Communication method and device of leading matrix

Abstract

A method is disclosed, comprising: generating a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble; obtaining a sequence encoded using the preamble matrix; and transmitting the sequence to one or more user devices.

Description

Technical Field

[0001] The following example embodiments relate to wireless communications. Background Art

[0002] Since resources are limited, it is desirable to improve the use of network resources. Summary of the Invention

[0003] The scope of protection sought by various exemplary embodiments is defined by the independent claims. Exemplary embodiments and features described in this specification that do not fall within the scope of the independent claims, if any, should be interpreted as examples that aid in understanding the various embodiments.

[0004] According to one aspect, an apparatus is provided that includes at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to at least: generate a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble; obtain a sequence encoded with the preamble matrix; and transmit the sequence to one or more user devices.

[0005] According to another aspect, an apparatus is provided that includes: means for generating a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble; means for obtaining a sequence encoded with the preamble matrix; and means for transmitting the sequence to one or more user equipment.

[0006] According to another aspect, a method is provided that includes generating a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble; obtaining a sequence encoded with the preamble matrix; and transmitting the sequence to one or more user devices.

[0007] According to another aspect, a computer program is provided, comprising instructions that, when executed by a device, cause the device to perform at least the following operations: generate a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble; obtain a sequence encoded with the preamble matrix; and transmit the sequence to one or more user devices.

[0008] According to another aspect, a computer-readable medium is provided, comprising program instructions that, when executed by a device, cause the device to perform at least the following operations: generate a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble; obtain a sequence encoded with the preamble matrix; and transmit the sequence to one or more user devices.

[0009] According to another aspect, a non-transitory computer-readable medium is provided, comprising program instructions that, when executed by a device, cause the device to perform at least the following operations: generate a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble; obtain a sequence encoded with the preamble matrix; and transmit the sequence to one or more user devices.

[0010] According to another aspect, an apparatus is provided, comprising at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to at least: receive a sequence from a network element of a radio access network, wherein the sequence is encoded as a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble; decode the sequence; generate a root sequence and a cyclic shift based at least in part on the decoding; generate the random access preamble based on the root sequence and the cyclic shift; and transmit the random access preamble to the network element.

[0011] According to another aspect, an apparatus is provided that includes: means for receiving a sequence from a network element of a radio access network, wherein the sequence is encoded as a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble; means for decoding the sequence; means for generating a root sequence and a cyclic shift based at least in part on the decoding; means for generating a random access preamble based on the root sequence and the cyclic shift; and means for transmitting the random access preamble to the network element.

[0012] According to another aspect, a method is provided that includes receiving a sequence from a network element of a radio access network, wherein the sequence is encoded as a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble; decoding the sequence; generating a root sequence and a cyclic shift based at least in part on the decoding; generating a random access preamble based on the root sequence and the cyclic shift; and transmitting the random access preamble to the network element.

[0013] According to another aspect, a computer program is provided, comprising instructions that, when executed by a device, cause the device to perform at least the following operations: receive a sequence from a network element of a radio access network, wherein the sequence is encoded as a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble; decode the sequence; generate a root sequence and a cyclic shift based at least in part on the decoding; generate a random access preamble based on the root sequence and the cyclic shift; and transmit the random access preamble to the network element.

[0014] According to another aspect, a computer-readable medium is provided, comprising program instructions that, when executed by a device, cause the device to perform at least the following operations: receive a sequence from a network element of a radio access network, wherein the sequence is encoded as a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble; decode the sequence; generate a root sequence and a cyclic shift based at least in part on the decoding; generate a random access preamble based on the root sequence and the cyclic shift; and transmit the random access preamble to the network element.

[0015] According to another aspect, a non-transitory computer-readable medium is provided, comprising program instructions that, when executed by a device, cause the device to perform at least the following operations: receive a sequence from a network element of a radio access network, wherein the sequence is encoded with a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble; decode the sequence; generate a root sequence and a cyclic shift based at least in part on the decoding; generate a random access preamble based on the root sequence and the cyclic shift; and transmit the random access preamble to the network element.

