Network power savings using bandwidth scaling

Abstract

A method may include allocating, by a system, an adaptive cell-specific bandwidth portion for facilitating cellular network communications, where the adaptive cell-specific bandwidth portion includes a set of bandwidth sizes capable of achieving different energy consumption of a cellular network. The method may also include transitioning, by the system, from the first bandwidth size to a second bandwidth size based on determining that the second bandwidth size is sufficient to serve the predicted amount of network traffic, where the second bandwidth size is smaller. The method may also include, after transitioning to the second bandwidth size, transitioning, by the system, from the second bandwidth size to the first bandwidth size based on a determination that the second criterion has been satisfied. The method may also include facilitating, by the system, cellular network communication with the first bandwidth size after transitioning from the second bandwidth size to the first bandwidth size.

Description

Cross-reference to related applications

[0001] This application claims priority to U.S. nonprovisional patent application No. 18 / 364,442, filed August 2, 2023, entitled “NETWORK ENERGY SAVINGS USINGBANDWIDTH SCALING”, the entire contents of which are incorporated herein by reference. Background Technology

[0002] Base stations can communicate with user equipment to facilitate mobile or cellular network communication. In this process, base stations consume energy. Summary of the Invention

[0003] The following is a simplified overview of the disclosed subject matter to provide a basic understanding of some embodiments in various examples. This content is not a comprehensive overview of all embodiments. It is not intended to identify key or essential elements of the various embodiments, nor is it intended to define the scope of the various embodiments. Its sole purpose is to present some concepts of this disclosure in a concise form as a prelude to the specific implementations presented later.

[0004] An example method may include: allocating an adaptive cell-specific bandwidth portion by a system to facilitate cellular network communication, wherein the adaptive cell-specific bandwidth portion comprises a set of bandwidth sizes, and wherein a corresponding size in the set of bandwidth sizes is associated with a corresponding frequency range. The method may further include: the system switching from a first bandwidth size in the set of bandwidth sizes to a second bandwidth size in the set of bandwidth sizes, wherein the second bandwidth size is smaller than the first bandwidth size, based on a determination, according to a first criterion, that a second bandwidth size in the set of bandwidth sizes is sufficient to serve predicted network traffic. The method may further include: the system facilitating first cellular network communication based on the second bandwidth size. The method may further include: after switching from the first bandwidth size to the second bandwidth size, the system switching from the second bandwidth size back to the first bandwidth size based on a determination that a second criterion has been met. The method may further include: after switching from the second bandwidth size to the first bandwidth size, the system facilitating second cellular network communication based on the first bandwidth size.

[0005] An example system can operate as follows: The system can maintain a set of bandwidth sizes for cellular network communication, where each bandwidth size corresponds to a corresponding frequency range. The system can switch from a first bandwidth size to a second bandwidth size in the set of bandwidth sizes, where the second bandwidth size is smaller than the first bandwidth size, based on a determination, according to a first criterion, that a second bandwidth size in the set of bandwidth sizes is sufficient to serve predicted network traffic. The system can then transmit first cellular network communication based on the second bandwidth size. After switching from the first bandwidth size to the second bandwidth size, the system can switch from the second bandwidth size back to the first bandwidth size based on a determination that a second criterion has been met. The system can then transmit second cellular network communication based on the first bandwidth size.

[0006] An example non-transient computer-readable medium may include instructions that, in response to execution, cause a system including a processor to perform operations. These operations may include: switching from a first bandwidth size to a second bandwidth size in a set of bandwidth sizes, wherein the second bandwidth size is smaller than the first bandwidth size, based on determining that a second bandwidth size in a set of bandwidth sizes is a threshold capable of serving predicted cellular network traffic. These operations may also include: facilitating first cellular network communication according to the second bandwidth size. These operations may further include: after switching from the first bandwidth size to the second bandwidth size, switching from the second bandwidth size back to the first bandwidth size based on determining that a criterion has been met. These operations may also include: facilitating second cellular network communication according to the first bandwidth size. Attached Figure Description

[0007] Numerous embodiments, objects, and advantages of this disclosure will become apparent when considering the following specific implementations in conjunction with the accompanying drawings, in which like reference numerals refer to like parts, and wherein:

[0008] Figure 1 The illustration shows an example system architecture that can facilitate network energy saving using bandwidth scaling according to embodiments of the present disclosure;

[0009] Figure 2 The illustration shows an example of adaptive bandwidth usage of a base station based on operator spectrum allocation in fifth-generation new radio (5G NR) communications according to an embodiment of the present disclosure, which can facilitate network energy saving using bandwidth scaling;

[0010] Figure 3 An example bandwidth adaptation scheme for dynamic bandwidth adaptation scheduling according to an embodiment of the present disclosure is illustrated, which can facilitate network energy saving using bandwidth scaling.

[0011] Figure 4The illustration shows an example relative bandwidth portion that can facilitate bandwidth adaptation for downlink communication at a base station according to an embodiment of the present disclosure, which can facilitate network energy saving using bandwidth scaling.

[0012] Figure 5 The illustration shows an example lifecycle management technique for bandwidth adaptation at the base station using a portion of the bandwidth according to an embodiment of the present disclosure, which can facilitate network energy saving using bandwidth scaling.

[0013] Figure 6 The illustration shows an example flow that facilitates network energy saving in a base station by using bandwidth scaling according to an embodiment of the present disclosure;

[0014] Figure 7 The illustration shows another example flow that facilitates network energy saving in a base station by using bandwidth scaling, according to an embodiment of the present disclosure;

[0015] Figure 8 The illustration shows another example flow that facilitates network energy saving in a base station by using bandwidth scaling, according to an embodiment of the present disclosure;

[0016] Figure 9 The illustration shows an example flow for reducing bandwidth that can facilitate network energy saving in a base station by using bandwidth scaling according to an embodiment of the present disclosure;

[0017] Figure 10 The illustration shows an example flow for reducing bandwidth that can facilitate network energy saving in a base station by using bandwidth scaling according to an embodiment of the present disclosure;

[0018] Figure 11 An example block diagram of a computer operable to perform embodiments of the present disclosure is illustrated. Detailed Implementation Overview

[0019] The examples described in this article may involve optimizing things, such as base station calibration. It should be understood that this technique can be applied to improve similar metrics, even if the improvement may not be optimal. Similarly, it should be understood that when examples are described at the highest level, such as maximizing a certain metric, there may indeed be examples that can improve or enhance that metric.

[0020] While the examples in this article typically involve fifth-generation (5G) new radio networks, it should be understood that this technology can also be applied to other types of networks.

[0021] To address the growing demands for data rates and capacity, efficient spectrum use and utilization will become increasingly important for future cellular network deployments. The ever-increasing energy consumption of mobile networks is likely to be a key cost factor in the modernization and expansion to 5G networks and future networks. Furthermore, improving network energy efficiency can focus on reducing power consumption in the Radio Access Network (RAN), as it accounts for nearly 80% of total energy consumption. In some examples, one aspect contributing to this continued increase in energy costs is the generation-widening transmission bandwidth of cellular wireless communications. From the 5 to 20 MHz carrier bandwidth introduced in fourth-generation (4G) communications, to the maximum available bandwidth of up to 400 MHz in 5G networks (utilizing carrier aggregation), this is 20 times that of 4G. While this can significantly enhance overall throughput and support future low-latency applications such as augmented reality / virtual reality (AR / VR), it also introduces a significant power consumption burden due to the use of broadband power amplifiers (PAs) at the base station (BS). This technology can be implemented to facilitate a reduction in this enormous power consumption at the base station using adaptive bandwidth scaling for adaptation, translating this adaptive bandwidth scaling into energy efficiency gains.

[0022] Given the current emphasis on energy conservation, technologies beneficial for reducing overall operational power consumption are being sought. Up to 80% of energy usage in mobile networks can be attributed to RAN components, and while several measures have been implemented in specifications to minimize transmission overhead, this problem is likely to worsen with the increasing prevalence of 5G NR networks. In this context, the use of wider bandwidth in 5G NR could exacerbate power consumption-related issues if appropriate steps are not taken to optimize RAN energy usage.

