Selection of random access preamble

Abstract

A first network node of a first radio access technology (RAT) may determine a first information of the first network node to determine a first random access channel (RACH) preamble selection policy for a user device. The first network node may receive from a second network node of a second RAT, second information of the second network node to determine the first RACH preamble selection policy. The first network node may determine the first RACH preamble selection policy based on the first information of the first network node and the second information of the second network node. The first network node may transmit to the user device the first RACH preamble selection policy.

Description

[0001] SELECTION OF RANDOM ACCESS PREAMBLE

[0002] TECHNICAL FIELD

[0003] This description relates to wireless communications.

[0004] BACKGROUND

[0005] A communication system may be a facility that enables communication between two or more nodes or devices, such as fixed or mobile communication devices. Signals can be carried on wired or wireless carriers.

[0006] An example of a cellular communication system is an architecture that is being standardized by the 3rd Generation Partnership Project (3GPP). A recent development in this field is often referred to as the long-term evolution (LTE) of the Universal Mobile Telecommunications System (UMTS) radio-access technology. EUTRA (evolved UMTS Terrestrial Radio Access) is the air interface of 3GPP's Long Term Evolution (LTE) upgrade path for mobile networks. In LTE, base stations or access points (APs), which are referred to as enhanced Node AP (eNBs), provide wireless access within a coverage area or cell. In LTE, mobile devices, or mobile stations are referred to as user equipments (UE). LTE has included a number of improvements or developments. Aspects of LTE are also continuing to improve.

[0007] 5G New Radio (NR) development is part of a continued mobile broadband evolution process to meet the requirements of 5G, similar to earlier evolution of 3G and 4G wireless networks. In addition, 5G is also targeted at the new emerging use cases in addition to mobile broadband. A goal of 5G is to provide significant improvement in wireless performance, which may include new levels of data rate, latency, reliability, and security. 5G NR may also scale to efficiently connect the massive Internet of Things (loT) and may offer new types of mission-critical services. For example, ultrareliable and low-latency communications (URLLC) devices may require high reliability and very low latency. 6G and other networks are also being developed. SUMMARY

[0008] According to some aspects, there is provided the subject matter of the independent claims. Some further aspects are defined in the dependent claims. The details of one or more examples of embodiments are set forth in the accompanying drawings and the description below. Other features will be apparent from the description and drawings, and from the claims.

[0009] BRIEF DESCRIPTION OF THE DRAWINGS

[0010] FIG. 1 is a block diagram of a wireless network 130.

[0011] FIG. 2 is a diagram illustrating types of RACH preambles.

[0012] FIG. 3 is a diagram illustrating a comparison of preamble allocation using static uniform distribution and based on a dynamic ML based RACH preamble selection policy.

[0013] FIG. 4 is a diagram illustrating signaling diagram of exchange of parameters of the RACH preamble selection policy between the network nodes and the UE in case of a single vendor MRSS cell.

[0014] FIG. 5 is a diagram illustrating signaling diagram of exchange of parameters of the RACH preamble selection policy between the network nodes and the UE in case of a multi vendor MRSS cell.

[0015] FIG. 6 is a diagram illustrating a general ML based approach to optimize the preamble allocation in MRSS cell.

[0016] FIG. 7 is a diagram illustrating an example of a supervised learning model (or ML model) with offline training to learn the probability distribution parameter for RACH preamble selection in a MRSS cell.

[0017] FIG. 8 is a diagram illustrating an example of a reinforcement learning based method to learn the preamble probability distribution for PRACH in MRSS cell.

[0018] FIG. 9 is a flow chart illustrating operation of an apparatus (e.g., which may be a UE or user device, or other apparatus) according to an example embodiment.

[0019] FIG. 10 is a flow chart illustrating operation of an apparatus (e.g., which may be a UE or user device, or other apparatus) according to an example embodiment.

[0020] FIG. 11 is a block diagram of a wireless station or node (e.g., UE, user device, AP, BS, eNB, gNB, RAN node, network node, TRP, or other node) 1300 according to an example embodiment. DETAILED DESCRIPTION

[0021] It shall be understood that although the terms “first,” “second,”... , etc., in front of noun(s) and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another and they do not limit the order of the noun(s). For example, a first element could be termed a second element, and similarly, a second element could be termed a first element, without departing from the scope of example embodiments. As used herein, the term “and / or” includes any and all combinations of one or more of the listed terms.

[0022] As used herein, unless stated explicitly, performing a step “in response to A” does not indicate that the step is performed immediately after “A” occurs and one or more intervening steps may be included.

[0023] FIG. 1 is a block diagram of a wireless network 130. In the wireless network 130 of FIG. 1 , user devices 131 , 132, 133 and 135, which may also be referred to as mobile stations (MSs) or user equipment (UEs), may be connected (and in communication) with a base station (BS) 134, which may also be referred to as an access point (AP), an enhanced Node B (eNB), a gNB or a network node. The terms user device and user equipment (UE) may be used interchangeably. A BS may also include or may be referred to as a RAN (radio access network) node, and may include a portion of a BS or a portion of a RAN node, such as e.g., such as a centralized unit (CU) and / or a distributed unit (DU) in the case of a split BS or split gNB. At least part of the functionalities of a BS (e.g., access point (AP), base station (BS) or (e)Node B (eNB), gNB, RAN node) may also be carried out by any node, server or host which may be operably coupled to a transceiver, such as a remote radio head. BS (or AP) 134 provides wireless coverage within a cell 136, including to user devices (or UEs) 131 , 132, 133 and 135. Although only four user devices (or UEs) are shown as being connected or attached to BS 134, any number of user devices may be provided. BS 134 is also connected to a core network 150 via a S1 interface 151 . This is merely one simple example of a wireless network, and others may be used.

[0024] A base station (e.g., such as BS 134) is an example of a radio access network (RAN) node within a wireless network. A BS (or a RAN node) may be or may include (or may alternatively be referred to as), e.g., an access point (AP), a gNB, an eNB, or portion thereof (such as a centralized unit (CU) and / or a distributed unit (DU) in the case of a split BS or split gNB), or other network node.

[0025] Some functionalities of the communication network may be carried out, at least partly, in a central / centralized unit, CU, (e.g., server, host or node) operationally coupled to distributed unit, DU, (e.g., a radio head / node). Thus, 5G networks architecture may be based on a so-called CU-DU split. The gNB-CU (central node) may control a plurality of spatially separated gNB-DUs, acting at least as transmit / receive (Tx / Rx) nodes. In some embodiments, however, the gNB-DUs (also called DU) may comprise e.g., a radio link control (RLC), medium access control (MAC) layer and a physical (PHY) layer, whereas the gNB-CU (also called a CU) may comprise the layers above RLC layer, such as a packet data convergence protocol (PDCP) layer, a radio resource control (RRC) and an internet protocol (IP) layers. Other functional splits are possible too.

[0026] According to an illustrative example, a BS node (e.g., BS, eNB, gNB, CU / DU, ... ) or a radio access network (RAN) may be part of a mobile telecommunication system. A RAN (radio access network) may include one or more BSs or RAN nodes that implement a radio access technology, e.g., to allow one or more UEs to have access to a network or core network (CN). Thus, for example, the RAN (RAN nodes, such as BSs or gNBs) may reside between one or more user devices or UEs and a core network. According to an example embodiment, each RAN node (e.g., BS, eNB, gNB, CU / DU, ... ) or BS may provide one or more wireless communication services for one or more UEs or user devices, e.g., to allow the UEs to have wireless access to a network, via the RAN node. Each RAN node or BS may perform or provide wireless communication services, e.g., such as allowing UEs or user devices to establish a wireless connection to the RAN node, and sending data to and / or receiving data from one or more of the UEs. For example, after establishing a connection to a UE, a RAN node or network node (e.g., BS, eNB, gNB, CU / DU, ... ) may forward data to the UE that is received from a network or the core network, and / or forward data received from the UE to the network or core network. RAN nodes or network nodes (e.g., BS, eNB, gNB, CU / DU, ... ) may perform a wide variety of other wireless functions or services, e.g., such as broadcasting control information (e.g., such as system information or on- demand system information) to UEs, paging UEs when there is data to be delivered to the UE, assisting in handover of a UE between cells, scheduling of resources for uplink data transmission from the UE(s) and downlink data transmission to UE(s), sending control information to configure one or more UEs, and the like. These are a few examples of one or more functions that a RAN node or BS may perform.

[0027] A user device or user node (user terminal, user equipment (UE), mobile terminal, handheld wireless device, etc.) may refer to a portable computing device that includes wireless mobile communication devices operating either with or without a subscriber identification module (SIM), including, but not limited to, the following types of devices: a mobile station (MS), a mobile phone, a cell phone, a smartphone, a personal digital assistant (PDA), a handset, a device using a wireless modem (alarm or measurement device, etc.), a laptop and / or touch screen computer, a tablet, a phablet, a game console, a notebook, a vehicle, a sensor, and a multimedia device, as examples, or any other wireless device. It should be appreciated that a user device may also be (or may include) a nearly exclusive uplink only device, of which an example is a camera or video camera loading images or video clips to a network. Also, a user node may include a user equipment (UE), a user device, a user terminal, a mobile terminal, a mobile station, a mobile node, a subscriber device, a subscriber node, a subscriber terminal, or other user node. For example, a user node may be used for wireless communications with one or more network nodes (e.g., gNB, eNB, BS, AP, CU, DU, CU / DU) and / or with one or more other user nodes, regardless of the technology or radio access technology (RAT). In LTE (as an illustrative example), core network 150 may be referred to as Evolved Packet Core (EPC), which may include a mobility management entity (MME) which may handle or assist with mobility / handover of user devices between BSs, one or more gateways that may forward data and control signals between the BSs and packet data networks or the Internet, and other control functions or blocks. Other types of wireless networks, such as 5G (which may be referred to as New Radio (NR)) may also include a core network.

[0028] In addition, the techniques described herein may be applied to various types of user devices or data service types, or may apply to user devices that may have multiple applications running thereon that may be of different data service types. New Radio (5G) development may support a number of different applications or a number of different data service types, such as for example: machine type communications (MTC), enhanced machine type communication (eMTC), Internet of Things (loT), and / or narrowband loT user devices, enhanced mobile broadband (eMBB), and ultrareliable and low-latency communications (URLLC). Many of these new 5G (NR) - related applications may require generally higher performance than previous wireless networks. loT may refer to an ever-growing group of objects that may have Internet or network connectivity, so that these objects may send information to and receive information from other network devices. For example, many sensor type applications or devices may monitor a physical condition or a status and may send a report to a server or other network device, e.g., when an event occurs. Machine Type Communications (MTC, or Machine to Machine communications) may, for example, be characterized by fully automatic data generation, exchange, processing and actuation among intelligent machines, with or without intervention of humans. Enhanced mobile broadband (eMBB) may support much higher data rates than currently available in LTE.

[0029] Ultra-reliable and low-latency communications (URLLC) is a new data service type, or new usage scenario, which may be supported for New Radio (5G) systems. This enables emerging new applications and services, such as industrial automations, autonomous driving, vehicular safety, e-health services, and so on. 3GPP targets in providing connectivity with reliability corresponding to block error rate (BLER) of 10-5 and up to 1 ms U-Plane (user / data plane) latency, by way of illustrative example. Thus, for example, URLLC user devices / UEs may require a significantly lower block error rate than other types of user devices / UEs as well as low latency (with or without requirement for simultaneous high reliability). Thus, for example, a URLLC UE (or URLLC application on a UE) may require much shorter latency, as compared to an eMBB UE (or an eMBB application running on a UE).