[0016] According to another aspect, an apparatus is provided, comprising at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to at least: create a first matrix comprising a plurality of random access preamble sets; select a sample batch from the first matrix; input the sample batch to an encoder having at least one neural network layer; receive a sequence as output from the encoder, wherein the sequence is encoded together with the sample batch; input the sequence to a decoder having at least one neural network layer; receive a second matrix as output from the decoder; determine a binary cross entropy loss comparing the sample batch and the second matrix; propagate the binary cross entropy loss to the decoder and the encoder via an optimizer; and repeat the steps of inputting the sample batch, receiving the sequence, inputting the sequence, receiving the second matrix, determining, and propagating until the binary cross entropy loss is below a threshold.

[0017] According to another aspect, an apparatus is provided, comprising: means for creating a first matrix comprising a plurality of random access preamble sets; means for selecting a sample batch from the first matrix; means for inputting the sample batch to an encoder having at least one neural network layer; means for receiving a sequence as output from the encoder, wherein the sequence is encoded together with the sample batch; means for inputting the sequence to a decoder having at least one neural network layer; means for receiving a second matrix as output from the decoder; means for determining a binary cross entropy loss comparing the sample batch and the second matrix; means for propagating the binary cross entropy loss to the decoder and the encoder via an optimizer; and means for repeating the steps of selecting a sample batch, inputting a sample batch, receiving a sequence, inputting a sequence, receiving a second matrix, determining, and propagating until the binary cross entropy loss is below a threshold.

[0018] According to another aspect, a method is provided, comprising: creating a first matrix comprising a plurality of random access preamble sets; selecting a sample batch from the first matrix; inputting the sample batch into an encoder having at least one neural network layer; receiving a sequence as an output from the encoder, wherein the sequence is encoded together with the sample batch; inputting the sequence into a decoder having at least one neural network layer; receiving a second matrix as an output from the decoder; determining a binary cross entropy loss comparing the sample batch and the second matrix; propagating the binary cross entropy loss to the decoder and the encoder via an optimizer; and repeating the steps of selecting the sample batch, inputting the sample batch, receiving the sequence, inputting the sequence, receiving the second matrix, determining, and propagating until the binary cross entropy loss is below a threshold.

[0019] According to another aspect, a computer program is provided, comprising instructions that, when executed by a device, cause the device to perform at least the following operations: creating a first matrix including a plurality of random access preamble sets; selecting a sample batch from the first matrix; inputting the sample batch into an encoder having at least one neural network layer; receiving a sequence as output from the encoder, wherein the sequence is encoded together with the sample batch; inputting the sequence into a decoder having at least one neural network layer; receiving a second matrix as output from the decoder; determining a binary cross entropy loss comparing the sample batch and the second matrix; propagating the binary cross entropy loss to the decoder and the encoder through an optimizer; and repeating the steps of selecting a sample batch, inputting a sample batch, receiving a sequence, inputting a sequence, receiving a second matrix, determining, and propagating until the binary cross entropy loss is below a threshold.

[0020] According to another aspect, a computer-readable medium is provided, comprising program instructions that, when executed by a device, cause the device to perform at least the following operations: creating a first matrix comprising a plurality of random access preamble sets; selecting a sample batch from the first matrix; inputting the sample batch to an encoder having at least one neural network layer; receiving a sequence as an output from the encoder, wherein the sequence is encoded together with the sample batch; inputting the sequence to a decoder having at least one neural network layer; receiving a second matrix as an output from the decoder; determining a binary cross entropy loss comparing the sample batch and the second matrix; propagating the binary cross entropy loss to the decoder and the encoder through an optimizer; and repeating the steps of inputting the sample batch, receiving the sequence, inputting the sequence, receiving the second matrix, determining, and propagating until the binary cross entropy loss is below a threshold.