[0023] 5G NR networks can include the concept of a numberology to accommodate different subcarrier bandwidths, thereby meeting the varying latency and throughput requirements of target use cases. In some examples, 5G NR networks can support wide carrier bandwidths: up to 200 MHz for frequency range 1 (FR1, e.g., below 6 GHz); and up to 400 MHz for frequency range 2 (FR2, e.g., 24-52 GHz). This can lead to significant power consumption on user equipment (UEs) as they need to monitor extremely large bandwidths to maintain connectivity and request services. Since in some examples, most UE downlink (DL) traffic can be served by narrower bandwidth slices, a certain amount of flexibility is needed in the bandwidth monitored by the UE to reduce the potentially enormous demands on its battery, as most devices have limited power. To provide a unified solution for both numberology and carrier bandwidth (BW) flexibility, the 3GPP Release 15 specification identifies the Bandwidth Partition (BWP) of 5G NR. The BWP specified in the 3GPP standard only allows receive-side bandwidth adaptation for the 5G NR access interface, meaning that BWP use is only applicable to the UE. Generally, a subset of the total cell bandwidth can be referred to as a BWP, and bandwidth adaptation (BA) can be achieved by configuring (multiple) BWPs for the UE, where the base station informs the UE via control signaling which of the configured BWPs is currently active. Since the UE does not always require high data rates, using a wide BW may mean higher power consumption from a radio frequency (RF) and baseband signal processing perspective, even during idle periods. Therefore, a BWP can provide a smaller BW for a connected UE than the configured cell bandwidth, thus providing a more energy-efficient operating mode for the UE while supporting wideband operation. In some examples, the base station can use the same downlink control signaling used for scheduling information to activate and deactivate the UE's BWP. In other examples, the UE can be scheduled so that it only transmits or receives within a specific frequency range. The difference between this approach and BWP is that the UE does not need to transmit or receive outside the configured frequency range of the active BWP, which contributes to power savings.

[0024] The RF baseband interface can be operated using a lower sampling rate, and the baseband processing required for transmission or reception with a narrower bandwidth is reduced.

[0025] At least when the default carrier bandwidth is relatively large before adaptation, UE RF bandwidth adaptation can provide UE power savings.

[0026] In some examples, actual power consumption can vary depending on each modem and RF implementation, making it subjective to provide a generalized power-saving gain figure. In other examples, power-saving gain measurements are crucial in defining the user's actual quality of experience (QoE). User applications may have different requirements for downlink / uplink (UL) throughput, packet latency, or resource demands from the network. Therefore, there may be an increasing focus on improving the end-user experience and adapting network capacity to meet diverse data usage patterns. To this end, achieving the highest possible data rate may not always be the primary requirement for applications and user experience; factors such as battery life may also be important. This could mean that power consumption should be considered based on a service profile of the service type and its data rate requirements.

[0027] For example, this technology can be implemented in the following way. Based on the ML model, the bandwidth of the DL base station can be adapted to promote network energy saving. This ML model can take into account the characteristics of network services and use an ML-based power consumption model, so that the bandwidth adaptation is only implemented when the pre-established NES saving threshold is considered to be achievable.

[0028] ML-based service prediction models can be used to develop dynamic base station power consumption behavior under the assumption of full service and to use the optimal base station power consumption (BWP) based on service requirements, where the difference between the two can be used as a decision metric for network energy saving.

[0029] Power reduction achieved by leveraging bandwidth scaling enables gradual adaptation of DL bandwidth and allows for the use of schemes as described in this paper for network power saving processes.

[0030] Figure 1 An example system architecture 100 is illustrated according to an embodiment of the present disclosure, which can facilitate network energy saving in a base station by using bandwidth scaling.

[0031] System architecture 100 includes base station 102 and user equipment 104. Base station 102 further includes a determined bandwidth size 106 and a network energy-saving component 108 using bandwidth scaling.

[0032] Each of base station 102 and / or user equipment 104 can utilize Figure 11 The computing environment 1100 is implemented in multiple parts. Base station 102 typically includes one or more antennas and electronic communication devices to facilitate network communication with user equipment 104. User equipment 104 typically includes computing devices having one or more antennas for end users to communicate with base station 102.

[0033] As part of communication with user equipment 104 (including examples of communication with multiple user equipments), base station 102 can determine the bandwidth size used for communication from a determined bandwidth size 106. When the determined bandwidth size is lower than the current bandwidth size, the base station can achieve associated energy savings by using this lower new bandwidth size.

[0034] In some examples, the network power-saving component 108 using bandwidth scaling can achieve this. Figures 6-10 The process involves (multiple) parts, thereby enabling network energy saving through bandwidth scaling at the base station.

[0035] Figure 2 The illustration depicts an example adaptive bandwidth usage 200 for a base station based on operator spectrum allocation according to an embodiment of the present disclosure, which can facilitate network energy saving using bandwidth scaling. In some examples, portions(s) of adaptive bandwidth usage 200 can be used to achieve... Figure 1 The system architecture consists of 100 (multiple) parts.

[0036] Adaptive bandwidth usage 200 includes spectrum allocated to the operator 202, cell-specific total channel bandwidth 204, BWP adaptation at the base transceiver (Tx) 206 (which can be implemented according to this technology), adaptive cell-specific bandwidth portion 208, UE-specific carrier offset 210, UE-specific bandwidth portion 212, bandwidth portion 214, and network power saving components 216 using bandwidth scaling (which can be similar to...). Figure 1 (Using a network power-saving component 108 with bandwidth scaling).

[0037] Figure 2 The text describes BWP and its utility. While the widest bandwidth can be the cell bandwidth or the bandwidth that a mobile network operator (MNO) can use in a frequency band based on the spectrum allocated to them, this technique can involve a smaller bandwidth segment from that total wide cell bandwidth. According to the Nyquist sampling theorem, to sample a signal with a total bandwidth spanning "B" MHz, the minimum required sampling rate would be 2BMHz (typically higher). This results in power consumption in the form of data converters (analog-to-digital converters (ADCs), which are necessary for sampling wideband signals. While some UEs can mitigate this requirement by using UE-specific BWPs, for base stations, using reduced transmission bandwidth to communicate with UEs currently served by the BS (or even those potentially served in the future) during idle or low-traffic periods may involve more steps. This is because, in addition to the signaling overhead required to transmit using such reduced bandwidth, there may be additional strategies to mitigate the risk of transmission delay and other critical performance indicators (KPIs) degradation, while prioritizing energy efficiency.

[0038] Some previous methods typically utilize the concept of BWP (Broadband Power Plan) applied to the transmit or receive bandwidth of a specific UE to reduce UE power consumption, but they do not contribute to reducing base station power consumption in any way. Base stations can typically cover the entire spectrum allocation of a particular operator. However, for efficient network design, base stations can have power profiles that scale according to changes in data rate / total service demand. RAN power consumption may not scale linearly with a decrease in transmitted data rate, such as when an active base station utilizes full bandwidth to transmit a lower data rate service profile. Furthermore, in macrocell environments, it may be necessary to support the operator's full bandwidth, but 5G NR scenarios may involve the deployment of dedicated wireless networks with lower requirements for consistently supporting full bandwidth. The pervasive power saving needs in base stations may necessitate considering methods that can deliver substantial gains in overall network energy efficiency. This technique can be implemented to facilitate adaptive bandwidth utilization by base stations, thereby saving power consumption.

[0039] In some examples, the energy consumption resulting from the wide bandwidth consumed by cellular radio can be excessively high and unsustainable, potentially putting significant pressure on network operators' operating expenses (OPEX). Furthermore, with the push towards higher throughput, 5G networks can reach maximum bandwidths of 400 MHz for some frequency bands, four times that of previous generations before Release 15. Supporting such wide bandwidth leads to significant energy consumption in RF and physical layer (PHY) modules, including parts of the radio unit (RU) and distributed unit (DU). Additionally, despite measures such as compression, the fronthaul providing connectivity between the RU and DU can also face significant pressure due to the need to carry large amounts of in-phase quadrature (I / Q) signals at very high throughput. From both an operational and energy efficiency perspective, these and other factors can make supporting the maximum bandwidth at the base station extremely challenging. While measures can be taken according to 3GPP standards to reduce the burden on UEs supporting wide bandwidth, few measures can be taken to enhance the overall energy efficiency of the base station.

