MODEM Data Sensing and Recovery Method for Data Disconnection in Mobile Communication Systems

Abstract

The present invention discloses a method for MODEM data perception and recovery of data interruption in a mobile phone communication system, including: (1) training a random forest algorithm binary classifier according to the actual network data and / or simulation data of each layer of the MODEM, and configuring the random forest algorithm binary classifier in a data interruption recovery module; (2) obtaining the data of each layer of the current MODEM and using it as the input of the random forest algorithm binary classifier, and determining whether data interruption occurs by the random forest algorithm binary classifier; if so, proceed to step (3), otherwise continue step (2); (3) triggering a data interruption recovery program. The present invention can automatically perceive data interruption and recover data by obtaining the data of each layer of the MODEM and determining whether data interruption occurs by the random forest algorithm binary classifier, thereby triggering a data interruption recovery program. The present invention can automatically perceive data interruption and recover data, improving the user experience in the data scenario.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communication, and particularly relates to a method for data perception and recovery of a MODEM (modem) for data interruption in a mobile phone communication system. Background Art

[0002] In the field of wireless communication, especially in the LTE (Long Term Evolution) communication field, the link reliability between a UE (User Equipment) and a base station is an important link to ensure the stability of user services.

[0003] The behavior of a UE (User Equipment) is controlled by network scheduling. In the case of a network handover failure, if the UE perceives slowly or does not perceive the handover failure at all, the UE data will be interrupted, and the user cannot normally use a browser, WeChat, etc. for data interaction. Currently, the prior art lacks effective means to automatically perceive data interruption and recover the data. Summary of the Invention

[0004] The present invention provides a method for data perception and recovery of a MODEM for data interruption in a mobile phone communication system, which can automatically perceive data interruption and recover the data. The present invention is realized by the following technical solutions:

[0005] A method for data perception and recovery of a MODEM for data interruption in a mobile phone communication system, characterized by comprising:

[0006] (1) Training a random forest algorithm binary classifier according to the actual network data and / or simulation data of each layer of the MODEM, and configuring the random forest algorithm binary classifier in a data interruption recovery module;

[0007] (2) Obtaining the data of each layer of the current MODEM and using it as the input of the random forest algorithm binary classifier, and determining whether data interruption occurs by the random forest algorithm binary classifier; if yes, go to step (3), otherwise continue step (2);

[0008] (3) Triggering a data interruption recovery program.

[0009] Specifically, in step (2), the MODEM first collects the network data of each current layer, and after the collection is completed, sends a data recovery indication message to the RILD process of the RIL layer of the mobile phone communication system to notify the RILD process to prepare to obtain the data of each current layer of the MODEM.

[0010] Specifically, in step (2), when the data interruption recovery module is started, it monitors whether the MODEM reports the data recovery indication information to the RILD process of the mobile communication system. If so, it obtains the data of each layer of the current MODEM through the RILD process and uses it as the input of the random forest algorithm binary classifier; otherwise, it does nothing.

[0011] Specifically, the data interruption recovery module is deployed in the RILD process of the mobile communication system, and is started with the RILD process and exits with the RILD process.

[0012] More specifically, in step (2), after the RILD process obtains the data of each layer of the current MODEM, a data recovery event is generated inside the RILD process, and at the same time, the data of each layer of the current MODEM is input into the random forest algorithm binary classifier.

[0013] Specifically, the data interruption recovery module is deployed in the Telephony process of the Telephony module of the mobile communication system, and is started with the Telephony process and exits with the Telephony process.

[0014] More specifically, in step (2), after the RILD process obtains the data of each layer of the current MODEM, it forwards it to the Telephony process. A data recovery event is generated inside the Telephony process, and at the same time, the data of each layer of the current MODEM is input into the random forest algorithm binary classifier.

[0015] Specifically, the actual network data and / or simulation data of each layer of the MODEM include the following items and sorted data: EMM status, 5GMM status, 4G-RRC status, 5G-RRC status, LTE-CQI, LTE-MCS, LTE-TBS, LTE-IBLER, LTE-MAC-Padding, NR-CQI, NR-MCS, NR-TBS, LTE-IBLER, NR-MAC-Padding, NR-TBS, NR-IBLER, NR-MAC-Pading, LTE-CS-RSRP, LTE-SINR, NR-SS-RSRP, NR-SINR.