[0021] According to another aspect, a non-transitory computer-readable medium is provided, comprising program instructions that, when executed by a device, cause the device to perform at least the following operations: creating a first matrix comprising a plurality of random access preamble sets; selecting a sample batch from the first matrix; inputting the sample batch to an encoder having at least one neural network layer; receiving a sequence as an output from the encoder, wherein the sequence is encoded together with the sample batch; inputting the sequence to a decoder having at least one neural network layer; receiving a second matrix as an output from the decoder; determining a binary cross entropy loss comparing the sample batch and the second matrix; propagating the binary cross entropy loss to the decoder and the encoder through an optimizer; and repeating the steps of inputting the sample batch, receiving the sequence, inputting the sequence, receiving the second matrix, determining, and propagating until the binary cross entropy loss is below a threshold.

[0022] According to another aspect, a system is provided, comprising at least one or more user devices and a network element of a radio access network. The network element is configured to: generate a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble; obtain a sequence encoded using the preamble matrix; and transmit the sequence to the one or more user devices. The one or more user devices are configured to: receive the sequence from the network element; decode the sequence; generate a root sequence and a cyclic shift based at least in part on the decoding; generate a random access preamble based on the root sequence and the cyclic shift; and transmit the random access preamble to the network element.

[0023] According to another aspect, a system is provided, comprising at least one or more user equipment and a network element of a radio access network. The network element comprises: means for generating a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble; means for obtaining a sequence encoded using the preamble matrix; and means for transmitting the sequence to one or more user equipment. The one or more user equipment comprises: means for receiving the sequence from the network element; means for decoding the sequence; means for generating a root sequence and a cyclic shift based at least in part on the decoding; means for generating a random access preamble based on the root sequence and the cyclic shift; and means for transmitting the random access preamble to the network element. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In the following, various example embodiments will be described in more detail with reference to the accompanying drawings, in which

[0025] Figure 1 An example of a cellular communication network is shown;

[0026] Figure 2 shows a signaling diagram according to an example embodiment;

[0027] Figure 3 shows a flow chart according to an example embodiment;

[0028] Figure 4 shows a flow chart according to an example embodiment;

[0029] Figure 5 An example for configuring a network element to utilize a preamble matrix as a preamble set is shown;

[0030] Figure 6 An example of encoding a leading matrix into a sequence is shown;

[0031] Figure 7A A first decoder option for decoding a preamble matrix from a sequence is shown;

[0032] Figure 7B A second decoder option for decoding a preamble matrix from a sequence is shown;

[0033] Figure 8 An example for generating a random access preamble is shown;

[0034] Figure 9 shows a flow chart according to an example embodiment;

[0035] Figure 10 shows a flow chart according to an example embodiment;

[0036] Figure 11 An example scenario is illustrated;

[0037] Figure 12 An example of a device is shown;

[0038] Figure 13 An example of a device is shown;

[0039] Figure 14 An example of a device is shown;

[0040] Figure 15 Illustrated an example of an artificial neural network

[0041] Figure 16 An example of a compute node is illustrated. DETAILED DESCRIPTION

[0042] The following embodiments are illustrative. Although the specification may refer to "one," "an," or "some" embodiments in several places in the text, this does not necessarily mean that each reference refers to the same embodiment or that a particular feature applies to only a single embodiment. Individual features of different embodiments may also be combined to provide other embodiments.

[0043] In the following, different example embodiments will be described using different example embodiments as examples of access architectures to which the example embodiments may be applied, a radio access architecture based on Long Term Evolution (LTE Advanced, LTE-A), New Radio (NR, 5G), Beyond 5G or Sixth Generation (6G), without limiting the example embodiments to such architectures, however. It will be apparent to a person skilled in the art that the example embodiments may also be applied to other kinds of communication networks with suitable devices by appropriately adapting the parameters and procedures. Some examples of other options for applicable systems may be Universal Mobile Telecommunications System (UMTS) Radio Access Network (UTRAN or E-UTRAN), Long Term Evolution (LTE, essentially the same as E-UTRA), Wireless Local Area Network (WLAN or Wi-Fi), WiMAX for Microwave Access (WiMAX), Personal Communications Service (PCS), Worldwide interoperability of protocols for interconnection between networks based on Wideband Code Division Multiple Access (WCDMA), systems using Ultra-Wideband (UWB) technology, sensor networks, Mobile Ad Hoc Networks (MANETs) and Multimedia Subsystems (IMS), or any combination thereof.

