Dynamic Base Station Energy Saving System with Multivariate Environmental Data and Grid Load Estimation

Abstract

The invention is a system that optimizes energy consumption in mobile communication networks, taking into account weather conditions (precipitation, temperature, etc.).The system includes the following: an environmental input layer (100) into which data from the network and regional events (matches, concerts, rallies) are transferred to the system; a network input layer (101) from which real-time traffic, user count, and occupancy (KPI) data are received from base stations; a data fusion and synchronization module (102) that combines external data (100) and technical data (101) to create a "contextual data set"; an AI-based traffic forecasting unit (103) that proactively predicts future cell load with enriched data from the fusion phase; a dynamic energy mode assignment layer (104) that switches the base station to the most efficient sleep or capacity mode according to the predicted load; a hardware control and application interface (105) that transmits the agreed energy modes to physical network elements; and an energy saving and quality control panel (106) that monitors network quality.

Description

1 TARIFF Dynamic Base Station with Multivariate Environmental Data and Grid Load Estimation Energy Saving System Technical Area The invention has implications for the telecommunications and mobile communications sectors, particularly for the energy needs of mobile operators. to optimize consumption, reduce carbon footprint and lower operational costs to integrate external environmental data with network traffic density; AI-powered It is related to dynamic energy management systems. 10 State of the Art Base stations, which make up the mobile communications infrastructure, account for the operators' total energy... It constitutes the largest part of consumption. Current energy saving methods generally Simple threshold values ​​based on fixed time (such as day / night) or instantaneous traffic load only 15 It works with these approaches. However, these approaches take into account constantly changing external factors (weather, regional either insufficient energy saving because it does not take into account activities (mobility fluctuations). failing to provide or causing a decrease in network quality (KPI) during unexpected peak periods. is happening. In the mobile communications sector, energy saving processes for base stations have traditionally taken 20 years. It is carried out using two main methods: Static Time Control: Equipment is controlled during nighttime hours when traffic is assumed to be low. It is switched off or put into low power mode. Threshold-Based Reactive Control: Traffic on the cell (PRB usage, etc.) is controlled according to a specific threshold. The energy saving mode activates when the power level falls below the threshold. 25 However, these methods evaluate network dynamics using only internal and real-time data. It does not take into account developments in the outside world. Existing systems; for example, a cell It can switch to power saving mode because traffic is low, but it will start in 15 minutes. from a major event or emergency traffic that will be created by an approaching storm Because they are unaware, this can lead to sudden deteriorations in service quality (KPIs). 30 These shortcomings can be summarized as follows: Lack of External Data: Current systems only provide on-network performance data. focusing on traffic such as weather conditions, regional event calendars, or city mobility. It does not integrate external data that directly affects the situation. Limitations of Predictive Capability: Current applications make proactive projections. 35 Instead, it makes decisions (reactively) based on the current traffic situation. This situation causes sudden traffic jams. 2 Increases in these systems lead to delays in the "wake-up" time and negatively impact the user experience. This leads to it being affected. Flexibility and Scalability Issue: Energy saving policies are generally applicable to all regions. It is applied with similar rules (generalization approach). However, each cell Depending on its location (shopping mall, highway, residential area, etc.), traffic characteristics and external factors 5 His reaction was different. KPI and Energy Balance Optimization: Quality of Service through Savings Modes the balance between them is achieved manually rather than through machine learning-based optimization. It is based on established safe threshold values. This means that either potential savings... This can lead to the project not being completed or the quality being compromised. 10 Data-Intensive Analytics Inadequacy: Massive data from thousands of base stations. An advanced machine capable of processing and correlating large amounts of external data in real-time. Learning approaches are not actively used in current systems. In conclusion, current practices consider energy efficiency not only in terms of grid load but also external factors. Connecting to global variables; balancing savings and quality with proactive forecasts 15 There is no comprehensive and intelligent system capable of managing this deficiency. This invention addresses this gap. by shutting down; integrating multidimensional data sources, continuously learning and energy modes It offers a unique infrastructure that manages preventive measures. Purpose of the Invention 20 The invention is due to the inadequacy of the methods used in the current art on the subject. It was created with the aim of solving the technical problems mentioned above. The main objective of the invention is to optimize energy consumption in mobile communication networks. To achieve this, it can analyze external environmental data and technical network traffic data together. It operates entirely data-driven and helps operators reduce energy costs while improving service quality by 25%. The aim is to offer a new generation decision support system that protects against repetitive behavioral changes (RPRs). In this context, the invention focuses on the use of PRBs and... Network load data such as RRC links, weather forecasts, and regional activity. It analyzes external factors such as calendars by combining them on the same platform; thus, it analyzes traffic. The time intervals in which it will rise or fall are determined not only based on past network data, but also on external factors. It is envisioned holistically, taking into account the global context. The system uses XGBoost and LightGBM 30. or cell-based traffic using regression and classification algorithms such as LSTM It estimates density with high accuracy and in this respect differs from static time planning. or it deviates from simple threshold-based rule sets. Current systems consume energy during sudden traffic surges. While there