[0016] Specifically, when obtaining the data of each layer of the current MODEM, if the UE is in the NSA mode, the data of the items and their sorting of the LTE L1 and NRL1 layers, LTE MAC and NR MAC layers, LTE RRC and NR RRC layers, and LTE NAS and NR NAS layers are obtained; if the UE is in the SA mode, the data of the items and their sorting of the NR L1 layer, NR MAC layer, NR RRC layer, and NR NAS layer are obtained; the data of the items that do not exist in the corresponding mode are replaced with invalid values.

[0017] Specifically, triggering the data disconnection recovery program in step (3) specifically includes: the RIL layer sends a data recovery instruction request to the MODEM, and the MODEM side sends OTA signaling for detachment, reattachment, and PDU session establishment to the network.

[0018] The beneficial effects of the present invention are as follows: by obtaining the data of each layer of the MODEM and using the binary classifier of the random forest algorithm to determine whether data disconnection occurs, and then triggering the data disconnection recovery program. The present invention can automatically sense data disconnection and recover the data, improving the user experience in the data scenario. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is the main flowchart of the MODEM data perception and recovery method for data disconnection in the mobile communication system provided in Embodiment 1 of the present invention.

[0020] Figure 2 It is a schematic diagram of obtaining the data of each layer of the current MODEM in the MODEM data perception and recovery method for data disconnection in the mobile communication system provided in Embodiment 1 of the present invention.

[0021] Figure 3 It is a schematic diagram of the position of DSRM in the system and the data flow and control flow in the MODEM data perception and recovery method for data disconnection in the mobile communication system provided in Embodiment 1 of the present invention.

[0022] Figure 4 It is a schematic diagram of the position of DSRM in the system and the data flow and control flow in the MODEM data perception and recovery method for data disconnection in the mobile communication system provided in Embodiment 1 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] Embodiment 1

[0024] Combined with Figure 1 As shown, this embodiment provides a MODEM data perception and recovery method for data disconnection in a mobile communication system, including:

[0025] (1) Train a random forest algorithm binary classifier based on the actual network data and / or simulation data of each layer of the MODEM, and configure the random forest algorithm binary classifier in a Data Stall Recovery Module (DSRM);

[0026] (2) Obtain the data of each layer of the current MODEM and use it as the input of the random forest algorithm binary classifier. Determine whether there is a data stream interruption by the random forest algorithm binary classifier. If yes, go to step (3); otherwise, continue with step (2);

[0027] (3) Trigger the data stream interruption recovery program.

[0028] Among them, when training the random forest algorithm binary classifier, the actual network data and / or simulation data of each layer of the MODEM include the following items and sorted data: EMM status, 5GMM status, 4G-RRC status, 5G-RRC status, LTE-CQI, LTE-MCS, LTE-TBS, LTE-IBLER, LTE-MAC-Padding, NR-CQI, NR-MCS, NR-TBS, LTE-IBLER, NR-MAC-Padding, NR-TBS, NR-IBLER, NR-MAC-Pading, LTE-CS-RSRP, LTE-SINR, NR-SS-RSRP, NR-SINR. See Table 1 below for details.

[0029]

[0030]

[0031] Table 1

[0032] Based on the above items and sorted data, set the training data format as: (X_vector; predict_y). Among them, X_vector is defined as: (EMM-status, 5GMM-status, LTE-RRC-status, 5G-RRC-status, LTE-CQI, LTE-MCS, LTE-TBS, LTE-IBLER, LTE-MAC-Padding, NR-CQI, NR-MCS, NR-TBS, NR-IBLER, NR-MAC-pading, LTE-CS-RSRP, LTE-SINR, NR-SS-RSRP, NR-SINR); Y takes values of Yes or No.

[0033] The training of the binary classifier of the random forest algorithm is based on known techniques in data mining and machine learning. Its discrimination criterion is as follows: the output is a probability value. If this probability value > 0.5, it is judged as "yes", and then the data stream interruption recovery program is triggered; if the probability value ≤ 0.5, it is judged as "no", and then the data stream interruption recovery program is not triggered.

[0034] Combined with Figure 2 As shown, when obtaining the data of each layer of the current MODEM, if the UE is in the NSA mode, the data of the items and their sorting of the LTE L1 and NR L1 layers, LTE MAC and NR MAC layers, LTE RRC and NR RRC layers, and LTE NAS and NR NAS layers are obtained; if the UE is in the SA mode, the data of the items and their sorting of the NR L1 layer, NR MAC layer, NR RRC layer, and NR NAS layer are obtained. The data of the items that do not exist in the corresponding mode are replaced with the invalid value NULL.