[0044] Figure 1 An example of a simplified system architecture is depicted showing some components and functional entities which are logical units and whose implementation may differ from what is shown. Figure 1 The connections shown in are logical connections; the actual physical connections may be different. It will be apparent to those skilled in the art that the system may also include Figure 1 Other functions and structures than those shown in .

[0045] However, the exemplary embodiments are not limited to the systems given as examples, but a person skilled in the art may apply the solution to other communication systems having the necessary properties.

[0046] Figure 1 The example of FIG. 1 shows a portion of an exemplary radio access network.

[0047] Figure 1User equipment 100 and 102 are shown configured to communicate with an access node (AN) 104, such as an evolved Node B (abbreviated as eNB or eNodeB) or a next generation Node B (abbreviated as gNB or gNodeB), in a wireless connection over one or more communication channels, providing a radio cell of the radio cell. The physical link from the user equipment to the access node can be called an uplink (UL) or reverse link, and the physical link from the access node to the user equipment can be called a downlink (DL) or forward link. The user equipment can also communicate directly with another user equipment via sidelink (SL) communications. It should be understood that an access node or its functionality can be implemented using any entity such as a node, host, server, or access point suitable for such usage.

[0048] A communication system may include more than one access node, in which case the access nodes may also be configured to communicate with each other via wired or wireless links designed for this purpose. These links may be used for signaling purposes or to route data from one access node to another. An access node may be a computing device configured to control the radio resources of the communication system to which it is coupled. An access node may also be referred to as a base station, a base transceiver station (BTS), an access point, or any other type of interface device, including a relay station, capable of operating in a wireless environment. An access node may include or be coupled to a transceiver. From the access node's transceiver, a connection may be provided to an antenna unit that establishes a bidirectional radio link to a user device. The antenna unit may include multiple antennas or antenna elements. The access node may also be connected to the core network 110 (CN or Next Generation Core NGC). Depending on the deployed technology, the counterpart to which the access node may be connected on the CN side may include a serving gateway (S-GW, which routes and forwards user data packets), a packet data network gateway (P-GW) for providing user devices with connectivity to external packet data networks, a user plane function (UPF), a mobility management entity (MME), or an access and mobility management function (AMF), among others.

[0049] User equipment describes a type of device that can allocate and assign resources on the air interface, and thus any features of user equipment described herein can be implemented using corresponding devices (eg, relay nodes).

[0050] An example of such a relay node may be a layer 3 relay towards an access node (self-backhaul relay). A self-backhaul relay node may also be referred to as an integrated access and backhaul (IAB) node. An IAB node may include two logical parts: a mobile terminal (MT) part responsible for the backhaul link (i.e., the link between the IAB node and the donor node, also referred to as the parent node) and a distributed unit (DU) part responsible for the access link (i.e., the sub-link between the IAB node and the user equipment, and / or the sub-link between the IAB node and other IAB nodes (multi-hop scenario)).

[0051] Another example of such a relay node may be a layer 1 relay, also known as a repeater, which may amplify signals received from an access node and forward them to a user device, and / or amplify signals received from a user device and forward them to an access node.

[0052] User equipment may also be referred to as a subscriber unit, mobile station, remote terminal, access terminal, user terminal, terminal device, or user equipment (UE), to name a few names or devices. User equipment may refer to a portable computing device, including wireless mobile communication devices with or without a subscriber identity module (SIM), including but not limited to the following types of devices: mobile stations (mobile phones), smartphones, personal digital assistants (PDAs), handsets, devices using wireless modems (such as alarm or measurement devices), laptop computers and / or touch screen computers, tablet computers, game consoles, notebooks, multimedia devices, reduced capability (RedCap) devices, wireless sensor devices, or any device integrated into a vehicle.

