A middleware layer for performing high processing tasks in a RAN base station

Abstract

A method and system for offloading high processing tasks by a user equipment to a base station of a radio access network. The method includes the user equipment communicating with a middleware layer executing on a processor of the base station and requesting at least a portion of the processing of a task to be performed by computing resources available to the base station. The middleware layer processes the task utilizing the computing resources available to the base station and transmits the results back to the user equipment.

Description

Technical Field

[0001] This invention relates to a middleware layer integrated within a radio access network (RAN) base station. This middleware layer is designed to offload and perform high-processing tasks, such as artificial intelligence (AI) computations and other demanding processing tasks, from mobile devices. By utilizing the base station's spare computing power, the middleware layer enhances the performance and efficiency of mobile devices, reduces latency, and improves the overall user experience. Background Technology

[0002] High-processing tasks, also known as CPU-bound tasks, are those that require a significant amount of processing power from a computer's central processing unit (CPU) or processor. These tasks include, among others, video and graphics editing programs, games with high-resolution graphics, some complex mathematical calculations, and image compression.

[0003] Memory-intensive tasks are those that require a significant amount of RAM to run smoothly. These tasks typically involve processing large amounts of data or performing complex operations, such as gaming, video editing, AR / VR applications, and even some rich media web browsing sessions.

[0004] Mid-range to high-end devices have limited amounts of RAM, or even just a few gigabytes, which restricts multitasking capabilities and can create performance bottlenecks or lead to higher battery consumption. This is particularly problematic in high-end gaming or professional video editing use cases.

[0005] Integrating high-processing and / or memory-intensive tasks, especially those involving artificial intelligence (AI), into mobile devices has transformed how users interact with technology. As mobile devices become increasingly powerful, with advanced processors and improved energy efficiency, they are now capable of handling complex computing tasks that were once reserved for desktop computers and dedicated hardware.

[0006] One major application of AI in mobile devices is photography. Modern smartphones utilize AI algorithms for image processing, improving photo quality through features such as scene recognition, automatic adjustment, and real-time enhancement. The latest mobile (cell) phones employ machine learning to optimize photos, providing features that significantly enhance nighttime photography, such as night vision.

[0007] Another area where AI is making progress is speech recognition and natural language processing (NLP). Virtual assistants such as Apple's Siri, Google Assistant, and Amazon's Alexa rely on AI to understand and respond to user commands. These assistants can process voice commands in real time to perform tasks such as setting reminders, answering questions, and controlling smart home devices. The underlying AI models continuously learn from user interactions, improving their accuracy and responsiveness over time.

[0008] AI is also revolutionizing user experience through personalization. Mobile apps use AI to analyze user behavior, preferences, and usage patterns to deliver customized recommendations. For example, streaming services like Netflix and Spotify use AI algorithms to suggest content based on individual user tastes, improving engagement and satisfaction.

[0009] Furthermore, AI plays a crucial role in security through biometric authentication methods such as facial recognition and fingerprint scanning. Smartphones equipped with AI-driven algorithms can securely identify users, enhance device security, and provide a seamless unlocking experience.

[0010] In gaming, mobile devices now support high-processor AI tasks that enhance gameplay. AI-driven non-player characters (NPCs) in mobile games create more immersive and responsive experiences. Games can leverage AI to improve NPC behavior, making them more realistic and challenging.

[0011] As mobile devices continue to evolve, the realization of high-processing tasks such as AI not only enhances functionality but also enriches the overall user experience, paving the way for innovative applications and services in everyday life. The ongoing advancements in AI technology are likely to further expand these capabilities, making mobile devices even more indispensable for personal and professional activities.

[0012] However, mobile hardware is struggling to keep up with the increasing demands for high-performance applications. These applications require ever-increasing computing speeds and higher power consumption. Even if mobile devices are capable of handling the required processing power, users are still required to purchase increasingly expensive hardware if they want to fully utilize the benefits of these applications.

[0013] Therefore, the potential of high-processing applications (such as AI) may be limited by the user's hardware.