[0035] For the two modes of the UE (NSA mode and SA mode), examples of specific values and prediction results are as follows:

[0036] NSA mode example 1: (EMM-Registered, 5GMM-Registered, LTE-RRC-Connected, 5G-RRC-Connected, LTE-CQI 13, LTE-MCS 25, LTE-TBS1650, LTE-IBLER 5%, LTE-MAC-Padding3%, NR-CQI 13, NR-MCS23, NR-TBS 2400, NR-IBLER 5%, NR-MAC-pading 3%, LTE-CS-RSRP-90, LTE-SINR 15, NR-SS-RSRP-80, NR-SINR 18; No).

[0037] NSA mode example 2: (EMM-Registered, 5GMM-Registered, LTE-RRC-Connected, 5G-RRC-Connected, LTE-CQI 5, LTE-MCS 7, LTE-TBS1650, LTE-IBLER 5%, LTE-MAC-Padding20%, NR-CQI 13, NR-MCS23, NR-TBS 2600, NR-IBLER 5%, NR-MAC-pading 13%, LTE-CS-RSRP-90, LTE-SINR 15, NR-SS-RSRP-110, NR-SINR 18; Yes).

[0038] SA mode example 1: (NULL, 5GMM-Registered, NULL, 5G-RRC-Connected, NULL, NULL, NULL, NULL, NULL, NR-CQI 13, NR-MCS 25, NR-TBS1650, NR-IBLER 1%, NR-MAC-pading 8%, NULL, NULL, SS-RSRP-80, NR-SINR 3; No).

[0039] SA mode example 2: (NULL, 5GMM-Registered, NULL, 5G-RRC-Connected, NULL, NULL, NULL, NULL, NULL, NR-CQI 7, NR-MCS 11, NR-TBS1650, NR-IBLER 1%, NR-MAC-pading 10%, NULL, NULL, SS-RSRP-80, NR-SINR 6; Yes).

[0040] In step (2) of this embodiment, the MODEM first collects the current network data of each layer. After the collection is completed, it sends a data recovery indication message (RIL_MODEM_DATA_RECOVERY_INDICATION) to the RILD process of the RIL layer (Radio Interface Layer) of the mobile communication system to notify the RILD process to prepare to obtain the current data of each layer of the MODEM. Correspondingly, when the data interruption recovery module is started, it listens whether the MODEM reports the data recovery indication message to the RILD process of the mobile communication system. If so, it obtains the current data of each layer of the MODEM through the RILD process and uses it as the input of the random forest algorithm splitter. Otherwise, it does nothing.

[0041] Combined Figure 3 As shown, the data interruption recovery module in this embodiment is deployed in the RILD process of the mobile communication system, or in other words, the data interruption recovery module (DSRM) is used as a service of the RIL (radio interface layer) module and starts and exits with the start and exit of the RILD (Radio interface layer Daemon) process. In step (2), after the RILD process obtains the current data of each layer of the MODEM, a data recovery event (RIL_MODEM_DATA_RECOVERY_EVENT) is generated inside the RILD process, and at the same time, the current data of each layer of the MODEM is input into the random forest algorithm splitter.

[0042] In addition, triggering the data disconnection recovery program in step (3) specifically includes: the RIL layer sends a data recovery instruction request to the MODEM, and the MODEM side sends OTA signaling for detachment, reattachment, and PDU session establishment to the network.

[0043] If the UE is in the NSA mode, the RIL sends an RIL_Mode_NSA_DATA_RECOVERY_REQ request to the MODEM. The specific actions on the MODEM side are to send the following signaling to the network:

[0044] 1. Send a De-attach OTA message to the Network

[0045] 2. Then send a Re-attach OTA message to the Network

[0046] 3. The UE sends a PDN connection setup OTA message to the Network

[0047] If the UE is in the SA mode, the RIL sends an RIL_Mode_SA_DATA_RECOVERY_REQ request to the MODEM. The specific actions on the MODEM side are to send signaling to the network:

[0048] 1. Send a De-registration OTA message to the Network

[0049] 2. Then send a registration OTA message to the Network

[0050] 3. The UE sends a PDU session establish OTA message to the Network

[0051] Embodiment 2

[0052] Combined with Figure 4As shown in the figure, the difference between the second embodiment and the first embodiment is that: the data interruption recovery module is deployed in the Telephony process of the Telephony module of the mobile phone communication system, starts with the start of the Telephony process, and exits with the exit of the Telephony process. Correspondingly, in step (2), after the RILD process obtains the data of each layer of the current MODEM, it encapsulates it into a Parcel data packet and forwards it to the Telephony process. A data recovery event is generated inside the Telephony process, and at the same time, the data of each layer of the current MODEM is input into the random forest algorithm binary classifier for discrimination. In addition, when the discrimination output at the Telephony end is Yes, a data disconnection recovery program is triggered, and a RILJ_Mode_NSA_DATA_RECOVERY_REQ / RILJ_Mode_SA_DATA_RECOVERY_REQ request is sent to the RILD process. The RILD process sends a data recovery instruction request to the MODEM, and the MODEM side then sends OTA signaling for detachment, reattachment, and PDU session establishment to the network.