[0030] The techniques described herein may be applied to a wide variety of wireless technologies or wireless networks, such as 5G (New Radio (NR)), cmWave, and / or mmWave band networks, loT, MTC, eMTC, eMBB, URLLC, 6G, etc., or any other wireless network or wireless technology. These example networks, technologies or data service types are provided only as illustrative examples. A user device (or UE) may measure various signals and may transmit one or more measurement reports to the network. For example, a UE may measure reference signals received from one or more network nodes (e.g., gNBs or DUs), including channel state information-reference signals (CSI-RSs) and / or synchronization signal block (SSB) reference signals, demodulation references signals, and / or other reference signals. Based on received reference signals, the UE may measure various signal parameters, e.g., such as reference signal received power (RSRP), reference signal received quality (RSRQ), signal to interference plus noise ratio (SINR), received signal strength indicator (RSSI), or other signal parameter.

[0031] The PHY (physical) layer may refer to layer 1 (L1 ) and MAC (media access control) may refer to layer 2 (L2). RSRP, RSRQ, SINR and RSSI are signal quantities measured at layer 1 (L1). The UE may send L1 measurement reports (e.g., CSI-RS reports, which include measurements of one or more signal parameters for one or more cells) to a gNB, source DU or serving cell. These L1 measurement reports may be sent periodically, for example, or aperiodically. L1 / L2 measurement reports may include no averaging or filtering of measurement values or may include less averaging or filtering than what is performed for L3 measurement reports. L1 (or L1 / L2) measurement reports may be transmitted by a UE to a serving network node or source DU and may cause the network node to trigger or initiate a L1 / L2 triggered mobility (LTM) handover of the UE to another cell. L1 measurements (e.g., RSRP RSRQ, RSSI) may be provided or reported periodically to the DU (MAC / PHY).

[0032] A machine learning (ML) model may be used within a wireless network to perform (or assist with performing) one or more tasks. In general, one or more nodes (e.g., BS, gNB, eNB, RAN node, user node, UE, user device, relay node, or other wireless node) within a wireless network may use or employ a ML model, e.g., such as, for example a neural network model (e.g., which may be referred to as a neural network, an artificial intelligence (Al) neural network, an Al neural network model, an Al model, a machine learning (ML) model or algorithm, a model, or other term) to perform, or assist in performing, one or more ML-enabled tasks. Other types of models may also be used. A ML-enabled task may include tasks that may be performed (or assisted in performing) by a ML model, or a task for which a ML model has been trained to perform or assist in performing). ML-based algorithms or ML models may be used to perform and / or assist with performing a variety of wireless and / or radio resource management (RRM) and / or RAN-related functions or tasks to improve network performance, such as, e.g., in the UE for beam prediction (e.g., predicting a best beam or best beam pair based on measured reference signals), antenna panel or beam control, RRM (radio resource measurement) measurements and feedback (channel state information (CSI) feedback), link monitoring, Transmit Power Control (TPC), etc. In some cases, ML models may be used to improve performance of a wireless network in one or more aspects or as measured by one or more performance indicators or performance criteria.

[0033] Models (e.g., neural networks or ML models) may be or may include, for example, computational models used in machine learning made up of nodes organized in layers. The nodes are also referred to as artificial neurons, or simply neurons, and perform a function on provided input to produce some output value. A neural network or ML model may typically require a training period to learn the parameters, i.e., weights, used to map the input to a desired output. The mapping may occur via the function that is learned from a given data for the problem in question. Thus, the weights are weights for the mapping function of the neural network. Each neural network model or ML model may be trained for a particular task.

[0034] To provide the output given the input, the ML functionality of a neural network model or ML model should be trained, which may involve learning the proper value for a large number of parameters (e.g., weights and / or biases) for the mapping function (or of the ML functionality of the ML model). For example, the parameters may be used to weight and / or adjust terms in the mapping function. This training may be an iterative process, with the values of the weights and / or biases being tweaked over many (e.g., tens, hundreds and / or thousands) of rounds of training episodes or training iterations until arriving at the optimal, or most accurate, values (or weights and / or biases). In the context of neural networks (neural network models) or ML models, the parameters may be initialized, often with random values, and a training optimizer iteratively updates the parameters (e.g., weights) of the neural network to minimize error in the mapping function. In other words, during each round, or step, of iterative training the network updates the values of the parameters so that the values of the parameters eventually converge to the optimal values.

[0035] ML models may be trained in either a supervised or unsupervised manner, as examples. In supervised learning, training examples are provided to the ML model or other machine learning algorithm. A training example includes the inputs and a desired or previously observed output. Training examples are also referred to as labeled data because the input is labeled with the desired or observed output. In the case of a neural network (which may be a specific case of ML model), the network (or ML model) learns the values for the weights used in the mapping function or ML functionality of the ML model that most often result in the desired output when given the training inputs. In unsupervised training, the ML model learns to identify a structure or pattern in the provided input. In other words, the model identifies implicit relationships in the data. Unsupervised learning is used in many machine learning problems and typically requires a large set of unlabeled data.

[0036] According to an example embodiment, a ML model may be classified into (or may include) two broad categories (supervised and unsupervised), depending on whether there is a learning “signal” or “feedback” available to a model. Thus, for example, within the field of machine learning, there may be two main types of learning or training of a model: supervised, and unsupervised. The main difference between the two types is that supervised learning is done using known or prior knowledge of what the output values for certain samples of data should be. Therefore, a goal of supervised learning may be to learn a function that, given a sample of data and desired outputs, best approximates the relationship between input and output observable in the data. Unsupervised learning, on the other hand, does not have labeled outputs, so its goal is to infer the natural structure present within a set of data points.

[0037] Supervised learning: The computer is presented with example inputs and their desired outputs, and the goal may be to learn a general rule that maps inputs to outputs. Supervised learning may, for example, be performed in the context of classification, where a computer or learning algorithm attempts to map input to output labels, or regression, where the computer or algorithm may map input(s) to a continuous output(s). Common algorithms in supervised learning may include, e.g., logistic regression, naive Bayes, support vector machines, artificial neural networks, and random forests. In both regression and classification, a goal may include finding specific relationships or structure in the input data that allow us to effectively produce correct output data. In some example cases, the input signal may be only partially available, or restricted to special feedback. Semi-supervised learning: the computer may be given only an incomplete training signal; a training set with some (often many) of the target outputs missing. Active learning: the computer can only obtain training labels for a limited set of instances (based on a budget), and also may optimize its choice of objects for which to acquire labels. When used interactively, these can be presented to the user for labeling.

[0038] Unsupervised learning: No labels are given to the learning algorithm, leaving it on its own to find structure in its input. Some example tasks within unsupervised learning may include clustering, representation learning, and density estimation. In these cases, the computer or learning algorithm is attempting to learn the inherent structure of the data without using explicitly-provided labels. Some common algorithms include k-means clustering, principal component analysis, and auto-encoders. Since no labels are provided, there may be no specific way to compare model performance in most unsupervised learning methods.

[0039] Continual Learning (CL) may refer to or may include a capability of the ML model to adapt to ever-changing (or continuously changing, or periodically changing) surrounding environment or data by learning or adapting the ML model continually based on incoming data (or new or updated data), e.g., without forgetting original or previous knowledge or ML model settings, and, e.g., which may be based on less than a full or complete set of data. For example, given a (e.g., potentially unlimited or continuous) stream of data (e.g., data reflecting changing or updated conditions or environment upon which the ML model should be updated), a continual learning (CL) algorithm may (or should) learn, e.g., by updating or adapting weights or other parameters of the ML model, based on a sequence of partial experiences or partial data (e.g., a most recent set of data) where all data may not be available at once, since new or updated data will be received later (thus, the new data potentially renders the weights or parameter settings of the ML model obsolete or inaccurate). Thus, a full or complete set of data may not be considered available at that time of ML model updating or adaptation, since the data or environment may be continuously or continually changing over time. Thus, at any given point or moment in time, data (upon which the ML model may be updated or adapted) may be considered incomplete because there may be a continuous stream of data. Thus, a CL algorithm may include or may refer to iteratively updating or adapting weights or other parameters of the ML model based on an updated set of data, and then repeating the learning or adaptation process for the ML model when a second (or later) set of updated data is received subsequently.

[0040] Reinforcement learning (RL) may refer to, may be or may include an interdisciplinary area of machine learning and optimal control concerned with how an intelligent agent should perform actions in a dynamic environment in order to maximize a reward. RL may be a goal based optimization approach where an agent performs an action (e.g., an action performed by a ML model) based on the observed state / context (or inputs) and then receives a reward to learn the optimal policy or train the ML model. To distinguish the good (or preferred) actions from the bad (or non-preferred) actions, the agent may explore the action space (e.g., performing various actions) by performing various actions and observing or obtaining a reward (feedback that may be used to train the ML model). Due to its ability to optimize radio functions, e.g., such as various radio resource management (RRM) functions, based on a reward, RL may be used in future wireless networks. There are many radio functions for which a ML model and / or RL may be used to assist and / or improve performance of the radio function, e.g., such as beam selection (or beam management), power control, assisting in performing handovers, and many others.

[0041] In an example embodiment, a random access channel (RACH) preamble may be transmitted by a UE to a network node (such as a gNB, a gNB-DU, and / or the like) over a physical random access channel (PRACH) to obtain an uplink (UL) synchronization and / or as part of a random access procedure to establish a connection with the network node. There may be 64 RACH preambles defined in each time-frequency PRACH occasion (or RACH occasion). The RACH preamble may include two parts, e.g., a cyclic prefix (CP) and a RACH preamble sequence. FIG. 2 is a diagram illustrating types of RACH preambles. In an example, there are 13 types of RACH preamble formats supported known as Format 0, Format 1 , Format 2, Format 3, Format A1 , Format A2, Format A3, Format B1 , Format B2, Format B3, Format B4, Format CO, Format C1 . The 13 types of RACH preamble formats may be grouped into two categories: long RACH preamble, and short RACH preamble. Differences in the time domain of different RACH preamble formats may include different cyclic prefix (CP) length, sequence length, gap (GP) length and number of repetitions.

[0042] The UE may choose a RACH preamble from a pool of available RACH preambles for accessing a network node e.g., a BS-DU, a gNB-DU, and / or the like. In an example, there may be contention-based RACH preambles and contention free RACH preambles. In the case of contention free RACH preambles, the UE receives configurations (for example RRC configuration messages) from the network node while in RRC connected state, wherein the configurations may indicate which RACH preamble sequence to be used by the UE. In case of the contention-based RACH preambles, the UE selects a RACH preamble at random and may send the random access request message with the selected RACH preamble to the network node. In case that another UE sends the same RACH preamble (sequence) to the network, the network may respond with random access response (RAR) to the UEs. The RAR may include timing advance (TA), temporary C-RNTI (T-C-RNTI), UL grant for msg 3 (e.g., for L2 / L3 message), and / or the like. Therefore, the two UEs may use the elements of the RAR message to send L2 / L3 messages over the same resources e.g., because the two UEs received the same resource allocation (meaning with the same time / frequency location) in the RAR message. In an implementation, as the signals interfere with each other, the network node may not be able to detect and decode any of the two signals. As a result, both UEs may restart the random access procedure.

[0043] In another implementation, the network node may detect and decode one of the signals and respond to one of the UEs with msg 4 and does not respond to the other UE. In this case, the UE that did not receive the msg 4, may repeat the random access procedure. In another example, a RACH preamble contention (or collision) may occur when the same RACH preamble arrives at the network node from two (or multiple) UEs within a RACH occasion. In an implementation, the network node may collect the statics of RACH preamble contentions or collisions that occur due to arrival of the same RACH preamble.