[0040] One problem with previous approaches is that power consumption can be quite high when using wide bandwidth due to the use of wideband RF components (such as a single wideband PA or a multi-carrier PA (MCPA)) in the radio unit. While power-saving schemes can be used for MCPA when all carriers are completely off, this approach may be inflexible and may only be implemented when network load is very low, making the number of carriers budgeted for the network no longer necessary to meet capacity requirements. The possibility of achieving fine-grained adaptive scaling by turning component carriers (CCs) on and off is also nonexistent. Furthermore, using CCs may result in some waste due to the need for a certain amount of guard band in actual transmission.

[0041] Due to the control overhead signaling of wide-bandwidth transmission, there may be some issues in 5G NR. For example, the following issues regarding higher baseline power consumption can be attributed to system configurations present in 5G: • When the UE is in active / connected mode and no data is being transmitted, only the Physical Downlink Control Channel (PDCCH) is transmitted, which reflects the baseline power consumption of the BS. Even in this mode, the BS power consumption continues to increase as bandwidth and subcarrier spacing (SCS) increase. • For maximum throughput scenarios, higher throughput means higher power consumption. However, at the same time, despite the inherent efficiency built into the 5G standard specifications, normalized power can be higher compared to the previous generation.

[0042] These issues highlight the higher effective power dissipation over a wider bandwidth.

[0043] Another issue with previous approaches may involve system performance optimization using BWP adaptation. Using BWP adaptation for the UE will impact system performance. While the expected impact is minimal under high load (since there is little benefit from using BWP, as the base station can utilize most of the allocated spectrum to support overall service demand), bursty service requests from the UE and intermittent high transmissions can lead to interference problems. This can be understood as follows: when considering how the base station transmission bandwidth remains constant, and therefore as the UE expands and contracts the BWP according to demand, the bit and power loads of adjacent subcarriers also change intermittently, resulting in intercarrier interference.

[0044] This technology can be implemented to alleviate some of these problems while ensuring that the base station has the best (or improved) energy efficiency to meet service demands.

[0045] Another issue with the previous approach may involve RF tuning. When the UE is expected to perform RF adaptation due to BWP handover, the physical RF retuning time can be considered, which may be referred to as handover latency. The following measurements can be considered regarding handover time: • For in-band operation (below 6 GHz), the conversion time can be up to 20 microseconds (μs) if the center frequency is the same before and after bandwidth adaptation. • For in-band operation (at least below 6 GHz), the conversion time can be 50–200 μs if the center frequency is different before and after bandwidth adaptation.

[0046] This technology can be implemented to facilitate the use and adaptation of DL BWP. It can also be implemented to adapt the transmission bandwidth of the base station radio unit (RU) according to service requirements, wherein the number of resource blocks (RBs) required to meet service requirements can be included within a bandwidth of "M" MHz, where the total cell bandwidth allocated by the network operator's spectrum can be "B" MHz, where B > M.

[0047] At any given time, only one DL BWP configuration is allowed to be active in the serving cell (SCell). In some examples, to comply with operator BW allocation, the configured BWP cannot be greater than the maximum BW allocated to the operator for that network, and connected UEs are not expected to receive signals outside of the active DL BWP configuration. Furthermore, if a UE's BWP implemented on its own terminal is smaller than the cell-level BWP used by the base station for that cell, the UE BWP may need to be smaller than the base station DLBWP. In other words, the base station BWP should be a superset of all UE BWPs to ensure that UE connectivity and requirements are always met.

[0048] The base station scheduler can further reconfigure the spectrum allocated to the UE to compress the allocated downlink bandwidth, thereby helping to achieve a smaller BWP without affecting the UE's Quality of Service (QoS). To activate bandwidth scaling, it should result in significant network energy savings. In some examples, bandwidth scaling can be further combined with other network energy-saving mechanisms, such as Advanced Sleep Mode (ASM) management of the base station, which allows various hardware modules within the base station to enter a sleep state based on time. This can facilitate reducing the overall energy consumption of the base station by simultaneously utilizing both time and frequency dimensions. Furthermore, where higher-level entities such as the RAN Intelligent Controller (RIC) can predict service patterns with high accuracy, this combination of bandwidth scaling and time-domain hardware downtime can be coordinated, for example, by using machine learning (ML)-based predictions, triggering one or both based on KPI constraints.

[0049] This technology can facilitate adaptive bandwidth activation / deactivation and handover. In some examples, reduced bandwidth operation based on total cell service demand can achieve network energy savings. However, changing cell bandwidth can impact overall network operation, including how UEs monitor control channels and request additional services from base stations. Therefore, the triggering and modification amount of bandwidth adaptation, as well as the duration for which the bandwidth will remain in the new state, need to be determined. Methods to facilitate these aspects are described below, including two specific triggers for bandwidth re-adaptation: • BW adaptation based on business thresholds (BWAdapt_DEM_TRIGGER), and • Duration-based BW adaptation (BWAdapt_DUR_TRIGGER).

[0050] BWAdapt_DEM_TRIGGER may include a signal generated by a bandwidth adaptation logic block (described herein) within the base station, which is based on a prediction of a reduction in traffic over a considerable period of time, such that the time required to reduce the transmission bandwidth and reset it to a wider bandwidth is negligible compared to the time it will take for the base station to operate with the reduced bandwidth.

[0051] BWAdapt_DUR_TRIGGER can include the following signals. For UE BW adaptation, a BWP inactivity timer can be used to return to the default BWP configuration. In contrast, according to this technique, base station duration triggering can be used in different ways. When the entire cell bandwidth is unused for a specific duration configurable by the system designer, BWAdapt_DUR_TRIGGER can be set to 1, and the base station can be configured for BWP usage reduced by a specific factor. An example of this method will be described herein. The duration can be measured by a counter: the counter begins to increment when the base station is configured to use the entire cell bandwidth and the normalized bandwidth usage (which can be defined as used bandwidth divided by the total cell bandwidth) is less than a threshold amount represented by NORM_BW_THRESH in consecutive subframes. In some examples, the value of NORM_BW_THRESH can be empirically set to 0.8. Once the base station switches to a lower bandwidth configuration, BWAdapt_DUR_TRIGGER can be reset to 0. The counter used to set BWAdapt_DUR_TRIGGER can remain at 0 until the system uses the full cell bandwidth again, and the count will start incrementing again when the normalized utilized bandwidth is below NORM_BW_THRESH.

[0052] In some examples, BWAdapt_DEM_TRIGGER can typically be expected to initiate BWP adaptation at the base station. However, a second trigger can create an opportunity for BW adaptation if the service thresholds have been set leniently and the base station is still adapting to appropriate service thresholds (in some examples this can be set by a deep reinforcement learning (DRL) agent that learns from the environment to establish optimal (or improved) thresholds). The operating environment can set a reward for the reinforcement learning (RL) agent, where improvements in network energy savings (NES) resulting from the agent's actions receive a positive reward, while latency caused by BA triggered by the RL agent's recommendations receives a negative reward.

[0053] This technology can facilitate adaptive bandwidth scaling for downlink transmissions. The transmission bandwidth used by the base station for downlink transmissions can dynamically vary according to network load to promote network energy efficiency (NES) at the cellular level. When scheduling large amounts of data for different UEs (whereby each UE can further implement a BWP on its terminal to reduce the bandwidth monitored by the UE itself), the base station can use wide bandwidth (up to the maximum bandwidth allocated to the operator for the cell) while operating on narrow bandwidth for the remaining time. The base station can support very wide channel bandwidths, which some UEs may not support. This technology can be implemented to support devices with different bandwidth capabilities by configuring the base station to communicate with the device using different BWPs. BWPs can provide a flexible mechanism for allocating radio resources so that the signal for the UE is limited to a portion of the base station channel bandwidth that the UE can support.

[0054] This technology can be implemented to facilitate BWP adaptation using services predicted by ML. RF front-end devices can include elements with finite activation (rise) and deactivation (fall) times. Device implementations need to consider these meta-states and provide some advance notice of when the transmission bandwidth transitions, as relevant control information may need to be provided to the connected UE via, for example, broadcast information or downlink control information (DCI) so that the UE can also reduce its scanning bandwidth according to the UE BWP process in the 5G NR standard. Furthermore, while AI / ML predicted services can create opportunities to use reduced bandwidth and power consumption (which may involve shutting down relevant baseband processing modules), in some examples, these bandwidth scaling processes may only be initiated if both of the following conditions are met: • The transition to higher / lower bandwidth usage can last for a considerable period of time, making the transition time negligible, for example, less than 0.1% of the time the base station is in reduced / increased BWP state. • Network energy savings measured by a predefined lookup table (LUT) associated with the radio unit / RAN module are higher than the target minimum savings granularity—for example, 5% in full-power mode.

