE to operate on the wireless communication network.

[0062] FIG. 5 is an example table illustrating the sample feature set for training the ML module by predicting the change in temperature and the cumulative time required by the RRC to operate on the wireless communication network. As illustrated in FIG. 5, the device configuration information can be collected from a device log and be converted into time-series data. The sample feature set can be represented in the table as illustrated in FIG. 5.

[0063] On extracting the time series data, the ML module 202 can be trained to predict change in the temperature for the given configuration and duration, by using a period of time (Δtime) as part of the input feature set and obtaining a change in temperature (ΔTemp) corresponding to the time as output of the ML module 202. Also, to predict expected time to reach certain temperature for a given configuration, by including change in temperature as part of the input feature set variable and obtaining the period of time corresponding to the temperature as output of the ML module 202.

[0064] FIGs. 6A and 6B are example diagrams illustrating the training involved in the ML module to predict the change in temperature of the UE operating on the wireless communication network. As illustrated in FIG. 6A, a sample deep neural network is designed to train the ML module 202 to predict the change in temperature value based on the current configuration of the RRC for the particular amount of time.

[0065] As illustrated in FIG. 6A, "X" represents the input values of the ML module 202 , "Y" is the final value of the ML module 202. Wirepresents the weights between the nodes of two layers of the ML module 202. The weight of each neural connection indicates strongness between the connection. Throughout the training of the ML module 202, the weight of the node changes.

[0066] "Bi" represents bias, which is an additional value present for each neuron. Bias is essentially a weight without an input term. It is used for having an extra bit of adjustability which is not dependant on a previous layer.

[0067] "Hi" represents the hidden layers, which are the intermediate layers between the input and output layer, the hidden layers can process the data by applying complex non-linear functions. Also, the linear relationship is not sufficient to capture the complexity of the task.

[0068] F() represents an activation function, which decides whether a neuron should be activated or not. The purpose of the activation function is to introduce non-linearity into the output of the neuron.

[0069] As illustrated in FIG. 6A, the input layer(x) comprises a scheduling rate which includes RSRP, Reference Signal Received Quality (RSRQ), Signal to noise ratio (SNR), TXPWR, bandwidth, MIMO, and frequency, temperature received from the thermistor, average ULTPUT, average DLTPUT, active Public Data Network (PDN), Protocol Data Unit (PDU), slice, and Δtime.

[0070] As illustrated in FIG. 6A, the hidden layers H1, and H2, along the output layer Y can be provided with the weights such as W1 (5*4), W2 (4*4), W3 (4*1) respectively.

[0071] As illustrated in FIG. 6B, the temperature change can be determined using the following formula, temperature change = f(W3 * f(W2 * f(W1 * X + B1) + B2) + B3), wherein the output layer (Y), hidden layer 2, hidden layer 1, and input layer (X) is used to estimate the temperature change using the ML module 202.

[0072] FIGs. 7A and 7B are example diagrams illustrating the training involved in the ML module to predict the time duration of at least one component of the RRC configuration to reach a certain temperature-based on the current configuration. As illustrated in FIG. 7A, the sample deep neural network can be designed to train the ML module 202, to predict the time duration to reach a certain temperature value on the basis of the current RRC configuration.

[0073] As illustrated in FIG. 7A, "X" represents the input values of the ML module 202. Y is the final value of the model. "Wi" represents the weights between the nodes of two layers of the ML module 202. The weight of each neural connection indicates the strongness between the connection. Throughout the training, the weights between the node can change.

[0074] "Bi" represents the bias, which is an additional value present for each neuron. Bias is essentially a weight without an input term. It is used for having an extra bit of adjustability which is not dependant on the previous layer.

[0075] "Hi" represents the hidden layers, which are the intermediate layers between the input and output layer. The hidden layers can process the data by applying complex non-linear functions to the layers, as a linear relationship is not sufficient to capture the complexity of the task.

[0076] f( ) is called an activation function which decides whether a neuron should be activated or not. The purpose of the activation function is to introduce non-linearity into the output of a neuron.

[0077] As illustrated in FIG. 7A, the input layer comprises a scheduling rate which includes RSRP, Reference Signal Received Quality (RSRQ), Signal to noise ratio (SNR), TXPWR, bandwidth, MIMO, and frequency, temperature received from the thermistor, average ULTPUT, average DLTPUT, active Public Data Network (PDN), Protocol Data Unit (PDU), slice, and Δtemperature.

[0078] As illustrated in FIG. 7A, the hidden layers H1, and H2, along the output layer Y can be provided with the weights such as W1 (5*4), W2 (4*4), W3 (4*1) respectively.