[0044] In an example implementation, the RACH preambles may include unique signatures to be distinguishable from each other, and therefore the probability of collision or interference among the RACH preambles may be reduced.

[0045] In an example, different radio access technologies (RATs) may share spectrum resources. A RAT, or radio access technology, may include or may refer to the technology and / or protocols used in wireless communication systems to establish a connection between a user device or UE and a network or network node. Different RATs may use different technology and / or different protocols for communication. There are several RATs used in different generations of cellular networks, such as, for example, 2G (GSM), 3G (UMTS), 4G (LTE), and 5G (NR), and now 6G. In an example, multi-RAT spectrum sharing (MRSS) may be supported by a communication system. The MRSS may be a feature for the migration to 6G as it allows new radio (NR) and 6G cells to share the same carrier(s) dynamically and to adapt to different traffic requirements. In a MRSS cell, the same radio resources may be shared by NR and 6G. Even though in the following the examples are mainly described with a viewpoint where RACH preambles are shared between two RATs, such as 5G and 6G, there may be more than two RATs in the scenario and the UEs may share the same set of preambles between many (more than two) RATs. The example embodiments are applicable to such scenarios as well.

[0046] Existing technologies support a limited number of RACH preambles per RACH occasion. For example, there may be 64 RACH preambles defined in each RACH occasion. Although the RACH preambles are designed to be unique and distinguishable from each other (e.g., to allow the UEs to access the network node without interference), a problem may arise in the case of the MRSS when sharing the spectrum between 5G RATs and 6G RATs (e.g., spectrum or resources may be shared between a 5G / NR network node or gNB and a 6G network node or gNB). In an example, the problem may arise when the UEs of both 5G RAT and 6G RAT may share the same pool of RACH preambles. Therefore, in a contention-based random access procedure (where contention-based RACH preambles are used), frequent RACH preamble collisions may occur. In other words, for the case of the multi RAT spectrum sharing (MRSS), e.g., when a first network node of a 5G RAT and a second network node of a 6G RAT share the same spectrum, allocation of RACH preambles in a static manner equally or randomly between the two RATs (the 5G RAT and the 6G RAT) may not be scalable. For example, a number of UEs of the 5G RAT that may send random access requests (or RACH requests, where each RACH request includes a RACH preamble) may be significantly larger while the number of the UEs of the 6G RAT may be significantly lower than the number of UEs in the 5G RAT. Therefore, if the pool of the RACH preambles is divided in a static manner equally between the UEs of the 5G RAT and UEs of the 6G RAT, then UEs of the 5G RAT may experience collisions while at least some of the RACH preamble resources of the 6G RAT may not be utilized, which is an inefficient use of RACH preamble resources among these two RATs. Thus, a more efficient RACH preamble resource allocation technique may be desirable to enhance the performance of the system and reduce the number of RACH preamble collisions.

[0047] Example embodiments may enhance the performance of the system by determining a dynamic RACH preamble selection policy for a UE and providing the RACH preamble selection policy to the UE. For example, a first network node of a first radio access technology (RAT) (such as a gNB-DU of the 5G RAT), may determine first information of the first network node to determine a first RACH preamble selection policy for a UE. In an example, the first network node may receive from a second network node of a second RAT (e.g., a gNB-DU of the 6G RAT), second information of the second network node to determine the first RACH preamble selection policy. In an example, the first network node may determine the first RACH preamble selection policy based on the first information of the first network node and the second information of the second network node. In an example, the first network node may transmit to the UE the first RACH preamble selection policy.

[0048] In addition, example embodiments enhance RACH preamble selection of the UE to reduce a probability of a RACH preamble collision. Example embodiments enable the UE to receive from the first network node of the first radio access technology (RAT), the first random access channel (RACH) preamble selection policy. For example, the first RACH preamble selection policy may be based on the first information of the first network node and the second information of the second network node of the second RAT. In an example, the UE may select a RACH preamble based on the first RACH preamble selection policy. The UE may then transmit a random access request to the first network node, using the selected RACH preamble.

[0049] Therefore, example embodiments are directed to determining a RACH preamble selection policy (e.g., the first RACH preamble selection policy and / or the second RACH preamble selection policy) for the UEs to reduce the probability of RACH preamble collision during a contention-based random access procedure. The network may employ the first information and the second information to perform a determination or a prediction based on an algorithm that uses machine learning (ML) techniques (or based on a ML model) for determining the RACH preamble selection policy (e.g., the first RACH preamble selection policy and / or the second RACH preamble selection policy). The second RACH preamble selection policy may be determined by the second network node, for example.

[0050] In an example embodiment, the first information may be (or may include) the first information of the first network node of the first RAT. In an example embodiment, the second information may be (or may include) the second information of the second network node of the second RAT. In an example, the first network node may be a first gNB distributed unit (gNB-DU). In an example, the second network node may be a second gNB-DU or a gNB centralized unit (gNB-CU).

[0051] In an example embodiment, the first RACH preamble selection policy may include at least one of: a probability distribution parameter to select a RACH preamble among a predetermined set of candidate RACH preambles (or a predetermined pool of candidate RACH preambles, the set of 64 RACH preambles, and / or the like). In an example, the probability distribution parameter may include at least one of a type of a distribution function (e.g., normal distribution, Gaussian distribution, exponential distribution, and / or the like), and / or a parameter associated with the distribution function such as mean, variance, and / or the like. In an example, the parameter associated with the distribution function may include at least one of a mean value, or a variance value, and / or the like. In an example, the first RACH preamble selection policy may include allocation information of RACH preambles for access of the user device via the first network node and the second network node. For example, the allocation may be based on the probability distribution parameter. The allocation information may define e.g. that a certain percentage of the available preambles are reserved for 5G UEs and the rest are reserved for 6G UEs. As another example the allocation information may define that a set of preambles (10% for example) is reserved for a certain category of UEs, for URLLC traffic as an example.

[0052] In an example embodiment, the first information of the first network node and the second information of the second network node may be used as inputs to the first ML model or ML algorithm to determine an output that may include the first RACH preamble selection policy.

[0053] In an example, the first information of the first network node may include at least one of a mode or state of operation of the first network node (e.g., power saving state), arrival rate of random access requests (from UEs) for the first network node, a number of RACH preamble collisions at the first network node, load information of the first network node, and / or the like. In an example, RACH preamble collisions may be among the UEs of the first RAT, among the UEs of the second RAT or among the UEs of the first RAT and second RAT. In an example, events at the first network node of the first RAT may be visible (detected) by the second network node of the second RAT. Therefore, when a RACH preamble is transmitted by a first UE to the first network node, and the same RACH preamble is transmitted by a second UE to the second network node, either (or both) of the first network node and / or the second network node may detect a RACH preamble collision.

[0054] In an example embodiment, the first network node may receive the second information of the second network node and use that to determine the first RACH preamble selection policy.

[0055] In an example, the second information of the second network node may include at least one of a mode or state of operation of the second network node (e.g., power saving state), arrival rate of random access requests (from UEs, wherein each request includes a RACH preamble) for the second network node, a number of RACH preamble collisions at the second network node, load information of the second network node. For example, the first network node may employ a first ML model to determine the first RACH preamble selection policy based on the first information and the second information. For example, the first ML model or the first ML algorithm, may be designed to determine the first RACH preamble selection policy wherein the allocation of the RACH preambles or the probability distribution parameters have minimal (or reduced) overlap with an allocation of RACH preambles for UEs of the second network node. For example, by taking the second information into account, the first ML model may produce a probability distribution parameter (for UEs of the first network node), wherein the mean, e g., p parameter is sufficiently different from the mean e.g., p parameter for UEs of the second network node so that the likelihood of overlap or potential RACH preamble collisions is reduced. In another example, an appropriate choice of the variance parameter o2may also impact the number of RACH preamble collisions. In an alternative example, load information of the second network node may assist in determining by the first network node whether a mean value and a variance can be selected that spans a larger portion of the pool of the RACH preambles. For example, when the load of the second network node is very low (as indicated by any of the information elements of the second information, such as arrival rate, number of collisions, load), the first network node may be free to allocate a larger portion of the pool of the RACH preambles (for the UEs to select from).

[0056] In another example, the second information of the second network node may include at least one of a second RACH preamble selection policy, and / or a probability distribution parameter associated with the second RACH preamble selection policy. In other words, the second network node may provide the second RACH preamble selection policy that is locally determined by the second network node, e.g., based on a second ML model and the second information available at the second network node. When the first network node receives the second RACH preamble selection policy (e.g., including a probability distribution parameter associated with the second RACH preamble selection policy), the first ML model or the first ML algorithm at the first network node may take into account the parameters associated with the probability distribution parameter (associated with the second RACH preamble selection policy) to determine the first RACH preamble selection policy such that the mean and / or variance parameters yield the first RACH preamble selection policy with minimum overlap with the second RACH preamble selection policy, thereby reducing the probability of RACH preamble collisions.

[0057] In an example embodiment, the first machine learning (ML) model of the first network node may be determined based on training an algorithm based on the first information of the first network node. For example, a supervised learning method may be employed to determine the first RACH preamble selection policy in a controlled network environment based on various data points of the first information (and / or the second information) at different times, and / or network conditions.

[0058] In another example, a second ML model of the second network node may be determined based on training an algorithm based on the second information available at the second network node. For example, a supervised learning method may be employed to determine the second RACH preamble selection policy in a controlled network environment based on various data points of the second information (and / or the first information) at different times, and / or network conditions.

[0059] In an example embodiment, the first network node may determine a probability distribution parameter based on the first information of the first network node and the second information of the second network node. In an example, the first RACH preamble selection policy may be based on the probability distribution parameter. In an example, the network node may transmit the probability distribution parameter to the UE as (part of) the first RACH preamble selection policy.

[0060] In an example, the first information (of the first network node) and / or the second information (of the second network node) may be used to predict a probability distribution parameter for selection of RACH preambles by UEs of different radio access technologies (RATs) that share the same spectrum. The prediction information (e.g. the predicted probability distribution parameter may be used to determine or create a random access channel (RACH) preamble selection policy that determines or indicates which RACH preambles may be selected (with a higher probability) by the UEs of the first radio access technology (RAT) and which RACH preambles may be selected (with a higher probability) by the UEs of the second RAT. In an example, the RACH preambles for the UEs of the first RAT and the UEs of the second RAT may include a same pool or set of RACH preambles, wherein a RACH preamble may be selected by a UE of the first RAT with a different probability (determined by the probability distribution parameter). For example, the UE of the first RAT (e.g., a 5G UE) may be more likely to select a certain preamble than the UE of the second RAT (e.g., a 6G UE). The probability may be determined based on the probability distribution parameter indicated by the RACH preamble selection policy (e.g., the first RACH preamble selection policy and / or the second RACH preamble selection policy). Therefore, the RACH preamble selection policy (e.g., the first RACH preamble selection policy and / or the second RACH preamble selection policy) may determine which RACH preambles are more likely to be selected (by a UE) and based on which probability. As an example, the probability distribution parameter may be a probability distribution function that may assign a probability for selection of each random variable. In an example, each random variable may be associated with an element of a pool of RACH preambles. Thus, according to the probability distribution function, different RACH preambles may be selected with a different probability based on a type of the probability distribution and parameters such as mean, variance, (p,o2) and / or the like.