[0055] In some examples, an AI / ML-based business forecaster can be implemented, which is configured to use previously collected business statistics and appropriate ML architectures, such as business demand forecasting based on Long Short-Term Memory (LSTM) networks, which a bandwidth adaptation (BA) logic block can use to predetermine whether a bandwidth change is needed.

[0056] In some examples, a similar goal can be achieved by considering reinforcement learning (RL) based approaches, where the RL agent learns service usage patterns and then recommends bandwidth portions suitable for base station use to meet KPI constraints and prioritize NES.

[0057] Figure 3 An example bandwidth adaptation scheme 300 for dynamic bandwidth adaptation scheduling according to an embodiment of the present disclosure is illustrated, which can facilitate network energy saving using bandwidth scaling. In some examples, multiple portions of the bandwidth adaptation scheme 300 can be used to implement... Figure 1 The system architecture consists of 100 (multiple) parts.

[0058] The bandwidth adaptation scheme 300 includes a buffer queue 302, UE_1_buffer 302A, UE_k_buffer 302B, UE_N_buffer 302N, a bandwidth adaptation logic component 304, an ML service predictor model update 306, a real-time overall service requirement 308, a media access control (MAC) scheduler 310, a PHY+RF component 312, a wireless channel 314, and user_1 316A, user_k 316B, and user_N 316N.

[0059] This technology can facilitate energy-efficient scheduling with dynamic bandwidth adaptation. In scheduling examples, even when available resources are fixed, there may be complexity issues related to changing the total available resources on a quasi-static basis. In some examples, it is possible to perform a quasi-static change in DL transmission bandwidth such that the overhead of sending change signaling to connected UEs and radio unit functions does not exceed the benefits of the operation.

[0060] exist Figure 3 The document describes a Bandwidth Adaptation (BA) logic component that can be used to set the correct bandwidth for transmission. The benefit of this component lies in its ability to predict the onset of bandwidth change demands and to communicate with all users via a common DCI block when bandwidth is cut off. In this process, the BA logic module can utilize an ML service predictor, which can be configured to use a combination of offline and online learning. ML Business Forecaster Model Update The component can be configured to receive input regarding real-time input requirements from UEs connected to the BS. This information can then be used to continuously refine the service predictor model and provide timely updates to the BA logic components.

[0061] The Business Logic (BA) component can use the predicted business demand information as follows: • If the predicted bandwidth requirement, based on the current modulation and coding scheme (MCS) level (per scheduled UE) and buffer queue state, is more than N_REDUCE physical resource blocks (PRBs) lower than the currently used bandwidth, the BA logic component can increment BA_LOWER_CNT by 1. (N_REDUCE can be a system-defined value, and in some examples, it can be equal to 12, 18, 24, 36... and higher.) • When BA_LOWER_CNT reaches BA_LOWER_THRESH, the BA logic component can recommend lower bandwidth usage to the MAC scheduler. System administrators can set BA_LOWER_THRESH to avoid frequent BA. BA_LOWER_CNT can be reset to 0. • When a small portion of the total cell bandwidth is in use and the service demand exceeds the currently used bandwidth, BA_HIGHER_CNT can be incremented. • When BA_HIGHER_CNT reaches BA_HIGHER_THRESH, the BA logic component can recommend higher bandwidth usage to the MAC scheduler. System administrators can set BA_HIGHER_THRESH to avoid frequent BA. BA_HIGHER_CNT can be reset to 0. BA_HIGHER_CNT can remain at 0 when the entire system bandwidth is utilized. • When the MAC scheduler accepts a BA change based on the recommendation of the BA logic component and the buffer queue status, it can send the corresponding control signal to the UE N subframes in advance (where N>3) via the PDCCH channel to allow the UE to also make the necessary changes.

[0062] When downlink bandwidth is extended, it may not necessarily mean that uplink bandwidth is also extended. This is because, in order to schedule the UE within the extended bandwidth, the base station may need to restart the uplink channel estimation module to send the recommended MCS information to the UE. Therefore, in some examples, for uplink BA, there may be modem-dependent lag in uplink bandwidth extension.

[0063] By reducing the used transmission bandwidth, the latency caused by the UE will hardly increase because the bandwidth reduction can be based on the total service demand. For a more aggressive approach to energy saving, a BWP reduction of more than the minimum required bandwidth of the total service demand can be used, but this has a corresponding impact on latency. This approach can be implemented in some examples when the base station primarily handles non-guaranteed bit rate (GBR) services.

[0064] The following is an example of how adaptive bandwidth scaling provides a more economical approach to network power consumption, taking into account service characteristics and adapting to DL operation bandwidth.

[0065] Examples of implementing this technique could involve building a bandwidth-adaptive power consumption model. Significant improvements in RAN operation energy efficiency can be achieved when base stations are allowed to change the overall DL transmission bandwidth at a finer granular scale than carrier all-on / all-off. However, the building blocks for this process could be scalable power consumption models that depend on the DL operating bandwidth and can potentially represent design optimizations within the radio unit for wide-bandwidth operation. Figure 4 An example is provided: how to construct a portion of the bandwidth from the entire cell bandwidth so that using this reduced BWP can achieve significant energy savings, even when choosing a suitable BWP involves additional overhead and computation.

[0066] In some examples, such tables can also incorporate power savings from baseband processing. These savings may be relatively smaller than those achievable through scaling via RF processing.

[0067] Figure 4 An example relative bandwidth portion 400 according to an embodiment of the present disclosure is illustrated, which can facilitate bandwidth adaptation for downlink communication at a base station and can facilitate network energy saving using bandwidth scaling. In some examples, portions(s) of the relative bandwidth portion 400 can be used to implement... Figure 1 The system architecture consists of 100 (multiple) parts.

[0068] The relative bandwidth section 400 includes bandwidth adaptation size 402, adaptive BWP_1 404, adaptive BWP_2 406, adaptive BWP_3 408, adaptive BWP_4 410, and a network power-saving component 412 using bandwidth scaling (which may be similar to...). Figure 1 (Using a network power-saving component 108 with bandwidth scaling).

[0069] Different examples of implementing this technology may involve adaptive BWP for DL ​​transmission. To facilitate the implementation of adaptive bandwidth scaling for DL, various variations of BW slicing can be considered, where the maximum number of physical resource blocks, denoted by PRB_max, can depend on the maximum bandwidth allocated to the operator for that cell. Figure 3 The image shows some of these power-saving combinations used for adaptive bandwidth scaling:

[0070] For some base station operations, all base stations are expected to start at full bandwidth upon startup, thus consuming "full power." Subsequently, depending on the service's bandwidth usage (BW), 25%, 50%, and 75% of the full bandwidth can be used as preset values ​​to adapt to service requirements. In these examples, the preset level number is 4, and these examples can more generally indicate the relatively low-overhead transmission bandwidth adaptation that may be needed to prevent frequent scaling. For examples that can tolerate larger overhead based on the transmission scenario, finer-grained steps of 10% or 20% can be implemented.

[0071] Figure 5 An example lifecycle management technique 500 for bandwidth adaptation at a base station, according to an embodiment of the present disclosure, is illustrated, which can facilitate network energy saving using bandwidth scaling. In some examples, portions(s) of the lifecycle management technique 500 can be used to implement... Figure 1 The system architecture consists of 100 (multiple) parts.

[0072] Lifecycle management technology 500 includes the initial BW phase 502, reduced BW phase 1 - BWP 1 504, reduced BW phase 2 - BWP 2 506, default BW phase - BWP_default 508, and a network power saving component 510 using bandwidth scaling (which can be similar to...). Figure 1 (Using a network power-saving component 108 with bandwidth scaling).

[0073] This technology can facilitate bandwidth adaptation throughout the base station's BWP (Bandwidth, Wiring, and Potential) lifecycle. For example... Figure 4 The diagram shown illustrates the use of various bandwidth components, from which... Figure 5 As can be seen, due to this bandwidth scaling, the base station can be in different BW phases, and thus in different power consumption phases. When the base station starts up, it may occupy the entire allocated spectrum, and thus, the initial state (502) may not provide any power reduction. In some examples, bandwidth reduction (or bandwidth increase when it is in a narrow BWP state) can be triggered by the triggers described above.