[0014] As demand for advanced mobile applications and services continues to grow, mobile devices are being required to perform complex computations, such as AI processing, real-time data analytics, and other high-processing tasks. These tasks require very high processing power, typically found only in high-end devices, and significantly deplete a mobile device's battery and processing power when performed, leading to reduced performance and user satisfaction.

[0015] A similar problem may occur with Internet of Things (IoT) devices, where the onboard processing power is not matched to the processing power required for certain tasks that it is originally capable of performing.

[0016] Therefore, there is a need for solutions that can offload these tasks from mobile or IoT devices to more capable infrastructure. Summary of the Invention

[0017] To address the aforementioned prominent issues, this invention is formulated to provide a middleware layer operating within a base station to perform high-processing tasks on behalf of the mobile device. The middleware layer is designed to integrate seamlessly with existing RAN infrastructure and communicate with the mobile device to receive and process computational tasks. By offloading these tasks to the base station, the middleware layer reduces the computational demands on the mobile device, improves battery life, and enhances the overall performance of mobile applications.

[0018] According to a first aspect of the present invention, a method is provided for a user equipment to offload a high-processing task to a base station of a radio access network, the method comprising: the user equipment communicating with a middleware layer executing on a processor of the base station, and requesting that at least a portion of the processing of the task be performed by computing resources available to the base station; and the middleware layer utilizing the computing resources available to the base station to process the task.

[0019] Optionally, while the task is being processed using the computing resources available to the base station, the middleware layer sends the processing results back to the user equipment.

[0020] High-processing tasks can be AI computing, data analysis, and other high-demand processing tasks.

[0021] The middleware layer can communicate with user equipment using a communication protocol, which may preferably be standardized.

[0022] Before a user equipment requests computing resources available from a base station to perform task processing, the user equipment may determine that its computing resources are insufficient to process the task independently.

[0023] The method may also include executing a virtualization layer on the processor of the user equipment, which performs: intercepting requests to perform tasks using computing resources available to the base station; and delegating assignments that control, allocate, schedule, and synchronize local and remote resources to perform tasks to the control and resource management layer of the user equipment.

[0024] The user equipment's control and resource management layer can be synchronized with the base station's middleware layer's control and resource management layer to determine the combined available computing resources for the user equipment and base station.

[0025] The method may also include allocating at least a portion of the task processing to computing resources available to the base station.

[0026] The computing resources available to a base station may include at least one of the processors and memory located in the base station.

[0027] The computing resources available to the base station may include remote computing resources connected to the base station via a radio access network, including network capabilities that can be reserved for the middleware layer in accordance with the available standard reservation process for network resources.

[0028] The availability of computing resources available at a base station for user equipment may depend on time.

[0029] The base station's middleware can be installed and upgraded via software updates.

[0030] User equipment can be a mobile device or an Internet of Things (IoT) device.

[0031] Mobile devices can be smartphones, tablets, laptops, or laptops.

[0032] According to another aspect of the invention, a base station for a radio access network is provided, which includes at least one processor executing a middleware layer arranged to perform the above-described method.

[0033] According to another aspect of the invention, a radio access network is provided, which includes at least one base station as described above. Attached Figure Description

[0034] Embodiments of the invention will now be described by way of example only and with reference to the accompanying drawings, wherein: Figure 1 A schematic diagram illustrating a user using a mobile device and an IoT device connected to a base station node is shown according to an embodiment of the present invention; Figure 2 A flowchart illustrating the steps performed by a base station after receiving a request from a mobile device to offload a processing task, according to an embodiment of the present invention, is shown. Figure 3 A schematic diagram of a base station node according to an embodiment of the present invention is shown; Figure 4 A schematic diagram illustrating the connection between a user equipment (UE) and a RAN base station node according to an embodiment of the present invention is shown; Figure 5 A schematic diagram illustrating the connection between an Internet of Things (IoT) device and a RAN base station node according to an embodiment of the present invention is shown. Detailed Implementation

[0035] Figure 1 A user is shown using user equipment 10 (such as a mobile device) connected to a base station 100 in a radio access network (RAN). According to the invention, user equipment 10 can offload some of the processing tasks requested to be performed to base station 100. In this way, when user equipment 10 identifies that it requires further processing capabilities, it can request, on its behalf, to base station 100 to perform at least a portion of the current processing. If base station 100 accepts the request, it uses its available computing resources to perform the processing of the offloaded task. Once the processing is performed, base station 100 can then send the result of the processing back to user equipment 10.