[0061] The RACH preamble selection policy (e.g., the first RACH preamble selection policy and / or the second RACH preamble selection policy) may include a probability distribution parameter or a probability distribution with one or more parameters depending on the distribution. The output of the ML model (e.g., the first ML model and / or the second ML model) may describe parameters related to the probability distribution of the RACH preamble selection. When a supervised learning method is used, an initial iteration of the ML model (or algorithm) output may be based on input parameters such as collision statistics, load of the network, arrival rate of the random access requests (from the UEs), and / or the like. Thus, the output of the ML model may vary based on the collision statistics, load of the network, arrival rate of the random access requests, and / or the like.

[0062] In an example, different input parameters of the ML model (e.g., the first ML model and / or the second ML model) may be given different weights or bias when the output of the ML model is calculated. The weights and bias values may be updated over time to achieve an optimal result, e.g. based on monitoring statistics of the collisions when the RACH preamble selection policy / ies are used by the UEs. The weights and bias values may impact the performance of the ML model or algorithm (e.g., the first ML model and / or the second ML model).

[0063] In an example implementation, when the result from the output of the ML model is transmitted to UEs as the RACH preamble selection policy, the UEs may select the RACH preambles based on the RACH preamble selection policy (e.g., the probability distribution parameter) and send random access requests including the selected RACH preamble. The network may monitor the statistics of collisions and then attribute or associate the statistics of collision to the performance of the implemented RACH preamble selection policy that was provided to the UEs. If the ML model (e.g., the first ML model and / or the second ML model) does not perform as expected (e.g., the number of RACH preamble collisions exceed a threshold), the ML model may be modified by altering the weights and bias values. Based on the monitoring, if a number of collisions at the first RAT is higher than a threshold while the load of the second RAT is not high, the RACH preamble selection policy may be adjusted by the ML model to assign more RACH preambles with high selection probability to the UEs of the first RAT.

[0064] In an example embodiment, the determination of a RACH preamble selection policy (e.g., the first RACH preamble selection policy or the second RACH preamble selection policy) may be based on an algorithm that utilizes an ML model or ML algorithm (e.g., the first ML model and / or the second ML model). The ML model (e.g., the first ML model and / or the second ML model) may learn (or may be trained) by evaluating the output with respect to the actual result in terms of the system performance such as a number of RACH preamble collisions. The system performance may be expressed in terms of a key performance indicator (KPI). In other words, the ML model may adapt to the dynamics of the system (such as rate of random access request arrivals (e.g., arrival rate of random access requests, where each random access (or RACH) request includes a RACH preamble), network load, collision statistics, and / or the like) and adjust the RACH preamble selection policy (that may include information of the probability distribution function) accordingly. For example, inputs of the ML model (e.g., the first ML model and / or the second ML model) may include collision statistics (e.g., number of RACH preamble collisions per unit of time, load statistics, and / or the like). For example, the arrival rate of the random access requests may be determined based on a Poisson distribution. Under the Poisson distribution with the expectation of ASG events in a given interval, the probability of k events (e.g., arrival of random access request of 5G UEs) in the same interval may be

[0065] As an example, a network node may receive, randomly, an average of ASG = k RACH requests per minute. If the RACH requests (which include a RACH preamble) are independent, and receiving one RACH request does not change the probability of when the next RACH request will arrive. Under these assumptions, the number k of RACH requests received during any minute has a Poisson probability distribution.

[0066] An example of a ML model (e.g., the first ML model and / or the second ML model) may be as follows. In an example implementation, the ML model may include input parameters and an output parameter such as the probability distribution for selection of RACH preambles. In an example, the input parameters may include RACH preamble collision statistics over a period of time. The RACH preamble collision statistics may indicate a number of the RACH preamble collisions between time T1 and time T2. The output of the ML model may determine the probability distribution for selection of RACH preambles in a way to keep the number of the RACH preamble collisions below a threshold. As an example, the probability distribution for selection of RACH preambles may allow the UE to select more RACH preambles with higher probability. In other words, the number of the RACH preambles that can be selected more frequently may be increased. In another example, the input parameters may include an arrival rate of random access requests. The arrival rate of random access requests may be predicted based on the Poisson distribution. In other words, the input parameter of the ML model (e.g., the first ML model and / or the second ML model) may indicate that the probability of receiving k number of random access requests may be equal to P1. Based on the probability P1 , the probability distribution for selection of RACH preambles may be determined in a way that if at the worst-case scenario of arrival rate of k random access requests, the number of collisions do not exceed a threshold. In another example, the input parameter may include a load information of the first network node of the first RAT and / or the second network node of the second RAT. The load information may be determined by available and / or utilized resources such as processing resources, memory resources, network resources, and / or the like. For example, the load information of the second network node of the second RAT may impact the output of the ML model (e.g., the first ML model) at the first network node of the first RAT. As an example, the load information may be transmitted from the second network node to first network node based on Xn messages over the Xn interface between the first network node and the second network node. For example, when the collision statistics at the first network node indicate that the number of collisions has exceeded a threshold or the probability of higher arrival rate of random access requests may increase, the UE may be allowed to select more RACH preamble with a high probability from the pool of RACH preambles for accessing the first network node if the load of the second network node allows allocation of larger pool of RACH preambles to the first network node.

[0067] In an example embodiment, the probability distribution for selection of RACH preambles may follow a normal distribution N(p,o2) where p is the mean and o2 is the variance. The normal distribution may be a distribution function for independent, randomly generated variables. In an example, the variable may be mapped to an event of selecting a RACH preamble. For example, the mean may indicate that an event of selecting the nthRACH preamble may be more likely. For example, the corresponding event to the mean may indicate that the nthRACH preamble may be selected with a high likelihood. The variance may indicate variation of likelihood for selecting adjacent events or variables.

[0068] An example implementation of a ML algorithm (e.g., the first ML model and / or the second ML model) may include determining the output based on one or more inputs such as input 1 =number of collisions, input 2=rate of random access request arrivals (e.g., arrival rate of random access requests), input 3=load information of the first network node, input 4= load information of the second network node, and / or the like. The algorithm may determine an output based on a combination of the inputs e.g., a linear combination of the inputs. For example, the output=F(input 1 , input 4) = c1*F1 (input 1 ) + c2*F2(input 2) + c3*F3(input 3)+ c4*F4(input 4), where c1 , ... c4 are coefficients determining a weight or bias of an input for calculation of the output and F, F1 , F2, F3, and F4 are functions that map the inputs to a value for determination of the probability distribution parameter.

[0069] In another example, the algorithm may determine the output based on an optimization of a utility function with a constraint. For example, the optimization problem may be solved to find optimal mean and variance values with an objective to minimize the number of RACH preamble collisions and subject to constraints of load of the first network node or the load of the second network node.

[0070] In an example, the output of the ML model (e.g., the first ML model and / or the second ML model) may be the probability distribution and associated parameters. The probability distribution may be a function indicating a probability of selecting a RACH preamble. Therefore, the probability distribution is a mathematical function that may result in the probabilities of occurrence of different possible outcomes for an event. The event may be selecting a certain RACH preamble from a pool of available (or a predetermined set of) RACH preambles.

[0071] To train the ML model (e.g., the first ML model and / or the second ML model) or the ML algorithm, a discrepancy value between a number of collisions expected based on a probability distribution that yields a desired (or optimal) outcome and a number of collisions as a result of the produced (or provided) probability distribution may be measured. For example, the desired outcome may be in terms of a number of collisions. The discrepancy may be measured based on mean squared error (MSE), negative log likelihood, and / or the like. In an example, the MSE may be an average squared difference between the estimated values related to an expected value of collision statistics and an actual value of the collision statistics. In an example, the negative log likelihood may be a cost function that is used as a loss for ML models, indicating how bad the ML model is performing. Therefore, the lower values indicate a better result. In the MRSS, the RACH occasions may be dynamically allocated based on the demand and availability of spectrum resources. The allocation may be managed by a network node such as a gNB, a gNB-Dll, a gNB-Cll, and / or the like. The RACH preambles used in each RACH occasion may be shared between a first network node of a first RAT (e.g., 5G RAT) and a second network node of a second RAT (e.g., 6G RAT). In an example, the first network node or the second network node may determine the number and timing of RACH occasions for each technology based on factors such as network load, interference conditions, and quality of service requirements of the UE.

[0072] FIG. 3 is a diagram illustrating a comparison of preamble allocation using static uniform distribution and preamble allocation using a dynamic ML based RACH preamble selection policy. In other words, a (dynamic ML based) RACH preamble selection policy 320, 330 may provide information on selection of a RACH preamble based on a probability determined by a probability distribution parameter. The probability of selecting the RACH preamble may be determined by a probability distribution function that is provided by the output of the ML model (e.g., the first ML model and / or the second ML model). As an example, the probability distribution function may assign a certain probability for selecting a RACH preamble. An advantage of an example embodiment is evident when compared to a method where the selection of the RACH preamble is static 310 and based on a uniform distribution e.g., random or with equal probability. Then the static 310 approach may not be scalable if the number of users of the first RAT and the second RAT are significantly different. However, according to an example embodiment, based on the network conditions and a number of users, the RACH preambles may be assigned accordingly such that if the first network receives larger number of users (e.g., high load RAT 1 and low load RAT 2 at 330), then more RACH preamble resources may be available for accessing the first network. Alternatively, according to an example embodiment, based on the network conditions and a number of users, the RACH preambles may be assigned accordingly such that if the second network receives larger number of users (e.g., low load RAT 1 and high load RAT 2 at 320), then more RACH preamble resources may be available for accessing the second network. According to example embodiments, RACH preamble selection of the UE may be based on a probability distribution parameter provided by the first network node (or the network). The technique of the example embodiments may be utilized to allocate the RACH preambles in a way to provide a larger pool of RACH preambles for a cell (or network node) of a RAT that is likely to receive a larger number of random access request arrivals (e.g., arrival rate of random access requests). Furthermore, with a larger pool of RACH preambles, a likelihood of RACH preamble collision may be reduced.

[0073] FIG. 4 is a diagram illustrating signaling diagram of exchange of parameters of the RACH preamble selection policy between the network nodes and the UE in case of a single vendor MRSS cell. At step 1 , the UE 410 may select a RACH preamble to start a RACH procedure. In an example, the UE may select the RACH preamble based on a default RACH preamble selection policy. At step 2, the UE may use the selected RACH preamble to transmit data and / or RACH request / message to the first network node (gNB-DU) 420.

[0074] At step 3 of FIG. 4, the first network node may detect collisions. In an example, the first network node 420 of a first radio access technology (RAT) (such as a gNB-DU of a 5G RAT), may determine first information of the first network node 420 to determine a first random access channel (RACH) preamble selection policy for the UE 410. In an example, the first network node may use a set of allocated preambles (by the default RACH preamble selection policy) and the measured preambles collision between the assigned UEs as an input to the first ML model or the first ML algorithm. In an example, the input may include the first information of the first network node (gNB-DU) 420d. In an example, the detection of the RACH preamble collision between two UEs may be performed as follows. In an example, when a UE 410 selects a RACH preamble at random and sends the RACH message with the selected RACH preamble to the first network node 420, another UE may also send the same RACH preamble (sequence) to the first network node 420, the first network node 420 may respond with a random access response (RAR) to both UEs. The RAR may include timing advance (TA), temporary C-RNTI (T-C-RNTI), UL grant for msg 3 (e.g., for L2 / L3 message), and / or the like. Therefore, the two UEs may send the L2 / L3 messages to the first network node 420 over the same resources e.g., because the two UEs received the same resource allocation (meaning with the same time / frequency location) in the RAR message. In an implementation, as the signals interfere with each other, the first network node 420 may not be able to detect and decode any of the two signals. As a result, both UEs may restart the random access procedure. In another implementation, the first network node 420 may detect and decode one of the signals and respond to one of the UEs with msg 4 and does not respond to the other UE. In this case, the UE that did not receive msg 4, may repeat the random access procedure. In another example, a contention may occur when the same RACH preamble arrives at the first network node 420 from two (or multiple) UEs. In an implementation, the first network node 420 may collect the statics of contentions.