[0053] It will be appreciated that a user device may also be an almost exclusively uplink-only device, an example of which may be a camera or camcorder that uploads image or video strips to a network. A user device may also be a device capable of operating in an Internet of Things (IoT) network, which is a scenario in which objects may be provided with the ability to transmit data over a network without requiring human-to-human or human-to-computer interaction. User devices may also utilize the cloud. In some applications, a user device may comprise a small portable or wearable device with a radio (e.g., a watch, headphones, or glasses), and computing may be performed in the cloud or in another user device. A user device (or, in some example embodiments, a layer 3 relay node) may be configured to perform one or more user device functions.

[0054] The various techniques described here can also be applied to cyber-physical systems (CPS)—systems of collaborative computing elements that control physical entities. CPS can implement and utilize a large number of interconnected ICT devices (sensors, actuators, processors, microcontrollers, etc.) embedded in physical objects at different locations. Mobile cyber-physical systems, where the physical systems in question may have inherent mobility, are a subclass of cyber-physical systems. Examples of mobile physical systems include mobile robots and electronic devices transported by humans or animals.

[0055] Additionally, although the device has been depicted as a single entity, different units, processors and / or memory units may be implemented (not all of which are shown). Figure 1 ).

[0056] 5G can use multiple-input, multiple-output (MIMO) antennas, more base stations or nodes than LTE (the so-called small cell concept), including macro sites operating in cooperation with smaller sites, and employ a variety of radio technologies depending on service requirements, use cases and / or available spectrum.

[0057] 5G mobile communications can support a wide range of use cases and related applications, including video streaming, augmented reality, different data sharing methods, and various forms of machine-type applications (such as (massive) machine-type communications (mMTC), including vehicle safety, different sensors, and real-time control. 5G may have multiple radio interfaces, namely sub-6 GHz, cmWave, and mmWave, and may also be integrated with existing legacy radio access technologies (such as LTE). Integration with LTE may be possible, at least in the early stages, as a system, where macro coverage may be provided by LTE and 5G radio interface access may come from small cells through aggregation to LTE. In other words, 5G may support inter-RAT operability (such as LTE-5G) and inter-RI operability (inter-radio interface operability, such as sub-6 GHz-cmWave-mmWave). One of the concepts being considered for 5G networks may be network slicing, in which multiple independent and dedicated virtual subnetworks (network instances) can be created within essentially the same infrastructure to run services with different requirements for latency, reliability, throughput, and mobility.

[0058] Current architectures in LTE networks may be fully distributed in the radio and fully centralized in the core network. Low-latency applications and services in 5G may require content to be placed close to the radio, leading to local bursting and multi-access edge computing (MEC). 5G may enable analytics and knowledge generation to occur at the source of data. This approach may require leveraging resources that may not be continuously connected to the network, such as laptops, smartphones, tablets, and sensors. MEC may provide a distributed computing environment for hosting applications and services. It may also include the ability to store and process content close to cellular users for faster response times. Edge computing may encompass a wide range of technologies, such as wireless sensor networks, mobile data collection, mobile signature analysis, cooperative distributed peer-to-peer ad hoc networks, and processing. It can also be categorized as local cloud / fog computing and grid / grid computing, dew computing, mobile edge computing, cloud computing, distributed data storage and retrieval, autonomous self-healing networks, remote cloud services, augmented and virtual reality, data caching, the Internet of Things (with massive connectivity and / or latency-criticality), and critical communications (autonomous vehicles, traffic safety, real-time analytics, time-critical control, and healthcare applications).