[0036] Other types of devices may be able to use the methods of the present invention and offload processing to base station 100, not just mobile devices, and some of these devices will be discussed later. Non-mobile devices (such as, but not limited to, desktop computers and Internet of Things (IoT) devices) can also benefit from the present invention. For example, Figure 1 Internet of Things (IoT) devices 20, such as cameras, connected to base station 100 are also shown. Such cameras may be able to perform numberplate or facial recognition, but some of the functions of such cameras require additional computing resources.

[0037] The following discussion largely concerns mobile devices, but the examples describing mobile device 10 are equally applicable to any other device that can connect to RAN base station 100 and request to offload its tasks, and these examples should not be considered limited to mobile device 10.

[0038] Figure 2 The function of base station 100 when it receives a request from mobile device 10 or other device to unload one or more tasks is summarized.

[0039] At step S100, a request message from a user equipment (UE) such as mobile device 10 is received by base station 100, requesting that a processing task be offloaded to base station 100 for processing. That is, the request message asks base station 100 to receive the processing task from mobile device 10 and to use the computing resources available to base station 100 to process the task on behalf of mobile device 10.

[0040] At step S110, base station 100 checks whether it has computing resources available to perform processing of tasks offloaded from mobile device 10. The evaluation of resource availability may require synchronization with conventional retention processes for relevant network capabilities (e.g., via slicing, QoS retention, APIs, etc.).

[0041] In step S120, if the base station 100 has available computing resources, the task is assigned to those resources for processing. If no available resources are identified, the request to process the task is rejected.

[0042] At step S130, the allocated computing resources represent the mobile device 10 in handling the unloaded tasks, so that the mobile device 10 does not have to handle the tasks itself.

[0043] In step S140, base station 100 transmits the result of the processing of the offloaded task to mobile device 10 that requested the offload task, and then the process ends. This may be applicable to any billing that triggers the service.

[0044] Base station 100 will have a processor and memory associated with it. The computing resources available to base station 100 may include these processors and memory (both permanent and volatile), allowing tasks offloaded from mobile device 10 to be processed within base station 100 itself. However, the computing resources available to base station 100 may also include resources external to base station 100. These could be resources at other connected base stations 100, cloud computing services, distributed computing resources, external server stations, or any other computing resources connected to base station 100. These could even be other mobile devices connected to base station 100 whose computing resources are available to base station 100.

[0045] In order for base station 100 to be adapted to perform the offloading process from user equipment 10, the middleware of base station 100 will need to be updated.

[0046] Middleware is a software layer running on the processor of base station 100 in the Radio Access Network (RAN) and acts as an intermediary between mobile device 10 and the RAN. It facilitates communication and data exchange, enabling mobile device 10 to connect to the network efficiently and effectively. Middleware reduces the complexity of underlying hardware and network protocols, allowing developers to create applications without needing to understand the details of how different devices and networks communicate. It further ensures compatibility between different mobile devices 10 and various RAN technologies. This is especially important given the wide range of operating systems, hardware capabilities, and network standards currently in use.

[0047] Middleware plays a crucial role in facilitating this connectivity, acting as an intermediary to enhance communication, data processing, and interoperability across different systems. In the context of mobile devices, which vary significantly in hardware capabilities, operating systems, and user interfaces, middleware is used to allow for more consistent interaction with the RAN's base station 100.

[0048] Middleware manages the processes of establishing, maintaining, and terminating connections between mobile device 10 and the RAN. It helps the devices negotiate the best possible connection parameters and optimize performance based on current network conditions. Middleware can also manage data processing and formatting, ensuring that sent and received information is compatible with both mobile device 10 and the RAN. This includes data compression, encryption, and error handling. It can also maintain session continuity during communication, manage user sessions, and ensure smooth data flow even when network conditions change or outages occur.

[0049] Middleware plays a key role in enhancing the performance, reliability, and user experience of mobile connectivity by simplifying the integration between the mobile device 10 and the radio access network.

