Method for reporting SRS from clusters of mobile devices

Abstract

The present invention discloses a method for reporting SRS from clusters of mobile devices, comprising: identifying an area with high mobile devices density; receiving mobile devices characteristics from mobile devices; analyzing the mobile devices characteristics; arranging mobile devices, with similar mobile device characteristics, into clusters with similar channel conditions; using a machine learning algorithm to optimize a boundary between each cluster; selecting, using the machine learning algorithm, the most suitable mobile devices to report SRS; estimating channel conditions of each cluster.

Description

"METHOD FOR REPORTING SRS FROM CLUSTERS OF MOBILE DEVICES"TECHNICAL FIELD

[0001] The present invention discloses a method for sounding reference signal "SRS" report. More specifically, the present invention discloses a method for SRS report that creates clusters for close devices.BACKGROUND OF THE INVENTION

[0002] The evolution of mobile telecommunications has witnessed substantial growth in recent decades, both in the number of users, devices, and applications, boosting a nonceasing demand for larger bandwidths, lower latency, and more reliable and efficient communications on newer Radio Access Technologies. In this scenario, SRS (Sounding Reference Signals) are vital in managing and optimizing data transmission traffic.

[0003] SRS were designed in the first draft of 4G LTE on 3GPP release 8, and they are an intrinsic part of this Radio Access Technology "RAT", also being used for 5G systems with some improvements. They are reference signals that allow for the dynamic evaluation of the radio channel between the mobile station / user equipment "MS / UE" and the base station "BS". Their primary goal is to provide essential information regarding the radio channel's conditions, including bandwidth, delay, and quality, among other relevant parameters. When an MS / UE transmits the SRS, since the transmitted signal is known, the base station can perform a detailed evaluation of the radio channel, allowing the network to optimize and fine-tune the transmission configurations. This feedback capability is crucial for optimizing the spectral efficiency of the transmission,enabling the radio system to achieve the quality of service (QoS) demanded by upper-layer services, assuring more reliable and efficient communication. Since SRS resources are positioned at specific intervals in the frequency domain, the network may interleave (multiplex) multiple SRS along the frequency domain, occupying the same OFDM symbols in the radio frame.

[0004] SRS implementation has some challenges to address to provide the desired measurements. Interference management is one of the difficulties that MS / UE / BS face when measuring the radio channel, impacting the measurements performed and compromising the service provided by SRS. The efficient allocation of spectrum resources is another challenge since no data transmission is allowed in the resources utilized by SRS, resulting in lower throughput during the utilization of SRS. Adequate synchronization methods are also required for the system to take full advantage of SRS benefits in 4G / 5G networks.

[0005] For a deep understanding of SRS and their impact on mobile networks, it is crucial to analyze the technique's architecture and operation. SRS operate in many layers, playing a vital role.

[0006] In the physical layer "PHY", SRS are generated by the user equipment "UE" and transmitted to the base station. These reference signals are modulated on many different sub-carriers of different frequencies, allowing for the radio channel analysis in many different chunks of the electromagnetic spectrum of the radio carrier. The gathering and interpretation of the SRS signals provide meaningful information for the dynamic adaptation of themodulation, coding of the transmission, and other parameters related to the radio channel.

[0007] On the Medium Access Control "MAC" layer, SRS are allocated at predefined time intervals synchronized via the uplink, allowing for a periodic evaluation of the radio channel. This is essential in efficiently allocating resources, minimizing the radio channel's interference, and maximizing the transmission capability.

[0008] The Radio Link Control "RLC" layer is highly dependent on the performance of the SRS since the RLC layer is responsible for functionalities like segmentation and reassembly of data packets and error correction. When the performance of the SRS is adequate, those functions will perform close to their best.

[0009] Some prior art documents try to improve the SRS report .