[0075] At step 4 of FIG. 4, the first network node 420 may train the first ML model. To train the first ML model, an appropriate loss function that measures the discrepancy between the predicted probability distribution parameters and the true or actual parameters of the probability distribution may be selected such as the mean squared error (MSE) or negative log-likelihood.

[0076] In an example embodiment, the first machine learning (ML) model of the first network node may be determined based on training an algorithm based on the first information of the first network node. For example, a supervised learning method may be employed to determine the first RACH preamble selection policy in a controlled network environment based on various data points of the first information (and / or the second information) at different times, and / or network conditions.

[0077] In another example, a second ML model of the second network node may be determined based on training an algorithm based on the second information available at the second network node. For example, a supervised learning method may be employed to determine the second RACH preamble selection policy in a controlled network environment based on various data points of the second information (and / or the first information) at different times, and / or network conditions.

[0078] To train the ML model (e.g., the first ML model and / or the second ML model) or the ML algorithm, a discrepancy value between a number of collisions expected based on a probability distribution that yields a desired (or optimal) outcome and a number of collisions as a result of the produced (or provided) probability distribution may be measured. For example, the desired outcome may be in terms of a number of collisions. The discrepancy may be measured based on mean squared error (MSE), negative log likelihood, and / or the like. In an example, the MSE may be an average squared difference between the estimated values related to an expected value of collision statistics and an actual value of the collision statistics. In an example, the negative log likelihood may be a cost function that is used as a loss for ML models, indicating how bad the ML model is performing. Therefore, the lower values indicate a better result.

[0079] At step 5 of FIG. 4, the first network node 420 (e.g., the 5G network node) may transmit the learnt probability distribution parameter as the first RACH preamble selection policy to the second network node (e.g., the 6G network node) 430. In an example embodiment, at step 6, the first network node 420 may receive from the second network node 430 a second RACH preamble selection policy. In an example, the first network node 420 may receive from a second network node 430 of a second RAT (e.g., a gNB-DU of a 6G RAT), the second RACH preamble selection policy or the probability distribution parameter of the second network node 430 to determine the first RACH preamble selection policy.

[0080] At step 7 of FIG. 4, the first network node 420 may determine or update the first RACH preamble selection policy based on the first information of the first network node 420 and the probability distribution parameter of the second network node 430. In an example, after receiving the second RACH preamble selection policy, the first network node 420 may adjust the probability distribution parameter(s) of the first RACH preamble selection policy based on the received probability distribution parameters (or the second RACH preamble selection policy) from second network node 430. In other words, either or both RATs e.g., the first network node 420 or the second network node 430 may adjust its own probability distribution parameters. In an example, if the network load information of the second network node 430 (e.g., 6G RAT) indicate that the 6G network is overloaded while the first network node 420 (5G RAT) is not overloaded, then the probability distribution parameter of the RACH preamble selection policy or allocation of the first network node (5G RAT) 420 may be adjusted based on the received load information of the second network node 430. In another example, if both of the first network node 420 and the second network node 430 are not overloaded, no adjustment to the probability distribution parameter(s) may be made because the collision statistics may remain below a threshold. At step 8, the first network node 420 may transmit the first RACH preamble selection policy to the UE 410 e.g., via a system information block (SIB) message, radio resource control (RRC) message, medium access control (MAC) control element (MAC-CE) and / or the like.

[0081] FIG. 5 is a diagram illustrating signaling diagram of exchange of parameters of the RACH preamble selection policy between the network nodes and the UE in case of a multi-vendor MRSS cell. Steps 1 to 4 and 7 to 8 of FIG. 5 are similar to the steps 1 to 4 and 7 to 8 of FIG. 4. At step 5, the first network node 440 may transmit the first information of the first network node 440 to the second network node 460. At step 6, the first network node 440 may receive the second information of the second network node 460 from the second network node 460. In an example, the first information of the first network node 440 may include at least one of a mode or state of operation of the first network node 440 (e.g., power saving state), arrival rate of random access requests for the first network node 440, a number of RACH preamble collisions at the first network node 440, load information of the first network node 440, and / or the like. In an example, the second information of the second network node 460 may include at least one of a mode or state of operation of the second network node 460(e.g., power saving state), arrival rate of random access requests for the second network node 460, a number of RACH preamble collisions at the second network node 460, load information of the second network node 460. In an example, the first information and / or the second information may include the inference model at the respective RAT. FIG. 6 is a diagram illustrating a general ML based approach to optimize the preamble allocation in a MRSS cell. The ML based approach may include an agent 610 that runs an artificial intelligence or ML algorithm (e.g., the first ML model and / or the second ML model). The ML algorithm of 610 may have inputs and output(s). The state 630 may be a representation of the current MRSS cell state, e.g., normal state, energy saving / efficiency state, high load state, and / or the like. The state 630 may be defined based on the load of each RAT of the MRSS cells e.g., load of the first RAT or load of the second RAT that represents a number of users requesting RACH preambles for random access. In an example, the number of users may be known to the shared network (environment) 650. In an example, the action 640 may be the probability distribution parameter (for selection of the RACH preambles) that may be learned by the environment (shared network) 650 and transmitted to the users to select RACH preambles. In an example, the reward 620 may include a utility (utility function) that the agent 610 receives for performing the right actions. In one example, the reward 620 may be preamble collision statistics that result from assigning a certain probability distribution parameter to the UEs for RACH preamble selection. In this case, the lower is the reward 620, the closer is the action 640 to the optimal value of the collision statistics.

[0082] FIG. 7 is a diagram illustrating an example of a supervised learning model (or ML model) with offline training to learn the probability distribution parameter for RACH preamble selection in a MRSS cell. An example of a ML model (e.g., the first ML model and / or the second ML model) module for predicting the RACH preamble selection policy (e.g., the probability distribution parameter) may be based on a (supervised) ML model. In an example, the supervised ML model may accept inputs such the first information and / or the second information, (e.g., as the collision statistics related, for instance, to allocated preambles and / or measured preambles, networks load statistics (e.g. 5G RAT load statistics, or6G RAT load statistics) and other auxiliary information). The information may be passed to the ML model internal layers that may include internal layers (hidden layer 1 , ... , hidden layer Nh). The internal layers may be based on neural networks (NN) blocks. The NN blocks may be of different types e.g., deep neural networks (DNNs), convolutional neural networks (CNNs), recurrent neural network (RNN), or long short-term memory (LSTM), and / or the like. The internal layers may be formed by trainable parameters. In one example, the information moves in the forward direction from the input to the output blocks. Therefore, in one variant of regression model, the output block may predict continuous values representing, for instance, the parameters that characterize the distribution of RACH preambles (o, p, A, and / or the like) for a given 5G RAT or 6G RAT network node. In an example, after training, the ML model may be deployed and used for inference. Each MRSS cell, may determine during a temporal window that defines the observation window at least one of collision statistic related to allocated preambles and / or network data traffic statistics.

[0083] FIG. 8 is a diagram illustrating an example of a reinforcement learning (RL) based method to learn the preamble probability distribution for PRACH in MRSS cell. The reinforcement learning agent (RL agent) 660 may include a policy TT module 663, a Q table 661 , and a state prediction module 662. The Q table 661 may receive a current observation and an action as inputs and may return a single scalar as output. The state prediction module 662 may include models such as the DNN, CNN, RNN, and / or the like. The policy TT module may be a function that takes as input a state s(t) 630 and return an action a(t) 640. The state 630 may be a representation of the current MRSS cell state, e.g., normal state, energy saving / efficiency state, high load state, and / or the like. In an example, the shared network (environment) 650 may include the MRSS cells such as the first network node and / or the second network node. In an example, action 640 may be the probability distribution parameter (for selection of the RACH preambles) that may be learned by the environment (shared network) 650 and transmitted to the users to select RACH preambles. In an example, the reward 620 may include a utility (utility function) that the RL agent 660 receives for performing the right actions. In one example, the reward 620 may be preamble collision statistics that result from assigning a certain probability distribution parameter to the UEs for RACH preamble selection. In this case, the lower is the reward 620, the closer is the action 640 to the optimal value of the collision statistics.

[0084] FIG. 9 is a flow chart illustrating operation of an apparatus (e.g., which may be a UE or user device, or other apparatus) according to an example embodiment. Operation 910 includes determining, by a first network node of a first radio access technology (RAT), first information of the first network node to determine a first random access channel (RACH) preamble selection policy for a user device. Operation 920 includes receiving, by the first network node from a second network node of a second RAT, second information of the second network node to determine the first RACH preamble selection policy. Operation 930 includes determining, by the first network node, the first RACH preamble selection policy based on the first information of the first network node and the second information of the second network node. Operation 940 includes transmitting to the user device the first RACH preamble selection policy.

[0085] With respect to the method of FIG. 9, the method may further include: wherein the first RACH preamble selection policy includes at least one of: a probability distribution parameter to select a RACH preamble among a predetermined set of candidate RACH preambles, wherein the probability distribution parameter includes at least one of: a type of a distribution function; or a parameter associated with the distribution function, the parameter including at least one of a mean value, or a variance value; or allocation information of RACH preambles for access of the user device via the first network node and the second network node.

[0086] With respect to the method of FIG. 9, the method may further include: wherein the first information of the first network node includes at least one of: a mode or state of operation of the first network node; arrival rate of random access requests for the first network node; a number of RACH preamble collisions at the first network node; or load information of the first network node.

[0087] With respect to the method of FIG. 9, the method may further include: wherein the second information of the second network node includes at least one of: a mode or state of operation of the second network node; arrival rate of random access requests for the second network node; a number of RACH preamble collisions at the second network node; or load information of the second network node.

[0088] With respect to the method of FIG. 9, the method may further include: wherein the second information of the second network node includes at least one of: a second RACH preamble selection policy; or a probability distribution parameter associated with the second RACH preamble selection policy.

[0089] With respect to the method of FIG. 9, the method may further include: wherein the second information is provided by a second machine learning (ML) model of the second network node.

[0090] With respect to the method of FIG. 9, the method may further include: determining a probability distribution parameter based on the first information of the first network node and the second information of the second network node, wherein the first RACH preamble selection policy is based on the probability distribution parameter; and transmitting the probability distribution parameter to the user device as the first RACH preamble selection policy.

[0091] With respect to the method of FIG. 9, the method may further include: wherein the first RACH preamble selection policy is based on: a first machine learning (ML) model of the first network node that is determined based on training an algorithm based on the first information of the first network node; and a second ML model of the second network node that is determined based on training an algorithm based on the second information available at the second network node.

[0092] With respect to the method of FIG. 9, the method may further include: wherein the first information of the first network node and the second information of the second network node that are used to train the algorithm of the first ML model and the second ML model, include at least one of: a mode or state of operation of the first network node and / or the second network node; arrival rate of random access requests for the first network node and / or the second network node; a number of RACH preamble collisions at the first network node and / or the second network node; or load information of the first network node and / or the second network node.

[0093] With respect to the method of FIG. 9, the method may further include: wherein the first information of the first network node and the second information of the second network node include at least one of: information of a probability distribution parameter for selection of a RACH preamble among a predetermined set of candidate RACH preambles; a trained ML model of the first ML model for prediction of the probability distribution parameter; a parameter for training of the first ML model; a trained ML model of the second ML model for prediction of the probability distribution parameter; or a parameter for training of the second ML model.