[0050] The middleware layer is implemented as a software module within base station 100. It includes interfaces for communicating with mobile device 10 and RAN infrastructure. The middleware layer is designed to be scalable and capable of handling multiple concurrent tasks from various mobile devices 10.

[0051] According to the invention, the middleware is also adapted to allow user equipment (UE) 10 (such as mobile device 10) to utilize the spare resources available to base station 100.

[0052] Figure 3A schematic diagram of a base station 100 node according to an embodiment of the present invention is shown. The base station 100 has hardware resources 110 located on the base station 100, including memory and a processor, some of which are allocated to perform the normal functions of the base station in connecting the mobile device 10 to a telecommunications network. Additional hardware resources are available to the base station 100, which are not located on the base station 100 but are still allocated for use by the base station 100.

[0053] Base station 100 has one or more processors with high processing power. Furthermore, power issues are unlikely because base station 100 is normally connected to the national power grid. Therefore, base station 100 can have spare resources 120 available for performing additional processing tasks. These spare resources 120 can be allocated to be shared with the connected mobile device 10.

[0054] Specifically, according to embodiments of the invention, base station 100 can be configured to use its spare resources 120 to offload tasks from connected mobile devices 10. In this case, a middleware layer running on base station 100 can receive high-processing tasks from the mobile devices, such as AI calculations, data analysis, and other demanding tasks. The middleware layer can then utilize the high-performance computing resources of base station 100 to efficiently execute the received tasks, rather than requiring the mobile devices 10 to perform the processing themselves. Upon task completion, the middleware layer sends the results back to the corresponding mobile device 10. Thus, mobile devices 10 can access higher-processing applications that they would normally not be able to access by offloading processing to base station 100 via middleware.

[0055] like Figure 3 As shown, the resources of base station 100 are shared between the base station's normal functions and the spare resources 120 available to the connected mobile device 10. A control and resource management layer 130 within the middleware provides an interface with user device 10.

[0056] To offload tasks from mobile device 10 to base station 100 in the RAN, the middleware layer will need to use communication protocols to interact with both mobile device 10 and the RAN infrastructure. Preferably, these protocols will be standardized. This ensures secure and reliable data transmission between mobile device 10 and base station 100.

[0057] Figure 4An example of the invention is illustrated below. Here, a user equipment (UE) 10, such as a mobile device, connects to a base station 100 and attempts to handle an application that may require more processing power than the processing power that the UE 10 can provide. Communication steps 1 to 5, discussed below, are identified by numbers in circles. Thus, at step 1, the application of the UE 10 requests the operating system 14 of the UE 10 to allocate hardware resources for one or more specific tasks. At step 2, the application in the UE 10, referred to as, for example, "..." virtual Mimicry layer The hidden layer (which may be part of the operating system 14) intercepts the request and delegates the task of controlling, allocating, scheduling, and synchronizing (orchestrending) local and remote resources to perform one or more tasks to the control and resource management layer 13 of the UE 10. At step 3, the operating system of the UE 10 or the control and resource management layer 13 within the UE checks whether the local hardware resources of the UE 10 are sufficient to perform the requested one or more tasks. At step 4, the UE control and resource management layer 13 continuously synchronizes with the control and resource management layer 130 of the base station to obtain a mapping of available resources on both sides. At step 5, if it is determined that the UE 10 will utilize the base station's spare resources 120, these resources 120 are then allocated to perform one or more UE tasks.

[0058] The present invention, which offloads high-processing tasks from mobile device 10 to base station 100, has various technical advantages.

[0059] A key advantage lies in the improved performance achieved through the increased computing power available at base station 100. While modern mobile devices 10 may be powerful compared to historical standards, they are still constrained by their processing capabilities. Therefore, offloading resource-intensive tasks (such as AI computation, video encoding / decoding, and 3D rendering) to a more powerful off-device processor enables more efficient execution of these tasks. This results in faster processing times and better overall performance. Offloading can even allow for tasks that would otherwise be impossible for mobile devices to perform, thus enabling additional functionality for the device.