[0074] Figure 5It is also shown that for the initial BW phase 502 (which utilizes full bandwidth), the base station can enter reduced BW phase 1 (BWP 1 504) via BWAdapt_DEM_TRIGGER, and subsequently enter the lower-power reduced BW phase 2 (BWP 2 506) via an additional BWAdapt_DEM_TRIGGER. However, if the base station remains in reduced BW phase 2 (BWP 2 506) for a sufficient period of time without receiving any further service triggers, BWAdapt_DUR_TRIGGER can reset the base station to 75% of the BW phase (e.g., the default BW phase - BWP_default 508), (for example) to ensure that the goal of reducing power consumption does not compromise the latency KPI.

[0075] Figure 6 The illustration depicts an example process 600 that facilitates network energy saving in a base station using bandwidth scaling, according to embodiments of the present disclosure. In some examples, one or more embodiments of process 600 may be derived from... Figure 1 Network power saving components using bandwidth scaling 108 or Figure 11 The computing environment is 1100 to achieve this.

[0076] It should be understood that the operation procedure of process 600 is an example operation procedure, and there may be embodiments that implement more or fewer operation procedures than the described procedure, or embodiments that implement the described operation procedures in a different order than the described order. In some examples, process 600 may be related to... Figure 7 Process 700 Figure 8 Process 800 Figure 9 Process 900, and / or Figure 10 The process 1000 is implemented by combining one or more embodiments of one or more of them.

[0077] Process 600 starts from 602 and moves to operation 604.

[0078] Operation 604 describes the allocation of an adaptive cell-specific bandwidth portion (BWP) for facilitating cellular network communication, wherein the adaptive cell-specific bandwidth portion comprises a set of bandwidth sizes, and wherein a corresponding size in this set of bandwidth sizes is associated with a corresponding frequency range around the same center frequency of the cell site. That is, multiple BWP sizes may exist that can be switched between, such as... Figure 4 The examples shown are those.

[0079] After operation 604, process 600 moves to operation 606.

[0080] Operation 606 describes a transition from a first bandwidth size to a second bandwidth size within the same set of bandwidth sizes, where the second bandwidth size is smaller than the first bandwidth size, based on a determination, according to a first criterion, that a second bandwidth size within the set of bandwidth sizes is sufficient to serve the predicted network traffic. This may include starting from an initial maximum bandwidth size (the first bandwidth size) and determining that the predicted traffic conditions allow for a transition to a smaller bandwidth size (the second bandwidth size).

[0081] In some examples, the conversion from a first bandwidth size in the same set to a second bandwidth size in the same set is performed for the base station's downlink bandwidth and is independent of the base station's uplink bandwidth. That is, adaptive bandwidth size adjustment can be performed for downlink communication without affecting uplink bandwidth.

[0082] In some examples, the transition from a first bandwidth size in the group of bandwidth sizes to a second bandwidth size in the group is performed based on determining that the energy savings associated with the transition to the second bandwidth size are greater than a threshold savings. That is, a transition can be made when a minimum energy saving is identified. For example, network energy savings measured by a predetermined LUT associated with the radio unit / RAN module are higher than the target minimum saving granularity, such as 5% in full-power mode.

[0083] In some examples, the transition from a first bandwidth size in the group of bandwidth sizes to a second bandwidth size in the group is performed based on the following: determining that the predicted amount of time for using the second bandwidth size is at least a threshold amount of time longer than the amount of time associated with the transition from the first bandwidth size to the second bandwidth size. That is, a transition to another bandwidth size can be performed if it is determined that the use of the higher / lower bandwidth can continue for a sufficiently long duration such that the transition time is negligible (e.g., less than 0.1% of the time the base station is in a reduced / increased bandwidth portion state).

[0084] After operation 606, process 600 moves to operation 608.

[0085] Operation 608 describes facilitating communication on the first cellular network based on a second bandwidth size. This could include using a smaller bandwidth size from operation 606 to communicate with user equipment.

[0086] After operation 608, process 600 moves to operation 610.

[0087] Operation 610 describes a process where, after switching from a first bandwidth size to a second bandwidth size, the process switches back to the first bandwidth size based on the determination that a second criterion has been met. This may include periodically switching back to the default bandwidth size and determining whether a smaller bandwidth size can continue to be used.

[0088] In some examples, the second criterion is specified relative to a threshold network traffic volume. This could be BWAdapt_DEM_TRIGGER. In other examples, the second criterion is specified relative to the amount of time elapsed. This could be BWAdapt_DUR_TRIGGER.

[0089] After operation 610, process 600 moves to operation 612.

[0090] Operation 612 describes facilitating second cellular network communication based on the first bandwidth size after switching from a second bandwidth size to a first bandwidth size. That is, the base station can continue to communicate with the UE while using the first bandwidth size.

[0091] In some examples, operation 612 includes: after switching from a second bandwidth size to a first bandwidth size, determining a new bandwidth size, wherein the new bandwidth size is used to facilitate third cellular network communication, wherein determining the new bandwidth size includes determining whether to remain at the first bandwidth size or to switch to another bandwidth size in the set of bandwidth sizes that is different from the first bandwidth size. That is, the purpose of periodically returning to the full bandwidth is to reassess the bandwidth size to be used.

[0092] After operation 612, process 600 moves to 614, where process 600 ends.

[0093] Figure 7 The illustration depicts an example process 700 according to embodiments of the present disclosure, which facilitates network energy saving in a base station by using bandwidth scaling. In some examples, one or more embodiments of process 700 may be derived from... Figure 1 Network power saving components using bandwidth scaling 108 or Figure 11 The computing environment is 1100 to achieve this.

[0094] It should be understood that the operation process of process 700 is an example operation process, and there may be embodiments that implement more or fewer operation processes than depicted, or embodiments that implement the depicted operation processes in a different order than depicted. In some examples, process 700 may be combined with... Figure 6 Process 600 Figure 8 Process 800 Figure 9 Process 900, and / or Figure 10The process 1000 is implemented by one or more embodiments of one or more of them.

[0095] Process 700 starts from 702 and moves to operation 704.

[0096] Operation 704 describes maintaining a set of bandwidth sizes for cellular network communication, where each bandwidth size corresponds to a corresponding frequency range. In some examples, operation 704 can be performed in accordance with... Figure 6 It can be implemented in a similar way to the 604 operation.

[0097] After operation 704, process 700 moves to operation 706.

[0098] Operation 706 describes a conversion from a first bandwidth size to a second bandwidth size in the same set of bandwidth sizes, where the second bandwidth size is smaller than the first bandwidth size, based on a determination, according to a first criterion, that a second bandwidth size in the set of bandwidth sizes is sufficient to serve the predicted network traffic. In some examples, operation 706 may be performed in accordance with... Figure 6 It can be implemented in a similar way to operation 606.

[0099] In some examples, operation 706 includes sending an indication to the user equipment (UE) to switch to a second bandwidth size, the UE reducing its scan bandwidth corresponding to the second bandwidth size. In some examples, the indication includes broadcast information. In some examples, the indication includes downlink control information. In some examples, the timing of sending the indication is determined based on the amount of time associated with bandwidth deactivation. That is, the UE can be notified in advance when the transmission bandwidth switch will occur, wherein the relevant control information is provided to the connected UE via broadcast information or via DCI, so that the UE can also reduce its scan bandwidth.

[0100] In some examples, Operation 706 is executed based on predicting future network traffic using a Long Short-Term Memory (LSTM) model trained on previous network traffic statistics. That is, previously collected traffic statistics and ML structures (such as LSTM-based traffic demand forecasting) can be applied to determine whether to change the bandwidth.

[0101] In some examples, operation 706 is performed based on using a reinforcement learning model to predict future network traffic, where the reward for the reinforcement learning model is defined based on meeting key performance indicator constraints of the future network traffic and on energy savings associated with switching to a second bandwidth size. That is, an RL method can be implemented whereby the RL agent learns traffic usage patterns and then recommends suitable bandwidth portions for base station use, prioritizing network energy conservation while meeting KPI constraints.

[0102] After operation 706, process 700 moves to operation 708.