[0079] As illustrated in FIG. 7B, the time duration can be determined using the following formula, time duration= f(W3 * f(W2 * f(W1 * X + B1) + B2) + B3), wherein the output layer (Y), hidden layer H2, hidden layer H1, and input layer (X) is used to estimate the time duration using the ML module 202.

[0080] FIGs. 8A and 8B are example flow diagrams illustrating a Deep Neural Network (DNN) involved in estimating the thermal cost of the radio configuration in the wireless communication network.

[0081] In an embodiment, the Optimal Configuration Identifier Module (OCIM)can identify the radio components, which can be employed in the RRC configuration for a given expected time on the 5G network. Hence, taking the time as input, the module may run through various possible configurations, feed each configuration as an input to the ML module 202 to obtain the time prediction and validate if the predicted time matches with the expected time on 5G network. Also, multiple models, which may include, but are not limited to Long short-term memory (LSTM), Deep Neural Network (DNN), and so on. The multiple models can be implemented to create OCIM, to derive and predict change in the temperature for the given radio configuration.

[0082] The information in the RRC configuration message as a feature set can be used for the regression model. The key components of the feature set, which may include, but are limited to, the frequency, the bandwidth, the MIMO, signals such as RSRP, RSRQ, SNR etc., Tx Power, scheduling rate, Tput: Avg. DL Tput, Avg. UL TPUT, temperature from the thermistor sensor, active application / slice / PDN.

[0083] In an embodiment herein, the OCIM can expose APIs used to obtain configuration for the given period of time. OCIM_Extend5GTime(t); → will run through its configurations and suggest the best one suited for this extension of 't' minutes. OCIM_ObtainConfiguration(t); →API will return the best possible radio configuration to sustain on the 5G network for 't' minutes.

[0084] As illustrated in FIG. 8A, the OCIM modules can be configured to use DNN to identify one or more possible radio configurations. The radio configurations such as MIMO, CA and the bandwidths can be configured to optimize the cost using the OCIM modules.

[0085] For an instance, the radio configuration of MIMO with the 1*1, the 1CA using the 50 MHz bandwidth may provide lowest thermal cost with the maximum time on the 5G network. In another instance, wherein the radio configuration with the MIMO 4*4, the 5CA using the 800MHz bandwidth can provide higher thermal cost with the minimum time on the 5G network.

[0086] As illustrated in FIG. 8B, the radio configuration can be provided as the input to the ML module 202, wherein the time required to remain on the 5G network is provided to the time prediction. On determining that the predicted time is greater than or equal to the expected 5G time, the UAI is initiated wherein the profile with the network using UAI is negotiated. On determining that the predicted time is less than the expected 5G time, the OCIM is initiated. The OCIM can further optimize the RRC configuration to provide the lower cost.

[0087] FIGs. 9A, 9B and 9C are example chart diagrams for obtaining various thermal costs for at least one component of the RRC configuration, according to embodiments as disclosed herein.

[0088] In an embodiment, thermal profiling of the radio configuration can be performed by estimating the thermal cost of radio configuration component, thermal cost of the network slice, and the thermal cost of the CA combination.

[0089] The UE can associate the thermal cost to each radio configuration component based on the effect on device temperature ramp. High thermal cost of the component may signify the faster temperature rise as the component being active, and low cost of a component may signify the lesser temperature rise. In an example, employing 4x4 MIMO component of high thermal cost, lower aggregated bandwidth implies lower thermal cost to the device.

[0090] The UE may have an initial thermal cost of each component which are already stored, which can be derived from the extensive simulation testing under controlled environment. The UE will update the cost related to each component in real time, based on the real time usage pattern of the device, real time performance etc. The UE can derive the thermal cost of the current radio configuration by adding the individual costs of each of the radio configuration components.

[0091] The thermal cost for the network slice can be estimated based on the respective radio configurations, associated with a Slice ID or PDU session ID. On estimating the cost for radio configuration, it can be associated with the PDU session ID. By combining the radio configuration with the cost, the Slice / PDU session IDs with the high thermal cost to the UE can be estimated. Therefore, the high thermal cost with the slice / PDU session ID can be shared with the Application (AP). Further, the requesting applications can be regulated, wherein the slice used by the UE is in a thermally constrained condition.

[0092] The thermal cost for one or more CA combinations can be estimated, wherein the UE can support various CA combinations. Each of the RRC configuration comprises at least one CA combination, that is supported by the UE. By profiling the configuration, the UE can estimate the thermal cost of the CA combination in use. On estimating the cost, the modem can restrict the combinations which are of high cost to the UE under thermally constrained conditions.