[0094] With respect to the method of FIG. 9, the method may further include: updating a parameter of a first ML model of the first network node based on at least one of: a number of RACH preamble collisions at the first network node and the second network node; or information of a mobility of the user device; and location information of the user device.

[0095] With respect to the method of FIG. 9, the method may further include: wherein the parameter of the of the first ML model includes at least one of: a weight (coefficient) associated with an input parameter; a parameter associated with a neural network block; or information of a regression model. With respect to the method of FIG. 9, the method may further include: wherein the information of the regression model includes at least one of: a linear regression; a polynomial regression; a support vector regression (SVR); a decision tree regression; or a random forest regression.

[0096] With respect to the method of FIG. 9, the method may further include: receiving a RACH preamble from two or more user devices; detecting a collision when a same RACH preamble is received at a same time from the two or more user devices; and determining a probability distribution parameter for selecting a RACH preamble based on at least one of: the first information of the first network node; the second information of the second network node; a first ML model of the first network node; and a second ML model of the second network node.

[0097] With respect to the method of FIG. 9, the method may further include: wherein the transmitting of the first RACH preamble selection policy is via a system information block (SIB) broadcast.

[0098] With respect to the method of FIG. 9, the method may further include: wherein the first network node is a first base station distributed unit (BS-DU).

[0099] With respect to the method of FIG. 9, the method may further include: wherein the second network node is a base station centralized unit (BS-CU) or a second BS-DU.

[0100] With respect to the method of FIG. 9, the method may further include: transmitting to the second network node the first information of the first network node.

[0101] With respect to the method of FIG. 9, the method may further include: wherein the first information of the first network node includes at least one of: the first RACH preamble selection policy; or a probability distribution parameter associated with the first RACH preamble selection policy.

[0102] With respect to the method of FIG. 9, the method may further include: receiving a RACH preamble from the user device based on the first RACH preamble selection policy. FIG. 10 is a flow chart illustrating operation of an apparatus (e.g., which may be a UE or user device, or other apparatus) according to an example embodiment. Operation 1010 includes receiving, by a user device from a first network node of a first radio access technology (RAT), a first random access channel (RACH) preamble selection policy, wherein the first RACH preamble selection policy is based on a first information of the first network node and a second information of a second network node of a second RAT. Operation 1020 includes selecting, by the user device, a RACH preamble based on the first RACH preamble selection policy. Operation 1030 includes transmitting, by the user device to the first network node, the selected RACH preamble.

[0103] With respect to the method of FIG. 10, the method may further include: wherein the first information of the first network node includes at least one of: a mode or state of operation of the first network node; arrival rate of random access requests for the first network node; a number of RACH preamble collisions at the first network node; or load information of the first network node.

[0104] With respect to the method of FIG. 10, the method may further include: wherein the second information of the second network node includes at least one of: a mode or state of operation of the second network node; arrival rate of random access requests for the second network node; a number of RACH preamble collisions at the second network node; or load information of the second network node.

[0105] With respect to the method of FIG. 10, the method may further include: wherein the first RACH preamble selection policy is based on: a first ML model of the first network node that is determined based on training an algorithm based on the first information available at the first network node; and a second ML model of the second network node that is determined based on training an algorithm based on the second information available at the second network node.

[0106] With respect to the method of FIG. 10, the method may further include: wherein the first RACH preamble selection policy includes at least one of: a probability distribution parameter to select a RACH preamble, wherein the probability distribution parameter includes at least one of: a type of a distribution function; or a parameter associated with the distribution function, the parameter including at least one of a mean value, or a variance value; or allocation information of RACH preambles for access of the user device via the first network node and the second network node.

[0107] With respect to the method of FIG. 10, the method may further include: receiving an update of the first RACH preamble selection policy.

[0108] With respect to the method of FIG. 10, the method may further include: wherein the selecting the RACH preamble based on the first RACH preamble selection policy includes: selecting a first RACH preamble from a set of RACH preambles when accessing the first network node of the first RAT; and selecting a second RACH preamble from the set of RACH preambles when accessing the second network node of the second RAT.

[0109] With respect to the method of FIG. 10, the method may further include: wherein the receiving of the first RACH preamble selection policy is via a system information block (SIB) broadcast.

[0110] With respect to the method of FIG. 10, the method may further include: receiving, by the user device from the second network node, a second RACH preamble selection policy, wherein the second RACH preamble selection policy is based on the first information of the first network node of the first RAT and the second information of the second network node of the second RAT; and selecting the RACH preamble based on the second RACH preamble selection policy.

[0111] Some examples will now be described, based on the description and figures provided herein.

[0112] Example A1 . An apparatus including: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to: determine, by the apparatus of a first radio access technology (RAT), first information of the apparatus to determine a first random access channel (RACH) preamble selection policy for a user device; receive from a second network node of a second RAT, second information of the second network node to determine the first RACH preamble selection policy; determine the first RACH preamble selection policy based on the first information of the apparatus and the second information of the second network node; and transmit to the user device the first RACH preamble selection policy.

[0113] Example A2. The apparatus of example A1 , wherein the first RACH preamble selection policy includes at least one of: a probability distribution parameter to select a RACH preamble among a predetermined set of candidate RACH preambles, wherein the probability distribution parameter includes at least one of: a type of a distribution function; or a parameter associated with the distribution function, the parameter including at least one of a mean value, or a variance value; or allocation information of RACH preambles for access of the user device via the apparatus and the second network node.

[0114] Example A3. The apparatus of any of examples A1 to A2, wherein the first information of the apparatus includes at least one of: a mode or state of operation of the apparatus; arrival rate of random access requests for the apparatus; a number of RACH preamble collisions at the apparatus; or load information of the apparatus.

[0115] Example A4. The apparatus of any of examples A1 to A3, wherein the second information of the second network node includes at least one of: a mode or state of operation of the second network node; arrival rate of random access requests for the second network node; a number of RACH preamble collisions at the second network node; or load information of the second network node.

[0116] Example A5. The apparatus of any of examples A1 to A4, wherein the second information of the second network node includes at least one of: a second RACH preamble selection policy; or a probability distribution parameter associated with the second RACH preamble selection policy.

[0117] Example A6. The apparatus of any of examples A1 to A5, wherein the second information is provided by a second machine learning (ML) model of the second network node. Example A7. The apparatus of any of examples A1 to A6, wherein the apparatus is further caused to determine a probability distribution parameter based on the first information of the apparatus and the second information of the second network node, wherein the first RACH preamble selection policy is based on the probability distribution parameter; and transmit the probability distribution parameter to the user device as the first RACH preamble selection policy.

[0118] Example A8. The apparatus of any of examples A1 to A7, wherein the first RACH preamble selection policy is based on: a first machine learning (ML) model of the apparatus that is determined based on training an algorithm based on the first information of the apparatus; and a second ML model of the second network node that is determined based on training an algorithm based on the second information available at the second network node.

[0119] Example A9. The apparatus of example A8, wherein the first information of the apparatus and the second information of the second network node that are used to train the algorithm of the first ML model and the second ML model, include at least one of: a mode or state of operation of the apparatus and / or the second network node; arrival rate of random access requests for the apparatus and / or the second network node; a number of RACH preamble collisions at the apparatus and / or the second network node; or load information of the apparatus and / or the second network node.

[0120] Example A10. The apparatus of any of examples A1 to A9, wherein the first information of the apparatus and the second information of the second network node include at least one of: information of a probability distribution parameter for selection of a RACH preamble among a predetermined set of candidate RACH preambles; a trained ML model of the first ML model for prediction of the probability distribution parameter; a parameter for training of the first ML model; a trained ML model of the second ML model for prediction of the probability distribution parameter; or a parameter for training of the second ML model.

[0121] Example A11 . The apparatus of any of examples A1 to A10, wherein the apparatus is further caused to update a parameter of a first ML model of the apparatus based on at least one of: a number of RACH preamble collisions at the apparatus and the second network node; or information of a mobility of the user device; and location information of the user device.

[0122] Example A12. The apparatus of example A11 , wherein the parameter of the of the first ML model includes at least one of: a weight (coefficient) associated with an input parameter; a parameter associated with a neural network block; or information of a regression model.

[0123] Example A13. The apparatus of example A12, wherein the information of the regression model includes at least one of: a linear regression; a polynomial regression; a support vector regression (SVR); a decision tree regression; or a random forest regression.

[0124] Example A14. The apparatus of any of examples A1 to A13, wherein the apparatus is further caused to: receive a RACH preamble from two or more user devices; detect a collision when a same RACH preamble is received at a same time from the two or more user devices; and determine a probability distribution parameter for selecting a RACH preamble based on at least one of: the first information of the apparatus; the second information of the second network node; a first ML model of the apparatus; and a second ML model of the second network node.

[0125] Example A15. The apparatus of any of examples A1 to A14, wherein the transmitting of the first RACH preamble selection policy is via a system information block (SIB) broadcast.

[0126] Example A16. The apparatus of any of examples A1 to A15, wherein the apparatus is a first base station distributed unit (BS-DU).

[0127] Example A17. The apparatus of any of examples A1 to A16, wherein the second network node is a base station centralized unit (BS-CU) or a second BS-DU.

[0128] Example A18. The apparatus of any of examples A1 to A17, wherein the apparatus is further caused to transmit to the second network node the first information of the apparatus. Example A19. The apparatus of any of examples A1 to A18, wherein the first information of the apparatus includes at least one of: the first RACH preamble selection policy; or a probability distribution parameter associated with the first RACH preamble selection policy.

[0129] Example A20. The apparatus of any of examples A1 to A19, wherein the apparatus is further caused to receive a RACH preamble from the user device based on the first RACH preamble selection policy.

[0130] Example B1 . An apparatus including: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to: receive from a first network node of a first radio access technology (RAT), a first random access channel (RACH) preamble selection policy, wherein the first RACH preamble selection policy is based on a first information of the first network node and a second information of a second network node of a second RAT; select a RACH preamble based on the first RACH preamble selection policy; and transmit to the first network node, the selected RACH preamble.

[0131] Example B2. The apparatus of example B1 , wherein the first information of the first network node includes at least one of: a mode or state of operation of the first network node; arrival rate of random access requests for the first network node; a number of RACH preamble collisions at the first network node; or load information of the first network node.

[0132] Example B3. The apparatus of any of examples B1 to B2, wherein the second information of the second network node includes at least one of: a mode or state of operation of the second network node; arrival rate of random access requests for the second network node; a number of RACH preamble collisions at the second network node; or load information of the second network node.

[0133] Example B4. The apparatus of any of examples B1 to B3, wherein the first RACH preamble selection policy is based on: a first ML model of the first network node that is determined based on training an algorithm based on the first information available at the first network node; and a second ML model of the second network node that is determined based on training an algorithm based on the second information available at the second network node.

[0134] Example B5. The apparatus of any of examples B1 to B4, wherein the first RACH preamble selection policy includes at least one of: a probability distribution parameter to select a RACH preamble, wherein the probability distribution parameter includes at least one of: a type of a distribution function; or a parameter associated with the distribution function, the parameter including at least one of a mean value, or a variance value; or allocation information of RACH preambles for access of the apparatus via the first network node and the second network node.

[0135] Example B6. The apparatus of any of examples B1 to B5, wherein the apparatus is further caused to receive an update of the first RACH preamble selection policy.

[0136] Example B7. The apparatus of any of examples B1 to B6, wherein the selecting the RACH preamble based on the first RACH preamble selection policy includes: selecting a first RACH preamble from a set of RACH preambles when accessing the first network node of the first RAT; and selecting a second RACH preamble from the set of RACH preambles when accessing the second network node of the second RAT.