[0103] Operation 708 describes transmitting first cellular network communication according to a second bandwidth size. In some examples, operation 708 can be performed in accordance with... Figure 6 It can be implemented in a similar way to operation 608.

[0104] After operation 708, process 700 moves to operation 710.

[0105] Operation 710 describes a process where, after switching from a first bandwidth size to a second bandwidth size, the user switches back to the first bandwidth size based on the determination that a second criterion has been met. In some examples, operation 710 can be performed in accordance with... Figure 6 It can be implemented in a similar way to operation 610.

[0106] After operation 710, process 700 moves to operation 712.

[0107] Operation 712 describes transmitting second cellular network communication according to a first bandwidth size. In some examples, operation 712 can be performed in accordance with... Figure 6 It can be implemented in a similar way to operation 612.

[0108] After operation 712, process 700 moves to 714, where process 700 ends.

[0109] Figure 8 The illustration depicts an example process 800 that facilitates network energy saving in a base station by using bandwidth scaling, according to embodiments of the present disclosure. In some examples, one or more embodiments of process 800 may be derived from... Figure 1 Network power saving components using bandwidth scaling 108 or Figure 11 The computing environment is 1100 to achieve this.

[0110] It should be understood that the operation process of process 800 is an example operation process, and there may be embodiments that implement more or fewer operation processes than depicted, or embodiments that implement the depicted operation processes in a different order than depicted. In some examples, process 800 may be combined with... Figure 6 Process 600 Figure 7 Process 700 Figure 9 Process 900, and / or Figure 10 The process 1000 is implemented by one or more embodiments of one or more of them.

[0111] Process 800 starts from operation 802 and moves to operation 804.

[0112] Operation 804 describes a process of transitioning from a first bandwidth size to a second bandwidth size in a set of bandwidth sizes, where the second bandwidth size is smaller than the first bandwidth size, based on determining that a second bandwidth size in a set of bandwidth sizes is a threshold capable of serving predicted cellular network traffic. In some examples, operation 804 can be performed in accordance with... Figure 6 It can be implemented in a similar way to operations 604-606.

[0113] In some examples, Operation 804 is executed based on predicting future network traffic using a trained model, where the trained model is trained using a combination of offline and online learning. That is, the determination of the bandwidth change can be based on an ML traffic predictor using a combination of offline and online learning.

[0114] In some examples, operation 804 is performed based on predicting future network traffic using a trained model, wherein the trained model is iteratively trained based on real-time input requests received from user equipment in a radio resource control connection state. That is, input from UEs connected to the BS regarding real-time input requests can be used to perform traffic prediction.

[0115] In some examples, the second bandwidth size is smaller than the bandwidth amount associated with the total service demand, and the conversion from the first bandwidth size in this set of bandwidth sizes to the second bandwidth size in this set of bandwidth sizes is performed based on determining that the guaranteed bit rate service volume in the total service demand is less than a threshold amount. That is, in some examples, to improve energy efficiency, a BWP reduction greater than the minimum required bandwidth of the total service demand can also be used, but this has a latency-related impact. For example, this approach can be implemented when the base station primarily handles non-GBR services.

[0116] In some examples, the conversion from a first bandwidth size in the group of bandwidth sizes to a second bandwidth size in the group is performed based on determining that an event has occurred a threshold number of times, wherein the event includes: determining physical resource blocks whose predicted bandwidth level, based on the current modulation and coding scheme and the buffer queue state, is at least a threshold number lower than the first bandwidth size. This is similar to... Figure 9 The process is 900.

[0117] After operation 804, process 800 moves to operation 806.

[0118] Operation 806 describes facilitating communication in the first cellular network according to the second bandwidth size. In some examples, operation 806 can be performed in accordance with... Figure 6 It can be implemented in a similar way to operation 608.

[0119] After operation 806, process 800 moves to operation 808.

[0120] Operation 808 describes a process where, after a transition from a first bandwidth size to a second bandwidth size, a transition from the second bandwidth size back to the first bandwidth size is performed based on the fact that a certain criterion has been met. In some examples, operation 808 can be performed in accordance with... Figure 6 It can be implemented in a similar way to operation 610.

[0121] In some examples, the criteria for determining whether business demand has been met include: identifying the number of times business demand exceeds a second bandwidth size threshold. This could be similar to... Figure 10 The process is 1000.

[0122] After operation 808, process 800 moves to operation 810.

[0123] Operation 810 describes facilitating second cellular network communication based on the first bandwidth size. In some examples, operation 810 can be performed in accordance with... Figure 6 It can be implemented in a similar way to operation 612.

[0124] After operation 810, process 800 moves to operation 812, where process 800 ends.

[0125] Figure 9 An example process 900 for reducing bandwidth in a base station to facilitate network energy saving using bandwidth scaling, according to an embodiment of the present disclosure, is illustrated. In some examples, one or more embodiments of process 900 may be derived from... Figure 1 Network power saving components using bandwidth scaling 108 or Figure 11 The computing environment is 1100 to achieve this.

[0126] It should be understood that the operation procedure of process 900 is an example operation procedure, and there may be embodiments that implement more or fewer operation procedures than depicted, or embodiments that implement the depicted operation procedures in a different order than depicted. In some examples, process 900 may be combined with... Figure 6 Process 600 Figure 8 Process 800 Figure 9 Process 900, and / or Figure 10 The process 1000 is implemented by one or more embodiments of one or more of them.

[0127] Process 900 starts from 902 and moves to operation 904.

[0128] Operation 904 describes determining whether the predicted bandwidth is more than N_REDUCE PRBs lower than the currently used bandwidth. This can be based on the predicted bandwidth requirement of each scheduled UE's current MCS level and buffer queue state. N_REDUCE can be a system-defined value.

[0129] If in operation 904 it is determined that the predicted bandwidth is more than N_REDUCE PRBs lower than the currently used bandwidth, process 900 moves to operation 906. Conversely, if in operation 904 it is determined that the predicted bandwidth is not less than or equal to more than N_REDUCE PRBs lower than the currently used bandwidth, process 900 moves to 912, where process 900 ends.

[0130] Operation 904 proceeds to operation 906, which determines that the predicted bandwidth is more than N_REDUCE PRBs lower than the currently used bandwidth. Operation 906 describes incrementing BA_LOWER_CNT. This may include adding 1 to the stored value of BA_LOWER_CNT.

[0131] After operation 906, process 900 moves to operation 908.

[0132] Operation 908 describes determining whether BA_LOWER_CNT is equal to BA_LOWER_THRESH. BA_LOWER_THRESH can be a parameter designed into the system.

[0133] If it is determined in operation 908 that BA_LOWER_CNT equals BA_LOWER_THRESH, then process 900 moves to operation 910. Conversely, if it is determined in operation 908 that BA_LOWER_CNT equals BA_LOWER_THRESH, then process 900 moves to operation 912, where process 900 ends.

[0134] Operation 908, which determines that BA_LOWER_CNT equals BA_LOWER_THRESH, leads to operation 910. Operation 910 describes recommending a lower bandwidth to the MAC scheduler. BA_LOWER_CNT can be set to zero for use in subsequent execution of flow 900.

[0135] After operation 910, process 900 moves to operation 912, where process 900 ends.

[0136] Process 900 can be implemented periodically to determine whether to reduce bandwidth while the base station facilitates network communication.

[0137] Figure 10The illustration depicts an example process 1000 for increasing bandwidth in a base station to facilitate network energy saving using bandwidth scaling, according to embodiments of the present disclosure. In some examples, one or more embodiments of process 1000 may be derived from... Figure 1 Network power saving components using bandwidth scaling 108 or Figure 11 The computing environment is 1100 to achieve this.

[0138] It should be understood that the operation process of process 1000 is an example operation process, and there may be embodiments that implement more or fewer operation processes than depicted, or embodiments that implement the depicted operation processes in a different order than depicted. In some examples, process 1000 may be combined with... Figure 6 Process 600 Figure 8 Process 800 Figure 10 Process 1000, and / or Figure 10 The process 1000 is implemented by one or more embodiments of one or more of them.

[0139] Process 1000 starts from 1002 and moves to operation 1004.

[0140] Operation 1004 describes determining whether less than the total cell bandwidth is being used. That is, less than 100% of the total available bandwidth may be being used.