[0093] Each radio configuration comprises a pre-defined set of radio parameters such as MIMO, CA, frequency and Bandwidth. The signal conditions of the UE, such as, but not limited to, TX power and RSRP of the radio signal are the variables causing the thermal ramp. The thermal temperature ramp of the device is a non-linear graph which can be expressed mathematically as a non-linear equation.

[0094] a1 f(x1) + a2 f(x2) + ... + an f(xn) = Temperature, wherein the functions f(x1), f(x2) .... f(xn) are the individual thermal functions of the components such as MIMO, CA. Frequency and Bandwidth can contribute to the thermal ramp. A1, a2 ... an are the co-efficient for the above-mentioned functions. Thereby, keeping all other parameters constant by means of a lab simulation and solving for one particular function will give the individual cost of that component towards the thermal ramp. Also, cost isolation for MIMO can be estimated by keeping other parameters constant.

[0095] As illustrated in FIG. 9A, on employing 4x4 MIMO, 1CA, 100 MHz Bandwidth, frequency f1, there may be a high thermal cost due to the higher MIMO. Thereby, providing a steep ramp with the time along the temperature.

[0096] As illustrated in FIG. 9B, on employing 2x2 MIMO, 1CA, 100 MHz Bandwidth, frequency f1, there may be a moderate thermal cost due to 2x2 MIMO. Thereby, providing a slow ramp with the time along the temperature.

[0097] As illustrated in FIG. 9C, on employing 1x1 MIMO, 1CA, 100 MHz Bandwidth, frequency f1, there may be a low thermal cost due to 1x1 MIMO. Thereby, providing the slowest ramp with the time along the temperature.

[0098] FIGs. 10, 11 and 12 are example scenarios wherein the ML modules are trained to provide extension of time in the Fifth Generation (5G) network, by optimizing the applications, and optimizing the modem protocols in the communication network.

[0099] In an example scenario (as illustrated in FIG. 10), assume that the user is using a 5G slicing service. Based on the ML module 202 prediction, the device can remain for ten more minutes on the 5G network before LTE fall back occurs. The modem can use OCIM to extend the operating time on the 5G network. The OCIM can run through the trained module and output a few radio configurations that will extend the remaining 5G network time from ten to fifteen minutes. The thermal profiling which was done by the algorithm for slice, CA combination and the components can estimate the components that can be cut down, so as to meet the 5G target time for fifteen minutes. Based on the inputs, the modem can use the UAI (UE assistance information) procedure to negotiate the required configuration with the network.

[0100] In another example scenario (as illustrated in FIG. 11), assume that application wants to receive the prediction of estimated 5G time from the ML module. Periodically, the ML module will respond the call backs with the estimated time on the 5G network, which can be used by the applications to optimize their operations. For example, if the estimated 5G network time is less and LTE fall back is expected within a stipulated time, it is better to scale down the Transmission Control Protocol (TCP) window size to prevent an abrupt shrinking of the window (due to LTE fall back). Abrupt window shrinking is bad for the TPUT stream, as it can consume some time before the window scales up again.

[0101] In an example scenario (as illustrated in FIG. 12), thermal profiling the radio configuration and estimating the thermal cost for each component has its advantages. Knowing the costliest radio component can help in deciding the configuration which is to cut down in the thermally constrained scenario. Similarly, knowing the highest thermal cost of the CA combination can help in deciding the configuration which is to cut down in the thermally constrained scenario. The most expensive CA combinations can be disabled by the UE and indicated to the network by the UE capability update. Knowing the highest thermal cost of the network slice, the slice can be disabled in the timely manner, as the device is approaching the LTE fall back deadline. In all of the above situations, the UE will check beforehand, whether the requested reduced configuration is sufficient to meet the current throughput need of the UE.

[0102] The embodiments disclosed herein can be implemented through at least one software program running on at least one hardware device and performing network management functions to control the network elements. The elements include blocks which can be at least one of a hardware device, or a combination of hardware device and software module.

[0103] The embodiments disclosed herein describe a circuit for performing analog calibration for a scalable multi-voltage memory interface driver. Therefore, it is understood that the scope of the protection is extended to such a program and in addition to a computer readable means having a message therein, such computer readable storage means contain program code means for implementation of one or more steps of the method, when the program runs on a server or mobile device or any suitable programmable device. The method is implemented in at least one embodiment through or together with a software program written in e.g., Very high-speed integrated circuit Hardware Description Language (VHDL) another programming language, or implemented by one or more VHDL or several software modules being executed on at least one hardware device. The hardware device can be any kind of portable device that can be programmed. The device may also include means which could be e.g., hardware means like e.g., an ASIC, or a combination of hardware and software means, e.g., an ASIC and an FPGA, or at least one microprocessor and at least one memory with software modules located therein. The method embodiments described herein could be implemented partly in hardware and partly in software. Alternatively, the invention may be implemented on different hardware devices, e.g., using a plurality of CPUs.