[0010] For example, document US 11, 664,874 B2, published on May 30, 2023, in the name of APPLE INC., and entitled "SYSTEM AND METHOD FOR DYNAMICALLY CONFIGURING USER EQUIPMENT SOUNDING REFERENCE SIGNAL (SRS) RESOURCES" describes techniques for selecting the number of sounding reference signal SRS resources to be used by a user equipment UE operating in a wireless communications system are provided. The UE determines a degree of partial beam correspondence of the UE and compares the degree to a threshold amount of partial beam correspondence support of the UE . In response to the comparison, UE selects a number of sounding reference signal SRS resources to request from a network node and sends an SRS resource request message tothe network node to request the selected number of SRS resources .

[0011] Document US 10,057,916 B2, published on August 21, 2018, in the name of COMMSCOPE TECHNOLOGIES LLC, and entitled "RADIO ACCESS NETWORK IN WHICH MOBILE DEVICES IN THE SAME COMMUNICATION CELL CAN BE SCHEDULED TO USE THE SAME AIRLINK RESOURCE" describes a communication system in a cellular network comprising: a processing system comprising a controller and remote units, with the remote units being configured to communicate with the controller and to communicate with mobile devices within a communication cell of the cellular network. At least part of the processing system is configured to perform operations comprising: estimating signal strength experienced by all or some of the mobile devices; identifying, based at least on the signal strength, one or more of the mobile devices that can be scheduled for communication with one or more of the remote units in the communication cell on a same airlink resource; and scheduling the communication.

[0012] Document EP 2,868,142 Bl, published on July 29, 2020, in the name of INTEL CORPORATION, and entitled "SOUNDING REFERENCE SIGNAL (SRS) MECHANISM FOR INTRACELL DEVICE-TO-DEVICE (D2D) COMMUNICATION" describes a technology for device discovery using a device-to-device D2D sounding reference signal SRS and device discovery using D2D SRS in a channel measurement group CMG. In an example, a user equipment UE configured for device discovery via a node using the D2D SRS can include a transceiver module. The transceiver module can send a radio resource control (RRC) device discovery request to a node, scan D2D SRS subframes ofproximity UEs using D2D SRS triggering, and send feedback to the node of detected D2D SRS information of the proximity UEs . The proximity UE can be located within a same cell as the UE .

[0013] Document US 11 , 700 , 073 B2 , publi shed on July 11 , 2023 , in the name of QUALCOMM INCORPORATED, and entitled "GROUP COMMON DCI FOR CLI MEASUREMENT AND REPORTING" provides systems , devices , apparatus and methods , including computer programs encoded on storage media, for grouping common DCI for CLI measurement and reporting . In a first aspect , a UE may receive DCI that is common to a group of UEs that includes the UE . The DCI may be configured to trigger at least one of an SRS transmission or a CLI measurement . The UE may transmit the SRS transmission and / or report the CLI measurement based on the DCI that is common to the group of UEs . In a second aspect , a base station may transmit at least one DCI that is common to the group of UEs to trigger the at least one of the SRS transmission or the CLI report for the group of UEs . The base station may receive the CLI report from at least a subset of UEs in the group of UEs .

[0014] However, a problem with the prior art is the fact that the solutions do not consider the location of the UEs for creating groups of similar wireless channel characteristics .

[0015] Another prior art problem lies in the methods are based on di f ferent approaches for energy saving related to SRS transmission, which include SRS reporting according to the device ' s duty cycle ( discontinuous reception - DRX ) , SRS report related to the UE speed, and energy saving mode attributed to antenna switching technique .

[0016] A further problem in the prior art is the fact that the methods provide battery savings related to j ust one device , being inef fective when considering the optimi zed reporting of SRS from multiple UEs , in which some UEs may be exempt from reporting .SUMMARY

[0017] An obj ective of the present invention is to provide a method for reporting SRS from clusters of mobile devices that solves the problems of the prior art .

[0018] This obj ective is achieved by means of a method for reporting SRS from clusters of mobile devices , comprising : identi fying an area with high mobile devices density; receiving mobile devices characteristics from mobile devices ; analyzing the mobile devices characteristics ; arranging mobile devices , with similar mobile device characteristics , into clusters with similar channel conditions ; using a machine learning algorithm to optimi ze a boundary between each cluster based on the channel conditions of each mobile device ; selecting, using the machine learning algorithm, the most suitable mobile devices to report SRS according to parameters defined by a network, wherein the machine learning algorithm selects the smallest possible number of mobile devices with the highest battery level to report SRS ; estimating channel conditions of each cluster, allowing the change of mobile devices reporting SRS based on theirbattery level and allowing the entrance of new mobile devices into a cluster .