[0141] If it is determined in operation 1004 that a bandwidth less than the total cell bandwidth is being used, process 1000 moves to operation 1006. Conversely, if it is determined in operation 1004 that the total cell bandwidth is being used, process 1000 moves to 1014, where process 1000 ends.

[0142] Operation 1004 proceeds to operation 1006, which determines that the bandwidth being used is less than the total cell bandwidth. Operation 1006 describes determining whether the service demand exceeds the current bandwidth.

[0143] If it is determined in operation 1006 that the service demand exceeds the current bandwidth, process 1000 moves to operation 1008. Conversely, if it is determined in operation 1006 that the service demand does not exceed the current bandwidth, process 1000 moves to 1014, where process 1000 ends.

[0144] The process proceeds from operation 1006, which determines that the service exceeds the current bandwidth, to operation 1008. Operation 1008 describes incrementing BA_HIGHER_CNT. This may include adding 1 to the stored value of BA_HIGHER_CNT.

[0145] After operation 1008, process 1000 moves to operation 1010.

[0146] Operation 1010 describes determining whether BA_HIGHER_CNT is equal to BA_HIGHER_THRESH. BA_HIGHER_THRESH can be a system-defined parameter.

[0147] If it is determined in operation 1010 that BA_HIGHER_CNT is equal to BA_HIGHER_THRESH, then process 1000 moves to operation 1012. Conversely, if it is determined in operation 1012 that BA_HIGHER_CNT is not equal to BA_HIGHER_THRESH, then process 1000 moves to operation 1014, where process 1000 ends.

[0148] Operation 1010, which determines that BA_HIGHER_CNT equals BA_HIGHER_THRESH, leads to operation 1012. Operation 1012 describes recommending a higher bandwidth to the MAC scheduler. BA_HIGHER_CNT can be reset to zero for future instances of process 1000. When the MAC scheduler accepts the change based on the recommendation and buffer queue state, it can send a corresponding control signal to the UE N subframes in advance (where N>3) via the PDCCH channel to allow the UE to also make the change.

[0149] After operation 1012, process 1000 moves to operation 1014, where process 1000 ends.

[0150] Process 1000 can be implemented periodically to determine whether to increase bandwidth while the base station facilitates network communication. Example operating environment

[0151] To provide additional context for the various embodiments described herein, Figure 11 The following discussion is intended to provide a brief, general description of a suitable computing environment 1100 that can implement the various embodiments described herein.

[0152] For example, the various parts of computing environment 1100 can be used to implement Figure 1 One or more embodiments of base station 102 and / or user equipment 104.

[0153] In some examples, computing environment 1100 can achieve this. Figures 6-10 One or more embodiments of the process are used to facilitate network energy saving in base stations by using bandwidth scaling.

[0154] Although the embodiments have been described above in the general context of computer-executable instructions that can run on one or more computers, those skilled in the art will recognize that these embodiments may also be implemented in combination with other program modules, and / or as a combination of hardware and software.

[0155] Typically, program modules include routines, programs, components, data structures, etc., that perform specific tasks or implement specific abstract data types. Furthermore, those skilled in the art will appreciate that various approaches can be implemented using other computer system configurations, including single-processor or multi-processor computer systems, minicomputers, mainframes, Internet of Things (IoT) devices, distributed computing systems, and personal computers, handheld computing devices, microprocessor-based or programmable consumer electronics, each of which can be operatively coupled to one or more associated devices.

[0156] The embodiments described herein can also be practiced in a distributed computing environment, where certain tasks are performed by remote processing devices linked via a communication network. In a distributed computing environment, program modules can reside on both local and remote memory storage devices.

[0157] Figure 11 An example block diagram 1100 of a computer operable to perform embodiments of the present disclosure is illustrated. The UEs (multiple) UEs 1104A, 1104B, and 1104N typically include devices used by end users to access a communication network. The UEs can be configured to communicate with the core network 1108.

[0158] Core network 1108 can receive and process messages from UE, and core network 1108 may include components of third-generation (3G), 4G, Long Term Evolution (LTE), 5G, or other wireless communication networks. Core network 1108 can be configured to establish a connection with UE 1104, for example by facilitating services such as connectivity and mobility management, authentication and authorization, subscriber data management, and policy management. Messages sent between UE and communication network 1106 can be propagated through one of base station 1118A (which includes centralized unit (CU) 1110, DU 1112, RU 1114, and antenna 1116), base station 1118B, and base station 1118N.

[0159] CU 1110 can be configured to handle non-real-time Radio Resource Control (RRC) and Packet Data Convergence Protocol (PDCP) communications. DU 1112 can be configured to handle communications transmitted according to the Radio Link Control (RLC), MAC, and PHY layers. RU 1114 can be configured to convert radio signals sent to antenna 1116 from digital packets to radio signals, and to convert radio signals received from antenna 1116 from radio signals to digital packets. Antenna 1116 (which may include a transceiver) can be configured to transmit and receive radio waves used to convey information. in conclusion

[0160] Computing devices typically include various media, which may include computer-readable storage media, machine-readable storage media, and / or communication media. These two terms are used interchangeably herein as follows: A computer-readable storage media or a machine-readable storage media can be any available storage medium that is accessible to a computer, and includes volatile and non-volatile media, removable and non-removable media. By way of example and not limitation, a computer-readable storage media or a machine-readable storage media can be implemented in conjunction with any method or technology used for storing information, such as computer-readable or machine-readable instructions, program modules, structured data, or unstructured data.

[0161] Computer-readable storage media may include, but are not limited to: random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, other optical disc storage devices, solid-state drives or other solid-state storage devices, or other tangible and / or non-transient media that can be used to store desired information. In this regard, the terms “tangible” or “non-transient” used herein to describe storage devices, memories, or computer-readable media should be understood to exclude the use of only transient signals as modifiers, and without waiving the rights to all standard storage devices, memories, or computer-readable media that do not only propagate transient signals.

[0162] Computer-readable storage media can be accessed by one or more local or remote computing devices (e.g., via access requests, queries, or other data retrieval protocols) to perform various operations on the information stored in the media.

[0163] Communication media typically embody computer-readable instructions, data structures, program modules, or other structured or unstructured data in the form of data signals (such as modulated data signals, like carrier waves or other transmission mechanisms), and include any medium for delivering or transmitting information. The term "(multiple) modulated data signals" refers to signals that encode information in one or more signals by setting or altering one or more of their characteristics. As an example, and not a limitation, communication media include wired media (such as wired networks or direct wired connections) and wireless media (such as acoustic waves, RF, infrared, and other wireless media).

[0164] In this specification, terms such as “data repository,” “data storage device,” “database,” “cache,” and any other information storage component substantially related to the operation and function of a component refer to a “memory component,” or an entity embodied in “memory” or a component that includes memory. It should be understood that the memory component or computer-readable storage medium described herein can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. By way of illustration and not limitation, non-volatile memory may include ROM, programmable ROM (PROM), EPROM, EEPROM, or flash memory. Volatile memory may include RAM, which acts as external cache memory. By way of illustration and not limitation, RAM may take many forms, such as synchronous RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct Rambus RAM (DRRAM). Additionally, the memory components of the systems or methods disclosed herein are intended to include, but are not limited to, the foregoing and any other suitable types of memory.

[0165] The illustrative embodiments of this disclosure can be practiced in a distributed computing environment, where certain tasks are performed by remote processing devices linked via a communication network. In a distributed computing environment, program modules can reside on both local and remote memory storage devices.

[0166] The systems and processes described above can be embodied in hardware, such as a single integrated circuit (IC) chip, multiple ICs, ASICs, etc. Furthermore, the order in which some or all of the boxes in a process box appear in each process should not be considered restrictive. Rather, it should be understood that some boxes in a process box can be executed in various orders, but not all orders are explicitly stated herein.

[0167] As used herein, the terms “component,” “module,” “system,” “interface,” “cluster,” “server,” “node,” etc., are generally intended to refer to computer-related entities that can be hardware, a combination of hardware and software, software, or software in execution, or entities associated with an operating machine having one or more specific functions. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, multiple computer-executable instructions, a program, and / or a computer. As an example, both an application running on a controller and the controller itself can be components. One or more components may reside within a process and / or an execution thread, and components may reside on a single computer and / or be distributed across two or more computers. As another example, an interface may include input / output (I / O) components and associated processors, applications, and / or application programming interface (API) components.