[0060] Offloading can also allow for parallel processing, as tasks such as machine learning (ML) or complex mathematical operations benefit from parallel processing. Therefore, using a separate processor, such as the processing power available at base station 100, enables tasks to be executed in parallel, significantly accelerating computation compared to the limited capabilities of a mobile device's processor.

[0061] Another important consideration is that high-processing-demand tasks consume a lot of power. Therefore, offloading at least a portion of this processing can reduce power consumption. Since mobile device 10 is limited by its battery life, running computationally intensive tasks on the main processor can lead to excessive power consumption. Therefore, offloading tasks to a more energy-efficient processor in external infrastructure can result in substantial power savings.

[0062] Excessive power consumption can lead to the secondary problem of excessive heat generation in mobile device 10. Mobile device 10 has limited heat dissipation capabilities, and therefore performing demanding tasks directly on the mobile processor generates excessive heat, which negatively impacts the overall performance of mobile device 10 and the corresponding user experience. Therefore, offloading computationally heavy tasks to base station 100 can help reduce the heat generated in the mobile device, keeping the device cooler and more stable.

[0063] The middleware layer can handle multiple tasks from various mobile devices 10 simultaneously, making it suitable for large-scale deployment. Offloading to the base station 100 provides significantly increased computing resources for extremely demanding tasks. Therefore, mobile devices 10 can handle complex workloads without being constrained by the hardware limitations of the mobile devices themselves. This also provides flexibility for handling tasks that require a large amount of computing power, such as large-scale data analysis and machine learning model training.

[0064] Another advantage is that new features do not necessarily require the latest version of the mobile device, because new features can be upgraded in the base station software, rather than in the hardware of the mobile device 10.

[0065] This leads to the advantage that the invention reduces the need for overpowered mobile hardware. By offloading complex tasks to an external processor at the base station, the mobile device 10 does not need to be equipped with the most expensive or powerful hardware to perform demanding tasks. This reduces the overall cost and weight of the device and allows for a slimmer form factor, making the device more affordable and portable.

[0066] Figure 5 A second example of the invention is shown below. Here, the device connected to base station 100 is Internet of Things (IoT) device 20, not user equipment (UE) 10. Again, device 20 can handle applications that may require more processing power than the processing power that IoT device 20 can provide. Communication steps 1 to 5, discussed below, are identified by numbers in circles. Thus, at step 1, the application of IoT device 20 requests the operating system 24 of IoT device 20 to allocate hardware resources for one or more specific tasks. At step 2, the application in the IoT device operating system 24 is referred to as, for example, "..." Virtualization layer The hidden layer (which may be part of the operating system 24) intercepts the request and delegates the task of controlling, allocating, scheduling, and synchronizing (or orchestrating) local and remote resources to perform one or more tasks to the control and resource management layer 23 of the IoT device 20. At step 3, the operating system 24 of the IoT device 20 or the control and resource management layer 23 of the IoT device 20 checks whether the local hardware resources of the IoT device 20 are sufficient to perform the requested one or more tasks. At step 4, the IoT device control and resource management layer 23 synchronizes with the base station control and resource management layer 130 to obtain a mapping of available resources on both sides. At step 5, if it is determined that the IoT device 20 will utilize the base station's spare resources 120, then these resources 120 are then allocated to perform one or more IoT device tasks.

[0067] This invention offers additional advantages for IoT devices. Many types of IoT devices can utilize this invention. For example, the incomplete list used for illustration includes only devices for traffic monitoring, license plate recognition, weather measurement, CCTV facial recognition, etc. These may be remote devices with limited power and computing resources, and therefore cannot provide any form of complex processing. However, as long as they have sufficient processing power to send requests for processing to a nearby base station 100 and receive corresponding results, IoT devices can perform functions that would normally be impossible to perform using their available computing resources.

[0068] In some cases, depending on a range of factors, the computing resources available to base station 100 can be prioritized for use by IoT devices. For example, if police officers want to use facial recognition from CCTV, and the facial recognition process is offloaded to base station 100, then when an emergency request for that service is made by the police, it can be given higher priority than other tasks that base station 100 is performing.