[0137] Example B8. The apparatus of any of examples B1 to B7, wherein the receiving of the first RACH preamble selection policy is via a system information block (SIB) broadcast.

[0138] Example B9. The apparatus of any of examples B1 to B8, wherein the apparatus is further caused to: receive, from the second network node, a second RACH preamble selection policy, wherein the second RACH preamble selection policy is based on the first information of the first network node of the first RAT and the second information of the second network node of the second RAT; and select the RACH preamble based on the second RACH preamble selection policy.

[0139] Example C1 . A method including: determining, by a first network node of a first radio access technology (RAT), first information of the first network node to determine a first random access channel (RACH) preamble selection policy for a user device; receiving, by the first network node from a second network node of a second RAT, second information of the second network node to determine the first RACH preamble selection policy; determining, by the first network node, the first RACH preamble selection policy based on the first information of the first network node and the second information of the second network node; and transmitting to the user device the first RACH preamble selection policy.

[0140] Example C2. The method of example C1 , wherein the first RACH preamble selection policy includes at least one of: a probability distribution parameter to select a RACH preamble among a predetermined set of candidate RACH preambles, wherein the probability distribution parameter includes at least one of: a type of a distribution function; or a parameter associated with the distribution function, the parameter including at least one of a mean value, or a variance value; or allocation information of RACH preambles for access of the user device via the first network node and the second network node.

[0141] Example C3. The method of any of examples C1 to C2, wherein the first information of the first network node includes at least one of: a mode or state of operation of the first network node; arrival rate of random access requests for the first network node; a number of RACH preamble collisions at the first network node; or load information of the first network node.

[0142] Example C4. The method of any of examples C1 to C3, wherein the second information of the second network node includes at least one of: a mode or state of operation of the second network node; arrival rate of random access requests for the second network node; a number of RACH preamble collisions at the second network node; or load information of the second network node.

[0143] Example C5. The method of any of examples C1 to C4, wherein the second information of the second network node includes at least one of: a second RACH preamble selection policy; or a probability distribution parameter associated with the second RACH preamble selection policy. Example C6. The method of any of examples C1 to C5, wherein the second information is provided by a second machine learning (ML) model of the second network node.

[0144] Example C7. The method of any of examples C1 to C6, further including determining a probability distribution parameter based on the first information of the first network node and the second information of the second network node, wherein the first RACH preamble selection policy is based on the probability distribution parameter; and transmitting the probability distribution parameter to the user device as the first RACH preamble selection policy.

[0145] Example C8. The method of any of examples C1 to C7, wherein the first RACH preamble selection policy is based on: a first machine learning (ML) model of the first network node that is determined based on training an algorithm based on the first information of the first network node; and a second ML model of the second network node that is determined based on training an algorithm based on the second information available at the second network node.

[0146] Example C9. The method of example C8, wherein the first information of the first network node and the second information of the second network node that are used to train the algorithm of the first ML model and the second ML model, include at least one of: a mode or state of operation of the first network node and / or the second network node; arrival rate of random access requests for the first network node and / or the second network node; a number of RACH preamble collisions at the first network node and / or the second network node; or load information of the first network node and / or the second network node.

[0147] Example C10. The method of any of examples C1 to C9, wherein the first information of the first network node and the second information of the second network node include at least one of: information of a probability distribution parameter for selection of a RACH preamble among a predetermined set of candidate RACH preambles; a trained ML model of the first ML model for prediction of the probability distribution parameter; a parameter for training of the first ML model; a trained ML model of the second ML model for prediction of the probability distribution parameter; or a parameter for training of the second ML model.

[0148] Example C11. The method of any of examples C1 to C10, further including updating a parameter of a first ML model of the first network node based on at least one of: a number of RACH preamble collisions at the first network node and the second network node; or information of a mobility of the user device; and location information of the user device.

[0149] Example C12. The method of example C11 , wherein the parameter of the of the first ML model includes at least one of: a weight (coefficient) associated with an input parameter; a parameter associated with a neural network block; or information of a regression model.

[0150] Example C13. The method of example C12, wherein the information of the regression model includes at least one of: a linear regression; a polynomial regression; a support vector regression (SVR); a decision tree regression; or a random forest regression.

[0151] Example C14. The method of any of examples C1 to C13, further including: receiving a RACH preamble from two or more user devices; detecting a collision when a same RACH preamble is received at a same time from the two or more user devices; and determining a probability distribution parameter for selecting a RACH preamble based on at least one of: the first information of the first network node; the second information of the second network node; a first ML model of the first network node; and a second ML model of the second network node.

[0152] Example C15. The method of any of examples C1 to C14, wherein the transmitting of the first RACH preamble selection policy is via a system information block (SIB) broadcast.

[0153] Example C16. The method of any of examples C1 to C15, wherein the first network node is a first base station distributed unit (BS-DU). Example C17. The method of any of examples C1 to C16, wherein the second network node is a base station centralized unit (BS-CU) or a second BS-DU.

[0154] Example C18. The method of any of examples C1 to C17, further including transmitting to the second network node the first information of the first network node.

[0155] Example C19. The method of any of examples C1 to C18, wherein the first information of the first network node includes at least one of: the first RACH preamble selection policy; or a probability distribution parameter associated with the first RACH preamble selection policy.

[0156] Example C20. The method of any of examples C1 to C19, further including receiving a RACH preamble from the user device based on the first RACH preamble selection policy.

[0157] Example D1. A method including: receiving, by a user device from a first network node of a first radio access technology (RAT), a first random access channel (RACH) preamble selection policy, wherein the first RACH preamble selection policy is based on a first information of the first network node and a second information of a second network node of a second RAT; selecting, by the user device, a RACH preamble based on the first RACH preamble selection policy; and transmitting, by the user device to the first network node, the selected RACH preamble.

[0158] Example D2. The method of example D1 , wherein the first information of the first network node includes at least one of: a mode or state of operation of the first network node; arrival rate of random access requests for the first network node; a number of RACH preamble collisions at the first network node; or load information of the first network node.

[0159] Example D3. The method of any of examples D1 to D2, wherein the second information of the second network node includes at least one of: a mode or state of operation of the second network node; arrival rate of random access requests for the second network node; a number of RACH preamble collisions at the second network node; or load information of the second network node. Example D4. The method of any of examples D1 to D3, wherein the first RACH preamble selection policy is based on: a first ML model of the first network node that is determined based on training an algorithm based on the first information available at the first network node; and a second ML model of the second network node that is determined based on training an algorithm based on the second information available at the second network node.

[0160] Example D5. The method of any of examples D1 to D4, wherein the first RACH preamble selection policy includes at least one of: a probability distribution parameter to select a RACH preamble, wherein the probability distribution parameter includes at least one of: a type of a distribution function; or a parameter associated with the distribution function, the parameter including at least one of a mean value, or a variance value; or allocation information of RACH preambles for access of the user device via the first network node and the second network node.

[0161] Example D6. The method of any of examples D1 to D5, further including receiving an update of the first RACH preamble selection policy.

[0162] Example D7. The method of any of examples D1 to D6, wherein the selecting the RACH preamble based on the first RACH preamble selection policy includes: selecting a first RACH preamble from a set of RACH preambles when accessing the first network node of the first RAT; and selecting a second RACH preamble from the set of RACH preambles when accessing the second network node of the second RAT.

[0163] Example D8. The method of any of examples D1 to D7, wherein the receiving of the first RACH preamble selection policy is via a system information block (SIB) broadcast.

[0164] Example D9. The method of any of examples D1 to D8, further including: receiving, by the user device from the second network node, a second RACH preamble selection policy, wherein the second RACH preamble selection policy is based on the first information of the first network node of the first RAT and the second information of the second network node of the second RAT; and selecting the RACH preamble based on the second RACH preamble selection policy. FIG. 11 is a block diagram of a wireless station or node (e.g., UE, user device, AP, BS, eNB, gNB, RAN node, network node, TRP, or other node) 1300 according to an example embodiment. The wireless station 1300 may include, for example, one or more (e.g., two as shown in FIG. 11 ) RF (radio frequency) or wireless transceivers 1302A, 1302B, where each wireless transceiver includes a transmitter to transmit signals and a receiver to receive signals. The wireless station also includes a processor or control unit / entity (controller) 1304 to execute instructions or software and control transmission and receptions of signals, and a memory 1306 to store data and / or instructions.

[0165] Processor 1304 may also make decisions or determinations, generate frames, packets or messages for transmission, decode received frames or messages for further processing, and other tasks or functions described herein. Processor 1304, which may be a baseband processor, for example, may generate messages, packets, frames or other signals for transmission via wireless transceiver 1302 (1302A or 1302B). Processor 1304 may control transmission of signals or messages over a wireless network, and may control the reception of signals or messages, etc., via a wireless network (e.g., after being down-converted by wireless transceiver 1302, for example). Processor 1304 may be programmable and capable of executing software or other instructions stored in memory or on other computer media to perform the various tasks and functions described above, such as one or more of the tasks or methods described above. Processor 1304 may be (or may include), for example, hardware, programmable logic, a programmable processor that executes software or firmware, and / or any combination of these. Using other terminology, processor 1304 and transceiver 1302 together may be considered as a wireless transmitter / receiver system, for example.

[0166] In addition, referring to FIG. 11 , a controller (or processor) 1308 may execute software and instructions, and may provide overall control for the station 1300, and may provide control for other systems not shown in FIG. 11 , such as controlling input / output devices (e.g., display, keypad), and / or may execute software for one or more applications that may be provided on wireless station 1300, such as, for example, an email program, audio / video applications, a word processor, a Voice over IP application, or other application or software. In addition, a storage medium may be provided that includes stored instructions, which when executed by a controller or processor may result in the processor 1304, or other controller or processor, performing one or more of the functions or tasks described above.

[0167] According to another example embodiment, RF or wireless transceiver(s) 1302A / 1302B may receive signals or data and / or transmit or send signals or data. Processor 1304 (and possibly transceivers 1302A / 1302B) may control the RF or wireless transceiver 1302A or 1302B to receive, send, broadcast or transmit signals or data.

[0168] Example embodiments are provided or described for each of the example methods, including: An apparatus (e.g., 1300, FIG. 11) including means (e.g., processor 1304, RF transceivers 1302A and / or 1302B, and / or memory 1306, in FIG. 11 ) for carrying out any of the methods; a non-transitory computer-readable storage medium (e.g., memory 1306, FIG. 11) comprising instructions stored thereon that, when executed by at least one processor (processor 1304, FIG. 11 ), are configured to cause a computing system (e.g., 1300, FIG. 11) to perform any of the example methods; and an apparatus (e.g., 1300, FIG. 11 ) including at least one processor (e.g., processor 1304, FIG. 11 ), and at least one memory (e.g., memory 1306, FIG. 11) including computer program code, the at least one memory (1306) and the computer program code configured to, with the at least one processor (1304), cause the apparatus (e.g., 1300) at least to perform any of the example methods.

[0169] Embodiments of the various techniques described herein may be implemented in digital electronic circuitry, or in computer hardware, firmware, software, or in combinations of them. Embodiments may be implemented as a computer program product, i.e. , a computer program tangibly embodied in an information carrier, e.g., in a machine-readable storage device or in a propagated signal, for execution by, or to control the operation of, a data processing apparatus, e.g., a programmable processor, a computer, or multiple computers. Embodiments may also be provided on a computer readable medium or computer readable storage medium, which may be a non-transitory medium. Embodiments of the various techniques may also include embodiments provided via transitory signals or media, and / or programs and / or software embodiments that are downloadable via the Internet or other network(s), either wired networks and / or wireless networks. In addition, embodiments may be provided via machine type communications (MTC), and also via an Internet of Things (IOT).