[0168] Furthermore, various embodiments can be implemented as methods, apparatus, or articles of art, using standard programming and / or engineering techniques to produce software, firmware, hardware, or any combination thereof to control a computer to implement one or more embodiments of the disclosed subject matter. Articles of art can encompass computer programs accessible from any computer-readable device or computer-readable storage / communication medium. For example, computer-readable storage media can include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, magnetic stripes, etc.), optical disks (e.g., CDs, DVDs, etc.), smart cards, and flash memory devices (e.g., cards, sticks, key drives, etc.). Of course, those skilled in the art will recognize that many modifications can be made to this configuration without departing from the scope or spirit of the various embodiments.

[0169] Furthermore, the terms “example” or “exemplary” as used herein are intended to serve as examples, instances, or illustrations. Any embodiment or design described herein as “exemplary” is not necessarily to be construed as being more preferred or advantageous than other embodiments or designs. Rather, the use of “exemplary” is intended to present concepts in a specific manner. As used in this application, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or.” That is, unless otherwise specified or explicitly indicated from the context, “X adopts A or B” is intended to mean any natural inclusive arrangement. That is, if X adopts A; X adopts B; or X adopts both A and B, then any of the above satisfies “X adopts A or B.” Furthermore, unless otherwise specified or explicitly indicated from the context, the articles “a” and “an” as used in this application and the appended claims should generally be interpreted as meaning “one or more.”

[0170] The foregoing description includes examples from this specification. Of course, for the purposes of describing this specification, it is impossible to describe every conceivable combination of components or methods, but those skilled in the art will recognize that many further combinations and arrangements are possible. Therefore, this specification is intended to cover all such changes, modifications, and variations that fall within the spirit and scope of the appended claims. Furthermore, where the term "include" is used in the detailed description or claims, the term is intended to be inclusive in a similar manner to how the term "comprising" is interpreted when used as a transitional term in the claims.

Claims

1. A method comprising: An adaptive cell-specific bandwidth portion is allocated by the system to facilitate cellular network communication, wherein the adaptive cell-specific bandwidth portion comprises a set of bandwidth sizes, and wherein the respective size of the set of bandwidth sizes is associated with a corresponding frequency range; The system determines, based on a first criterion, that a second bandwidth size is sufficient to serve the predicted network traffic, and then switches from a first bandwidth size in the set of bandwidth sizes to a second bandwidth size in the set of bandwidth sizes, wherein the second bandwidth size is smaller than the first bandwidth size; The system facilitates communication on the first cellular network based on the second bandwidth size; After switching from the first bandwidth size to the second bandwidth size, the system switches from the second bandwidth size to the first bandwidth size based on determining that a second criterion has been met; and After switching from the second bandwidth size to the first bandwidth size, the system facilitates second cellular network communication based on the first bandwidth size.

2. The method according to claim 1, further comprising: After switching from the second bandwidth size to the first bandwidth size, the system determines a new bandwidth size to facilitate third cellular network communication using the new bandwidth size, wherein determining the new bandwidth size includes: determining to remain at the first bandwidth size, or determining to switch to another bandwidth size in the set of bandwidth sizes that is different from the first bandwidth size.

3. The method of claim 1, wherein the second criterion is specified relative to a threshold network traffic volume.

4. The method of claim 1, wherein the second criterion is specified relative to the amount of time elapsed.

5. The method of claim 1, wherein the conversion from the first bandwidth size in the set of bandwidth sizes to the second bandwidth size in the set of bandwidth sizes is performed for downlink bandwidth and is performed independently of uplink bandwidth.

6. The method of claim 1, wherein the conversion from the first bandwidth size in the set of bandwidth sizes to the second bandwidth size in the set of bandwidth sizes is performed based on determining that the predicted amount of time spent using the second bandwidth size is more than a threshold amount of time than the time associated with the conversion from the first bandwidth size to the second bandwidth size.

7. The method of claim 1, wherein the conversion from the first bandwidth size in the set of bandwidth sizes to the second bandwidth size in the set of bandwidth sizes is performed based on determining that the energy savings associated with the conversion to the second bandwidth size are greater than a threshold savings.

8. A system comprising: processor; as well as A memory storing executable instructions that, when executed by the processor, facilitate the execution of operations, including: Maintain a set of bandwidth sizes for cellular network communication, wherein each size in the set of bandwidth sizes corresponds to a corresponding frequency range; Based on the determination that the second bandwidth size is sufficient to serve the predicted network traffic according to the first criterion, the bandwidth size is converted from the first bandwidth size in the set of bandwidth sizes to the second bandwidth size in the set of bandwidth sizes, wherein the second bandwidth size is smaller than the first bandwidth size; The first cellular network communication is transmitted according to the second bandwidth size; After switching from the first bandwidth size to the second bandwidth size, based on the determination that the second criterion has been met, the bandwidth is switched from the second bandwidth size to the first bandwidth size; and The second cellular network communication is transmitted according to the first bandwidth size.

9. The system of claim 8, wherein the conversion from the first bandwidth size in the set of bandwidth sizes to the second bandwidth size in the set of bandwidth sizes comprises: An instruction to switch to the second bandwidth size is sent to the user equipment, which reduces its scan bandwidth to correspond to the second bandwidth size.

10. The system of claim 9, wherein the indication includes broadcast information.

11. The system of claim 9, wherein the indication includes downlink control information.

12. The system of claim 9, wherein the time at which the indication is sent is determined based on the amount of time associated with bandwidth deactivation.

13. The system of claim 8, wherein the conversion from the first bandwidth size in the set of bandwidth sizes to the second bandwidth size in the set of bandwidth sizes is performed based on predicting future network traffic using a long short-term memory model that has been trained on previous network traffic statistics.

14. The system of claim 8, wherein the conversion from the first bandwidth size in the set of bandwidth sizes to the second bandwidth size in the set of bandwidth sizes is performed based on predicting future network services using a reinforcement learning model, and wherein the reward of the reinforcement learning model is defined based on: satisfying key performance indicator constraints of the future network services, and energy savings associated with the conversion to the second bandwidth size.

15. A non-transient computer-readable medium comprising instructions that, in response to execution, cause a system including a processor to perform an operation, the operation comprising: Based on the determination that the second bandwidth size is a threshold that can serve the predicted cellular network traffic, the bandwidth size is switched from a first bandwidth size in a set of bandwidth sizes to a second bandwidth size in the set of bandwidth sizes, wherein the second bandwidth size is smaller than the first bandwidth size; Facilitate first cellular network communication based on the second bandwidth size; After switching from the first bandwidth size to the second bandwidth size, based on the determination criteria being met, the bandwidth size is switched back to the first bandwidth size; and Facilitate second cellular network communication based on the first bandwidth size.

16. The non-transient computer-readable medium of claim 15, wherein the conversion from the first bandwidth size in the set of bandwidth sizes to the second bandwidth size in the set of bandwidth sizes is performed based on predicting future network traffic using a trained model, and wherein the trained model is trained based on a combination of offline learning and online learning.

17. The non-transient computer-readable medium of claim 15, wherein the conversion from the first bandwidth size in the set of bandwidth sizes to the second bandwidth size in the set of bandwidth sizes is performed based on predicting future network traffic using a trained model, and wherein the trained model is iteratively trained based on real-time input requests received from user equipment in a radio resource control connection state.

18. The non-transient computer-readable medium of claim 15, wherein the second bandwidth is less than the bandwidth amount associated with total service requirements, and wherein the conversion from the first bandwidth size in the set of bandwidth sizes to the second bandwidth size in the set of bandwidth sizes is performed based on determining that the guaranteed bit rate service in the total service requirements is less than a threshold amount.

19. The non-transient computer-readable medium of claim 15, wherein the conversion from the first bandwidth size in the set of bandwidth sizes to the second bandwidth size in the set of bandwidth sizes is performed based on determining a threshold number of times an event has occurred, and wherein the event includes: Identify physical resource blocks whose predicted bandwidth level, based on the current modulation and coding scheme and the buffer queue state, is at least a threshold number lower than the first bandwidth size.

20. The non-transient computer-readable medium of claim 15, wherein determining that the criterion has been satisfied includes: Determine the number of times the business demand exceeds the second bandwidth size threshold.

Images