[0053] The above specific implementation manners are only for full disclosure and not for limiting the present invention. Replacements of equivalent technical features that can be obtained without creative labor based on the creative concept of the present invention should be regarded as the scope disclosed in this application.

Claims

1. A method for MODEM data perception and recovery of data interruption in a mobile phone communication system, characterized in that Including: (1) Train a random forest algorithm binary classifier according to the actual network data and / or simulation data of each layer of the MODEM, and configure the random forest algorithm binary classifier in a data interruption recovery module of a mobile communication system; (2) The MODEM first collects the current network data of each layer. After the collection is completed, it sends a data recovery indication message to the RILD process in the RIL layer of the mobile communication system, notifying the RILD process to prepare to obtain the current data of each layer of the MODEM. The mobile communication system obtains the current data of each layer of the MODEM and uses it as the input of the random forest algorithm binary classifier. The output of the random forest algorithm binary classifier is a probability value. If the probability value > 0.5, it is determined that a data disconnection occurs and step (3) is entered; otherwise, step (2) is continued; (3) Trigger a data disconnection recovery program, specifically including: the RIL layer of the mobile communication system sends a data recovery instruction request to the MODEM, and the MODEM side sends OTA signaling for detachment, reattachment, and PDU session establishment to the network.

2. The MODEM data perception and recovery method for the data disconnection of the mobile phone communication system according to claim 1, characterized in that, In step (2), when the data interruption recovery module is started, it listens whether the MODEM reports the data recovery indication message to the RILD process of the mobile communication system. If so, it obtains the current data of each layer of the MODEM through the RILD process and uses it as the input of the random forest algorithm binary classifier; otherwise, no action is taken.

3. The MODEM data perception and recovery method for the disconnection of mobile communication system data according to claim 2, characterized in that, The data interruption recovery module is deployed in the RILD process of the mobile communication system, and is started with the start of the RILD process and exits with the exit of the RILD process.

4. The MODEM data perception and recovery method for data interruption in the mobile phone communication system according to claim 3, characterized in that, In step (2), after the RILD process obtains the current data of each layer of the MODEM, a data recovery event is generated inside the RILD process, and at the same time, the current data of each layer of the MODEM is input into the random forest algorithm binary classifier.

5. The MODEM data perception and recovery method for data interruption in the mobile communication system according to claim 2, characterized in that, The data interruption recovery module is deployed in the Telephony process of the Telephony module of the mobile communication system, and is started with the start of the Telephony process and exits with the exit of the Telephony process.

6. The MODEM data perception and recovery method for data interruption in a mobile phone communication system according to claim 5, characterized in that In step (2), after the RILD process obtains the current data of each layer of the MODEM, it forwards it to the Telephony process. A data recovery event is generated inside the Telephony process, and at the same time, the current data of each layer of the MODEM is input into the random forest algorithm binary classifier.

7. The MODEM data perception and recovery method for the disconnection of mobile communication system data according to claim 1, characterized in that, The actual network data and / or simulation data of each layer of the MODEM include the following items and sorted data: EMM status, 5GMM status, 4G-RRC status, 5G-RRC status, LTE-CQI, LTE-MCS, LTE-TBS, LTE-IBLER, LTE-MAC-Padding, NR-CQI, NR-MCS, NR-TBS, LTE-IBLER, NR-MAC-Padding, NR-TBS, NR-IBLER, NR-MAC-Pading, LTE-CS-RSRP, LTE-SINR, NR-SS-RSRP, NR-SINR.

8. The MODEM data perception and recovery method for the data disconnection of the mobile phone communication system according to claim 7, characterized in that, When obtaining the data of each layer of the current MODEM, if the UE is in the NSA mode, obtain the data of the items and their sorting of the LTE L1 and NR L1 layers, LTE MAC and NR MAC layers, LTE RRC and NR RRC layers, and LTE NAS and NR NAS layers; if the UE is in the SA mode, obtain the data of the items and their sorting of the NR L1 layer, NR MAC layer, NR RRC layer, and NR NAS layer; the data of the items that do not exist in the corresponding mode are replaced with invalid values.

Images