[0170] As used in this application, the term ‘circuitry’ or “circuit” refers to all of the following: (a) hardware-only circuit implementations, such as implementations in only analog and / or digital circuitry, and (b) combinations of circuits and soft-ware (and / or firmware), such as (as applicable): (i) a combination of processor(s) or (ii) portions of processor(s) / software including digital signal processor(s), software, and memory(ies) that work together to cause an apparatus to perform various functions, and (c) circuits, such as a microprocessor(s) or a portion of a microprocessor(s), that require software or firmware for operation, even if the software or firmware is not physically present. This definition of ‘circuitry’ applies to all uses of this term in this application. As a further example, as used in this application, the term ‘circuitry’ would also cover an implementation of merely a processor (or multiple processors) or a portion of a processor and its (or their) accompanying software and / or firmware. The term ‘circuitry’ would also cover, for example and if applicable to the particular element, a baseband integrated circuit or applications processor integrated circuit for a mobile phone or a similar integrated circuit in a server, a cellular network device, or another network device.

[0171] The computer program may be in source code form, object code form, or in some intermediate form, and it may be stored in some sort of carrier, distribution medium, or computer readable medium, which may be any entity or device capable of carrying the program. Such carriers include a record medium, computer memory, read-only memory, photoelectrical and / or electrical carrier signal, telecommunications signal, and software distribution package, for example. Depending on the processing power needed, the computer program may be executed in a single electronic digital computer, or it may be distributed amongst a number of computers.

[0172] Furthermore, embodiments of the various techniques described herein may use a cyber-physical system (CPS) (a system of collaborating computational elements controlling physical entities). CPS may enable the embodiment and exploitation of massive amounts of interconnected ICT devices (sensors, actuators, processors microcontrollers, ...) embedded in physical objects at different locations. Mobile cyber physical systems, in which the physical system in question has inherent mobility, are a subcategory of cyber-physical systems. Examples of mobile physical systems include mobile robotics and electronics transported by humans or animals. The rise in popularity of smartphones has increased interest in the area of mobile cyberphysical systems. Therefore, various embodiments of techniques described herein may be provided via one or more of these technologies.

[0173] A computer program, such as the computer program(s) described above, can be written in any form of programming language, including compiled or interpreted languages, and can be deployed in any form, including as a standalone program or as a module, component, subroutine, or other unit or part of it suitable for use in a computing environment. A computer program can be deployed to be executed on one computer or on multiple computers at one site or distributed across multiple sites and interconnected by a communication network.

[0174] Method steps may be performed by one or more programmable processors executing a computer program or computer program portions to perform functions by operating on input data and generating output. Method steps also may be performed by, and an apparatus may be implemented as, special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application specific integrated circuit).

[0175] Processors suitable for the execution of a computer program include, by way of example, both general and special purpose microprocessors, and any one or more processors of any kind of digital computer, chip or chipset. Generally, a processor will receive instructions and data from a read-only memory or a random access memory or both. Elements of a computer may include at least one processor for executing instructions and one or more memory devices for storing instructions and data. Generally, a computer also may include, or be operatively coupled to receive data from or transfer data to, or both, one or more mass storage devices for storing data, e.g., magnetic, magnetooptical disks, or optical disks. Information carriers suitable for embodying computer program instructions and data include all forms of non-volatile memory, including by way of example semiconductor memory devices, e.g., EPROM, EEPROM, and flash memory devices; magnetic disks, e.g., internal hard disks or removable disks; magneto-optical disks; and CD-ROM and DVD-ROM disks. The processor and the memory may be supplemented by, or incorporated in, special purpose logic circuitry.

[0176] To provide for interaction with a user, embodiments may be implemented on a computer having a display device, e.g., a cathode ray tube (CRT) or liquid crystal display (LCD) monitor, for displaying information to the user and a user interface, such as a keyboard and a pointing device, e.g., a mouse or a trackball, by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback, e.g., visual feedback, auditory feedback, or tactile feedback; and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0177] Embodiments may be implemented in a computing system that includes a backend component, e.g., as a data server, or that includes a middleware component, e.g., an application server, or that includes a frontend component, e.g., a client computer having a graphical user interface or a Web browser through which a user can interact with an embodiment, or any combination of such backend, middleware, or frontend components. Components may be interconnected by any form or medium of digital data communication, e.g., a communication network. Examples of communication networks include a local area network (LAN) and a wide area network (WAN), e.g., the Internet.

[0178] While certain features of the described embodiments have been illustrated as described herein, many modifications, substitutions, changes and equivalents will now occur to those skilled in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit of the various embodiments.

Claims

CLAIMS:1 . An apparatus comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to: determine, by the apparatus of a first radio access technology (RAT), first information of the apparatus to determine a first random access channel (RACH) preamble selection policy for a user device; receive from a second network node of a second RAT, second information of the second network node to determine the first RACH preamble selection policy; determine the first RACH preamble selection policy based on the first information of the apparatus and the second information of the second network node; and transmit to the user device the first RACH preamble selection policy.

2. The apparatus of claim 1 , wherein the first RACH preamble selection policy comprises at least one of: a probability distribution parameter to select a RACH preamble among a predetermined set of candidate RACH preambles, wherein the probability distribution parameter comprises at least one of: a type of a distribution function; or a parameter associated with the distribution function, the parameter comprising at least one of a mean value, or a variance value; or allocation information of RACH preambles for access of the user device via the apparatus and the second network node.

3. The apparatus of any of claims 1 to 2, wherein the first information of the apparatus comprises at least one of: a mode or state of operation of the apparatus; arrival rate of random access requests for the apparatus; a number of RACH preamble collisions at the apparatus; or load information of the apparatus.

4. The apparatus of any of claims 1 to 3, wherein the second information of the second network node comprises at least one of: a mode or state of operation of the second network node; arrival rate of random access requests for the second network node; a number of RACH preamble collisions at the second network node; or load information of the second network node.

5. The apparatus of any of claims 1 to 4, wherein the second information of the second network node comprises at least one of: a second RACH preamble selection policy; or a probability distribution parameter associated with the second RACH preamble selection policy.

6. The apparatus of any of claims 1 to 5, wherein the second information is provided by a second machine learning (ML) model of the second network node.

7. The apparatus of any of claims 1 to 6, wherein the apparatus is further caused to determine a probability distribution parameter based on the first information of the apparatus and the second information of the second network node, wherein the first RACH preamble selection policy is based on the probability distribution parameter; and transmit the probability distribution parameter to the user device as the first RACH preamble selection policy.

8. The apparatus of any of claims 1 to 7, wherein the first RACH preamble selection policy is based on: a first machine learning (ML) model of the apparatus that is determined based on training an algorithm based on the first information of the apparatus; and a second ML model of the second network node that is determined based on training an algorithm based on the second information available at the second network node.

9. The apparatus of claim 8, wherein the first information of the apparatus and the second information of the second network node that are used to train the algorithm of the first ML model and the second ML model, comprise at least one of: a mode or state of operation of the apparatus and / or the second network node; arrival rate of random access requests for the apparatus and / or the second network node; a number of RACH preamble collisions at the apparatus and / or the second network node; or load information of the apparatus and / or the second network node.

10. The apparatus of any of claims 1 to 9, wherein the first information of the apparatus and the second information of the second network node comprise at least one of: information of a probability distribution parameter for selection of a RACH preamble among a predetermined set of candidate RACH preambles; a trained ML model of the first ML model for prediction of the probability distribution parameter; a parameter for training of the first ML model; a trained ML model of the second ML model for prediction of the probability distribution parameter; or a parameter for training of the second ML model.11 . The apparatus of any of claims 1 to 10, wherein the apparatus is further caused to update a parameter of a first ML model of the apparatus based on at least one of: a number of RACH preamble collisions at the apparatus and the second network node; information of a mobility of the user device; or location information of the user device.

12. The apparatus of claim 11 wherein the parameter of the of the first ML model comprises at least one of: a weight coefficient associated with an input parameter; a parameter associated with a neural network block; orinformation of a regression model.

13. The apparatus of any of claims 1 to 12, wherein the apparatus is further caused to: receive a RACH preamble from two or more user devices; detect a collision when a same RACH preamble is received at a same time from the two or more user devices; and determine a probability distribution parameter associated with the first RACH preamble selection policy, the probability distribution parameter being for selecting a RACH preamble based on at least one of: the first information of the apparatus; the second information of the second network node; a first ML model of the apparatus; and a second ML model of the second network node.

14. The apparatus of any of claims 1 to 13, wherein the transmitting of the first RACH preamble selection policy is via a system information block (SIB) broadcast.

15. The apparatus of any of claims 1 to 14, wherein: the apparatus is a first base station distributed unit (BS-DU); and the second network node is a base station centralized unit (BS-CU) or a second BS-DU.

16. The apparatus of any of claims 1 to 15, wherein the apparatus is further caused to transmit to the second network node the first information of the apparatus, wherein the first information of the apparatus comprises at least one of: the first RACH preamble selection policy; or a probability distribution parameter associated with the first RACH preamble selection policy.

17. An apparatus comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to:receive from a first network node of a first radio access technology (RAT), a first random access channel (RACH) preamble selection policy, wherein the first RACH preamble selection policy is based on a first information of the first network node and a second information of a second network node of a second RAT; select a RACH preamble based on the first RACH preamble selection policy; and transmit to the first network node, the selected RACH preamble.

18. The apparatus of claim 17, wherein the first information of the first network node comprises at least one of: a mode or state of operation of the first network node; arrival rate of random access requests for the first network node; a number of RACH preamble collisions at the first network node; or load information of the first network node.

19. The apparatus of any of claims 17 to 18, wherein the second information of the second network node comprises at least one of: a mode or state of operation of the second network node; arrival rate of random access requests for the second network node; a number of RACH preamble collisions at the second network node; or load information of the second network node.

20. The apparatus of any of claims 17 to 19, wherein the first RACH preamble selection policy comprises at least one of: a probability distribution parameter to select a RACH preamble, wherein the probability distribution parameter comprises at least one of: a type of a distribution function; or a parameter associated with the distribution function, the parameter comprising at least one of a mean value, or a variance value; or allocation information of RACH preambles for access of the apparatus via the first network node and the second network node.21 . The apparatus of any of claims 17 to 20, wherein the apparatus is further caused to receive an update of the first RACH preamble selection policy.

22. The apparatus of any of claims 17 to 21 , wherein the selecting the RACH preamble based on the first RACH preamble selection policy comprises: selecting a first RACH preamble from a set of RACH preambles when accessing the first network node of the first RAT; and selecting a second RACH preamble from the set of RACH preambles when accessing the second network node of the second RAT.

23. The apparatus of any of claims 17 to 22, wherein the receiving of the first RACH preamble selection policy is via a system information block (SIB) broadcast.

24. A method comprising: determining, by a first network node of a first radio access technology (RAT), first information of the first network node to determine a first random access channel (RACH) preamble selection policy for a user device; receiving, by the first network node from a second network node of a second RAT, second information of the second network node to determine the first RACH preamble selection policy; determining, by the first network node, the first RACH preamble selection policy based on the first information of the first network node and the second information of the second network node; and transmitting to the user device the first RACH preamble selection policy.

25. A method comprising: receiving, by a user device from a first network node of a first radio access technology (RAT), a first random access channel (RACH) preamble selection policy, wherein the first RACH preamble selection policy is based on a first information of the first network node and a second information of a second network node of a second RAT; selecting, by the user device, a RACH preamble based on the first RACH preamble selection policy; and transmitting, by the user device to the first network node, the selected RACH preamble.

Images