[0104] The foregoing description of the specific embodiments will so fully reveal the general nature of the embodiments herein that others can, by applying current knowledge, readily modify and / or adapt for various applications such specific embodiments without departing from the generic concept, and, therefore, such adaptations and modifications should and are intended to be comprehended within the meaning and range of equivalents of the disclosed embodiments. It is to be understood that the phraseology or terminology employed herein is for the purpose of description and not of limitation. Therefore, while the embodiments herein have been described in terms of embodiments and examples, those skilled in the art will recognize that the embodiments and examples disclosed herein can be practiced with modification within the scope of the embodiments as described herein.

Claims

A method for handling thermal mitigation in a User Equipment (UE) (100), the method comprising:estimating, by a Machine Learning (ML) module (202), a time required by at least one component of a Radio Resource Configuration (RRC) to operate on a wireless communication network before initiating the thermal mitigation; anddetermining, by the ML module (202), a cumulative time till the at least one component of the RRC can operate on the wireless communication network based on the estimated time of each component of the RRC.The method as claimed in claim 1, wherein the at least one component of the RRC is a Multiple-Input, Multiple-Output (MIMO), a Reference Signal Received Power (RSRP), Transmit power (TX power), a Carrier Aggregations (CA), frequency, a bandwidth, one or more slices, and a scheduling rate.The method as claimed in claim 1, wherein the method comprises identifying, by an Optimal Configuration Identifier Module (OCIM), a configuration required for the at least one component of the RRC, to extend time for using at least one service on the wireless communication network.The method as claimed in claim 1, wherein the method comprises predicting, by the ML module (202), a temperature of the UE using at least one service on the wireless communication network, based on the configuration of at least one component of the RRC, in which the thermal mitigation has been initiated.The method as claimed in claim 1, wherein the method comprises:profiling, by the ML module (202), at least one component of the RRC configuration to associate a thermal cost, wherein the thermal cost is amount of time consumed by the UE (100) to remain in the wireless communication network, on continuously using the configured RRC; andpredicting, by the ML module (202), amount of time that the RRC configuration remains in the wireless communication network, before the thermal mitigation is initiated and Fifth Generation network (5G) is disabled.The method as claimed in claim 1, wherein profiling, by the ML module (202), at least one radio configuration and associating the thermal cost for at least one component of the RRC configuration that is responsible for generating heat in the UE (100).An user equipment (100) comprising:a processor (102); anda memory (106) storing instructions;wherein the instructions, when executed by the processor (102), cause the user equipment to:control a machine learning module (202) implemented on the processor (102) to estimate a time required by at least one component of a Radio Resource Configuration (RRC) to operate on a wireless communication network before initiating the thermal mitigation; andcontrol the machine learning module (202) to determine a cumulative time till the at least one component of the RRC can operate on the wireless communication network based on the estimated time of each component of the RRC.The user equipment of claim 7, wherein the at least one component of the RRC is a Multiple-Input, Multiple-Output (MIMO), a Reference Signal Received Power (RSRP), Transmit power (TX power), a Carrier Aggregations (CA), frequency, a bandwidth, one or more slices, and a scheduling rate.The user equipment of claim 7, wherein instructions, when executed by the processor, cause the user equipment to identify, by an Optimal Configuration Identifier Module (OCIM), a configuration required for the at least one component of the RRC, to extend time for using at least one service on the wireless communication network.The user equipment of claim 7, wherein the instructions, when executed by the processor, cause the user equipment to predict a temperature of the UE using at least one service on the wireless communication network, based on the configuration of at least one component of the RRC, in which the thermal mitigation has been initiated.The user equipment of claim 7, wherein the instructions, when executed by the processor, cause the user equipment to:profile at least one component of the RRC configuration to associate a thermal cost, wherein the thermal cost is amount of time consumed by the UE to remain in the wireless communication network, on continuously using the configured RRC; andpredict amount of time that the RRC configuration remains in the wireless communication network, before the thermal mitigation is initiated and Fifth Generation network (5G) is disabled.The user equipment of claim 7, wherein profiling by the user equipment at least one radio configuration and associating the thermal cost for at least one component of the RRC configuration that is responsible for generating heat in the user equipment.

Images