may be a delay in exiting power saving modes, in this invention, model success is particularly noticeable in sudden bursts of intensity. It is optimized according to the accurate prediction of increases; for this purpose, the system's wake-up speed and 35 3 Accuracy metrics are used as the primary evaluation criteria. Furthermore, the system only... It doesn't work with an on / off logic; it involves reducing MIMO layers and dynamically adjusting frequency bands. such as deactivating and adjusting the depth of hardware-based sleep modes. It offers a tiered energy saving hierarchy. The process does not require manual interventions or regional planning. It is not based on fixed definitions; each cell has its own traffic pattern and external data. By learning its reaction, it can provide the most efficient energy source specifically for that cell, without needing human interpretation. It determines its policy. While energy saving is implemented, access quality metrics are monitored in real time. The system is monitored, and if there is an unexpected traffic development, it instantly switches from power saving mode. By exiting, it maintains service continuity. Thus, the invention provides total energy at the operator level. by providing a measurable decrease in consumption, both OPEX costs will be reduced and 10 It directly contributes to corporate sustainability goals. As a result, this invention not only monitors current traffic but also takes into account the influence of environmental data. able to predict how traffic will shape up and apply these predictions to proactive energy saving. a system architecture that translates into actions and does not compromise network quality It offers this. In this respect, compared to existing energy management systems in the sector, it provides both efficiency of 15%. It also provides a significant advantage in terms of network security. Figures that will help understand the invention. Figure 1 shows the general architecture of the system that is the subject of the invention. Explanation of Part References 100. Environmental Input Layer 101. Network Input Layer 102. Data Fusion and Synchronization Module 103. AI-Based Traffic Prediction Unit 25 104. Dynamic Energy Mode Assignment Layer 105. Hardware Control and Implementation Interface 106. Energy Saving and Quality Control Panel Detailed Description of the Invention 30 This detailed description of the preferred configurations of the invention provides a better understanding of the subject matter. In order to facilitate understanding and without imposing any limiting effects, the Invention, mobile In the communications sector, the energy consumption of base stations depends on both technical network performance. Advanced analytics based on both data and external environmental factors (weather, events, etc.) It is a proactive management system that optimizes using machine learning methods. The system has 35 4 cell-based technical metrics (101) and environmental variables (100) a central fusion By bringing them together in module (102), it estimates the future traffic load and this estimate Based on this, the system determines the most suitable energy saving mode. Thus, the system only uses that mode. Instead of looking at current traffic, consider energy efficiency by also taking into account upcoming external influences. It allows for the preservation of network quality while increasing performance. 5 The invention relates to: Environmental Input Layer (100), Network Input Layer (101), Data Fusion and Synchronization Module (102), Traffic Prediction Unit (103), Dynamic Energy Mode Assignment Layer (104), Hardware Control and Application Interface (105) and optional Energy Saving and consists of the Quality Control Panel (106) components. The system primarily receives weather information (precipitation, temperature, etc.) via the Environmental Input Layer (100). and collects regional event calendar data. This external data is used in the Network Input Layer (101) Data containing cell-based technical performance data (PRB usage, number of users, etc.) Integrated on a time and location basis via the Fusion and Synchronization Module (102). This data is processed and transformed into a coherent "contextual data set". The Traffic Forecasting Unit (103) uses this combined data set to predict various machines 15 future traffic with learning algorithms (e.g., XGBoost, LSTM, LightGBM) It develops models that proactively predict density and trains them at regular intervals. This Predictions obtained from the models are assigned by the Dynamic Energy Mode Assignment Layer (104). by evaluating which energy saving mode the base station is in (e.g., MIMO layer) (shutdown, micro-sleep or frequency-based capacity reduction) will automatically switch to 20 It will be decided. The agreed energy modes are controlled via the Hardware Control and Application Interface (105). Energy consumption is optimized by transmitting it to physical grid elements. Simultaneously, The system provides via the optional Energy Saving and Quality Control Panel (106). The amount of savings and network access quality are monitored in real-time. The estimates are 25%. In the event of an increase in traffic beyond what is expected, the system, thanks to its monitoring mechanism... It temporarily exits power saving mode to ensure network quality. Thus, the invention combines technical performance data with external variables. By evaluating, it enables more accurate and proactive energy management; it allows operators to perform operational tasks. It allows them to reduce their operating costs (OPEX) and decrease their carbon footprint. 30

Claims

REQUESTS 1. It is a system that optimizes energy consumption in mobile communication networks, and its feature is; weather data (precipitation, temperature, etc.) and regional event data (matches, concerts, rallies) environmental input layer to which it is transferred to the system (100), Real-time traffic, user count, and occupancy (KPI) data from base stations 5 received network input layer (101), By combining external data (100) and technical data (101) at the same time and region, "contextual Data fusion and synchronization module (102) which creates a "data set" By using enriched data from the fusion phase, we can proactively assess the future cellular load. AI-based traffic forecasting unit (103), 10 Switches the base station to the most efficient sleep or capacity mode based on the estimated load. dynamic energy mode assignment layer (104), Hardware control that transmits the agreed-upon energy modes to the physical network elements. application interface (105) It includes. 15 2. This system complies with System 1, and its feature is the energy efficiency and grid efficiency provided by the system. It includes an energy saving and quality control panel (106) that monitors its quality.

Images