[0019] Conveniently, the method according to the present invention consists o f the fact that the identi fication of the area with high mobile devices density further comprises identi fying mobile devices close to each other and in a low mobility pattern .

[0020] The method according to the present invention further consists in the fact that the mobile devices characteristics are at least one of location, SRS report , battery level , and Reference Signal Received Power .

[0021] Furthermore , the method according to the present invention consists in the fact that estimating channel conditions of each cluster comprises veri fying whether the SRS report reaches the expected SRS report performance .

[0022] Additionally, the method according to the present invention consists in the fact that changing mobile devices reporting SRS based on the battery level comprises changing mobile devices with low battery level by mobile devices with better battery level .

[0023] Further, the method according to the present invention consisting in the fact that the entrance of new mobile devices into a cluster is allowed i f the expected SRS report performance is not reached

[0024] An advantage of the present invention consists of using an area of high mobile device density divided into multiple clusters created according to a machine learning algorithm, where only some mobile devices report SRS to estimate the propagation channel while saving energy ofdevices with critical battery conditions and optimi zing the uplink channel for mobile devices with a large amount of data to be transmitted .

[0025] A further advantage of the present invention is the fact that the machine learning algorithm decides the most suitable devices to report SRS according to some filters defined by the network .BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The obj ects and advantages of the present invention will become clearer through the following detailed description of the examples and non-limiting drawings presented at the end of this document :

[0027] Figure 1 illustrates a high mobile device density area according to the present invention .

[0028] Figure 2 illustrates examples of high mobile devices density areas where the method according to the present invention is applied .

[0029] Figure 3 is a flowchart illustrating the step of identi fying an area with high mobile device density where the method according to the present invention may be applied .

[0030] Figure 4 is a flowchart illustrating the step of arranging devices with similar location characteristics according to the present invention .

[0031] Figure 5 is a flowchart illustrating the step of selecting the most suitable devices to report SRS , according to the present invention .

[0032] Figure 6 is a flowchart illustrating the step of estimating channel conditions for each cluster, according to the present invention .

[0033] Figure 7 is a flowchart illustrating the main steps of the method according to the present invention .DETAILED DESCRIPTION

[0034] Although the present invention may be susceptible to di f ferent embodiments , a preferred embodiment is shown in the following detailed discussion with the understanding that the present description should be considered an exempli fication of the principles of the invention and that the present invention is not intended to be limited to what has been illustrated and described here .

[0035] The present invention provides a method for wireless uplink channel estimation based on sounding reference signal "SRS" reporting from multiple mobile devices arranged in clusters , wherein the mobile devices have low mobility and relative proximity to each other . The mobile devices are divided into clusters by a machine learning algorithm according to similar characteristics . Figure 1 exempli fies a high mobile device density area, where multiple mobile devices are close to each other, sharing similar locations and propagation channel characteristics . The network operator defines a threshold related to the number of mobile devices in a speci fic area to start the method according to the present invention . The more devices in a given area, the more ef ficient the channel estimate will be , and the greater the overall battery savings of the mobile devices located on each cluster . Mobile devices are in the x-y plane , and the base station height is in the z- axis direction .

[0036] In this example , the mobile devices are divided into five clusters ( #1 to #5 ) , formed by relatively closedevices that share similar channel conditions. This method reduces the average battery consumption in a high mobile device density area, saving energy for mobile devices with low / critical-level batteries. Considering that the mobile devices are relatively close, not all mobile devices need to report SRS in the uplink channel since many are subject to the same channel conditions.

[0037] Figure 2 illustrates examples of high mobile devices density areas where the method according to the present invention is applied. According to Figure 2, high mobile device density area #1 illustrates a stadium where many mobile devices tend to stay relatively close to each other. This scenario includes musical shows, sports events, etc. High mobile device density area #2 represents a typical residential area, where many mobile devices share very similar channel conditions, especially in meetings, house parties, and homes shared by many people. A Mall is illustrated as the high mobile device density area #3, which may have different groups of relatively close devices, such as in the cinema and food court. High mobile device density area #4 represents a school where many mobile devices may share similar propagation channel conditions, for example, in the classrooms, meetings, and playground areas.