[0069] For other users of this invention, the priority of the computing resources of base station 100 can also be changed. For example, during peak hours, base station 100 may already have a high processing load due to normal telecommunications services. Suppressing mobile communications to handle tasks offloaded by other users would be inappropriate. In this case, the computing resources available to mobile device 10 or IoT device 20 may be limited at certain times, such as during periods of high demand for the RAN telecommunications network. The telecommunications network operator may also charge different rates for offloading tasks at different times based on demand.

[0070] Preferably, the middleware of existing base station 100 can be updated to add new functionality without requiring hardware updates. For example, the middleware of base station 100 can be updated via over-the-air (OTA) software updates. This is typically a complex and multi-stage process that ensures minimal disruption to service and ideally maintains the operational integrity of the network. Middleware in a RAN context typically comprises software components that handle basic tasks such as communication protocol processing, network resource management, and inter-process communication. Updates can be performed remotely to provide new functionality to the middleware and may also include additional updates to improve performance, fix bugs, or enhance security.

[0071] In summary, offloading high-processing-demand tasks from mobile device 10 (or IoT device 20) to a dedicated processor via a connection to base station 100 offers various technical advantages, including enhanced performance, power efficiency, reduced thermal load, and scalability. This enables mobile device 10 to handle more complex tasks without sacrificing battery life or user experience, while also allowing for faster response times and more advanced capabilities. By utilizing a dedicated high-performance processor, mobile device 10 can be developed to handle increasingly demanding applications with improved efficiency and cost-effectiveness.

[0072] Although specific embodiments have been described as examples to illustrate the invention, several alternatives and variations will be apparent to those skilled in the art without departing from the invention as set forth in the claims.

Claims

1. A method for offloading high-processing tasks from a user equipment to a base station of a radio access network, the method comprising: The user equipment communicates with a middleware layer that executes on the processor of the base station and requests that at least a portion of the processing of the task be performed by the computing resources available to the base station. as well as The middleware layer utilizes the computing resources available to the base station to process the task.

2. The method according to claim 1, wherein, After the task is processed by the computing resources available to the base station, the middleware layer sends the result of the processing back to the user equipment.

3. The method according to claim 1 or 2, wherein, The high-processing tasks are one of the following: AI computing, data analysis, and other high-demand processing tasks.

4. The method according to any of the preceding claims, wherein, The middleware layer communicates with user equipment using a communication protocol that is to be standardized.

5. The method according to any of the preceding claims, wherein, Before the user equipment requests computing resources available from the base station to perform the processing of the task, the user equipment determines that its computing resources are insufficient to process the task on its own.

6. The method according to any of the preceding claims further comprises executing a virtualization layer on the processor of the user equipment, the virtualization layer performing: Intercept the request to execute the task using the computing resources available to the base station; The task of assigning control, allocation, scheduling, and synchronization of local and remote resources to perform the task is delegated to the control and resource management layer of the user equipment.

7. The method according to claim 5, wherein, The control and resource management layer of the user equipment is synchronized with the control and resource management layer of the middleware layer of the base station to determine the combined available computing resources of the user equipment and the base station.

8. The method of claim 6, further comprising allocating at least a portion of the processing of the task to the computing resources available to the base station.

9. The method according to any of the preceding claims, wherein, The computing resources available to the base station include at least one of the processor and memory located in the base station.

10. The method according to any of the preceding claims, wherein, The computing resources available to the base station include remote computing resources connected to the base station via the radio access network, the radio access network including network capabilities, including network capabilities that can be reserved for the scope of the middleware layer in accordance with the available standard reservation process of network resources.

11. The method according to any of the preceding claims, wherein, The availability of the computing resources available to the base station for the user equipment depends on time.

12. The method according to any of the preceding claims, wherein, The middleware of the base station is installed and upgraded via software updates.

13. The method according to any of the preceding claims, wherein, The user equipment is a mobile device or an Internet of Things (IoT) device.

14. The method according to claim 13, wherein, The mobile device is a smartphone, tablet, laptop, or laptop.

15. A base station for a radio access network, comprising at least one processor, the at least one processor executing a middleware layer, the middleware layer being configured to perform the method according to any of the preceding claims.

16. A radio access network comprising at least one base station as claimed in claim 15.

Images