[0038] Other scenarios can use the proposed method, such as, churches, parks, theaters, gyms, street events, commercial buildings, or any other situation / scenario where mobile devices are relatively close to each other.

[0039] The method according to the present invention uses a machine learning algorithm to answer the following: How many clusters will each area have? Which mobile deviceswill belong to each cluster / group? Which mobile devices will report SRS from each cluster?

[0040] In this sense , in the method according to the present invention, critical battery level mobile devices do not report SRS for energy saving purposes , as other mobile devices can perform this task in the cluster . Additionally, mobile devices with significant needs of uplink data throughput do not report SRS .

[0041] Further according to method of the present invention, based on the average SRS reported by each cluster, the network defines pre-defined channel regions in which a new mobile device entering this geographic region receives its resources based on the average estimation of the current channel condition of the cluster . This optimi zes time and processing in resource allocation .

[0042] According to Figures 3 and 7 , the method identi fies a base station with a high number of mobile devices connected 110 and determines whether there are regions with many mobile devices close to each other and in a low mobility pattern 120 . The machine learning algorithm defines these situations , according to the preferences of each network, based on the previous applications of the method . For example , a first network might define the close mobile device value as " 1 meter" , while a second network might define the close mobile device value as " 50 cm" . For speed, the first network might define , for example , that the device is in low mobility pattern when it remains in the same location for at least 10 seconds , while second network would consider a time of 1 minute .

[0043] As such, the method according to the present invention identifies an area with high mobile device density 100.

[0044] In case there are no regions, the method maintains the current channel report settings.

[0045] In case there are regions, the method proceeds to arranging mobile devices, with similar mobile device characteristics, into clusters with similar channel conditions 200, as illustrated in Figures 4 and 7.

[0046] A machine learning algorithm receives mobile device characteristics from the mobile devices 210 such as location, SRS report, battery level, and Reference Signal Received Power.

[0047] Then, the machine learning algorithm divides the target area into clusters of mobile devices with similar channel conditions according to the mobile devices characteristics received and analyzed 220, such as Angle of Arrival "AoA" , which defines the direction in which the device is to the base station and determines the direction by measuring the path length difference at individual antenna array elements in the base station, Reference Signal Received Power "RSRP", which is linked to the distance between the mobile device and the base station helping to delimit each cluster, correlation of SRS report, location, and relative proximity between mobile devices.

[0048] The machine learning algorithm estimates the optimal boundary between each cluster 300 according to the channel condition reporting comparison from each mobile device. Very different SRS reports between two mobile devices indicate that they probably belong to different clusters.Nearby devices with the same SRS attenuation profile will belong to the same cluster. Angle of Arrival must define both the azimuth location "cp" and the elevation "0" to suit scenarios with groups of mobile devices at different heights. In addition, the network may use other tools to increase mobile device location accuracy, such as the Global Navigation Satellite System "GNSS".

[0049] According to Figures 5 and 7, the method then selects, using the machine learning algorithm, the most suitable mobile devices to report SRS 400 according to parameters defined by a network, wherein the machine learning algorithm selects the smallest possible number of mobile devices with the highest battery level to report SRS.

[0050] The network, by means of the machine learning algorithm, defines which mobile devices must report SRS 410 at a given time, being able to change both the update time interval in which mobile devices are reporting and which mobile devices report, according to the system's dynamics (mobility pattern, battery level, greater demand for the use of the uplink channel, etc.) . For example, the network is free to define that only mobile devices close to each cluster's edges / borders report SRS. Another scenario would be defining that devices with less than 10% battery do not report SRS, leaving this task to mobile devices with better battery levels. The network, for example, defines a longer time interval for low / critical battery level mobile device to report SRS, saving battery life. Suppose the channel estimate is below what the network wants (i.e., causing problems in the correct resource allocation) . In this case, the network requests more mobile devices in the cluster toreport SRS , improving the overall channel condition estimation . Another possible way to improve the average channel condition estimate is to choose more spread-out devices to better cover the entire cluster area . The machine learning algorithm play a fundamental role in this step, optimi zing the SRS reporting from each cluster and selecting the smallest possible number of mobile devices ( and those with better battery conditions ) to report SRS , aiming for an adequate estimate of the uplink channel .

[0051] According to Figures 6 and 7 , the method estimates the channel conditions of each cluster 500 , allowing the change of mobile devices reporting SRS based on their battery level and allowing the entrance of new mobile devices into a cluster .

[0052] Channel quality may be estimated through the SRS report from mobile devices in each cluster . In this way, the network maps clusters according to similar channel conditions , allowing quick resource allocation to new mobile devices that are entering into the cluster, in addition to reducing the energy consumption of devices with batteries in a low / critical state , as long as mobile devices with better battery conditions are more suitable for SRS reporting . Another advantage of the proposed method is the consequent reduction in network overhead due to the smaller number of mobile devices reporting SRS .

[0053] The method verifies whether the SRS report reaches the expected SRS report performance 510 and the entrance of new mobile devices into a cluster is al lowed i f the expected SRS report performance is not reached . Additionally, the method changes mobile devices reportingSRS based on the battery level in changing mobile devices with low battery level by mobile devices with better battery level .

[0054] In addition to the embodiments presented above , the same inventive concept may be applied to other alternatives or possibilities for using the invention .

[0055] Although the present invention has been described in relation to certain preferred embodiments , it should be understood that it is not intended to limit the invention to those particular embodiments . Rather, it is intended to cover all possible alternatives , modi fications and equivalences within the spirit and scope of the invention, as defined by the appended claims .

Claims

CLAIMS1 . A method for reporting SRS from clusters of mobile devices , comprising : identi fying an area with high mobile devices density; receiving mobile devices characteristics from mobile devices ; analyzing the mobile devices characteristics ; arranging mobile devices , with similar mobile device characteristics , into clusters with similar location and channel conditions ; using a machine learning algorithm to optimi ze a boundary between each cluster based on the channel conditions of each mobile device ; selecting, using the machine learning algorithm, the most suitable mobile devices to report SRS according to parameters defined by a network, wherein the machine learning algorithm selects the smallest possible number of mobile devices with the highest battery level to report SRS ; estimating channel conditions of each cluster, allowing the change of mobile devices reporting SRS based on their battery level and allowing the entrance of new mobile devices into a cluster .2 . The method as in claim 1 , wherein the identi fication of the area with high mobile devices density further comprises identi fying mobile devices close to each other and in a low mobility pattern .3 . The method as in claim 1 , wherein the mobile devices characteristics are at least one of location, SRS report , battery level , and Reference Signal Received Power .

4. The method as in claim 1, wherein estimating channel conditions of each cluster comprises verifying whether the SRS report reaches the expected SRS report performance.

5. The method as in claim 1, wherein changing mobile devices reporting SRS based on the battery level comprises changing mobile devices with low battery level by mobile devices with better battery level.

6. The method as in claim 1, wherein the entrance of new mobile devices into a cluster is allowed if the expected SRS report performance is not reached.

7. The method as in claim 1, wherein high mobile device density area is a place where many mobile devices tend to stay relatively close to each other, this place including but not limited to musical shows, sports events, meetings, house parties, and homes shared by many people, malls, cinema, food courts, schools, churches, parks, theaters, gyms, street events, commercial buildings, or any other situation / scenario where mobile devices are relatively close to each other.

8. The method as in claim 1, wherein, based on the average SRS reported by each cluster, the network defines pre-defined channel regions in which a new mobile device entering this geographic region receives its resources based on the average estimation of the current channel condition of the cluster.

9. The method as in claim 1, wherein a base station is identified with a high number of mobile devices connected and determines whether there are regions with many mobile devices close to each other and in a low mobility pattern, where the machine learning algorithm identifies thesesituations , according to the preferences of each network, based on the previous applications of the method .

Images