[0059] The communication system may also be capable of communicating with one or more other networks 113, such as the public switched telephone network or the Internet, or utilizing services provided by them. The communication network may also be capable of supporting the use of cloud services, for example, at least part of the core network operations may be performed as one or more cloud services (this is in the context of Figure 1 (depicted by "cloud" 114 in the diagram). There may also be multiple clouds, and clouds may be physically distributed in both virtual and physical forms. The communication system may also include a central control entity, etc., providing facilities for the networks of different operators to collaborate, for example, in terms of spectrum sharing.

[0060] The access node can also be divided into: a radio unit (RU), including a radio transceiver (TRX), i.e., a transmitter (Tx) and a receiver (Rx); one or more distributed units (DU) 105, which can be used for so-called layer 1 (L1) processing and real-time layer 2 (L2) processing; and a central unit (CU) 108 (also called a centralized unit), which can be used for non-real-time layer 3 (L2) processing. The CU can be connected to one or more DUs, for example, via an F1 interface. This split can achieve the centralization of the CU relative to the cell site and the DU, while the DU can be more distributed or even remain at the cell site. The CU and DU together can also be called a baseband or baseband unit (BBU). The CU and DU can also be included in a radio access point (RAP).

[0061] CU108 can be defined as a logical node that carries the higher layer protocols of the access node, such as radio resource control (RRC), service data adaptation protocol (SDAP) and / or packet data convergence protocol (PDCP). DU 105 can be defined as a logical node that carries the radio link control (RLC), medium access rights transformation (MAC) and / or physical (PHY) layers of the access node. The operation of the DU can be at least partially controlled by the CU. The CU can include a control plane (CU-CP), which can be defined as a logical node that carries RRC and the control plane part of the PDCP protocol of the CU for the access node. The CU can also include a user plane (CU-UP), which can be defined as a logical node that carries the user plane part of the PDCP protocol and the SDAP protocol for the CU access node.

[0062] The cloud computing platform can also be used to run the CU and / or DU. The CU can run in the cloud computing platform, which can be called a virtualized CU (vCU). In addition to the vCU, there can also be a virtualized DU (vDU) running in the cloud computing platform. In addition, there can also be a combination in which the DU can use a so-called bare metal solution, such as an application-specific integrated circuit (ASIC) or a customer-specific standard product (CSSP) system-on-chip (SoC) solution. It should also be understood that the functional distribution between the above-mentioned access node units, or between different core network operations and access node operations may be different.

[0063] By leveraging Network Function Virtualization (NFV) and Software Defined Networking (SDN), edge cloud can be introduced into the Radio Access Network (RAN). Using edge cloud can mean performing access node operations at least partially in a server, host, or node that is operatively coupled to a Remote Radio Head (RRH) or Radio Unit (RU), or performing access node operations in an access node that includes the radio portion. Node operations can also be distributed across multiple servers, nodes, or hosts. The Cloud RAN architecture enables RAN real-time functions to be performed on the RAN side (in distributed units, DU 105), while non-real-time functions can be performed in a centralized manner (in centralized units, CU1 08).

[0064] It should also be understood that the functional division between core network operations and access node operations may differ from LTE or even not exist. Some other technological advancements that may be used include big data and all-IP, which may change the way networks are built and managed. 5G (or new radio, NR) networks can be designed to support multiple hierarchical structures, where MEC servers can be placed between the core and access nodes. It should be understood that MEC can also be applied to 4G networks.

[0065] 5G can also leverage non-terrestrial communications, such as satellite communications, to enhance or supplement the coverage of 5G services, for example by providing backhaul. Possible use cases could be providing service continuity for machine-to-machine (M2M) or Internet of Things (IoT) devices or passengers on vehicles, or ensuring service availability for critical communications and future orbital / maritime / aerospace communications. Satellite communications can leverage geostationary (GEO) satellite systems as well as low Earth orbit (LEO) satellite systems, in particular mega-constellations (systems with hundreds of (nano)satellites deployed). A given satellite 106 in a mega-constellation can cover several satellite-enabled network entities creating a terrestrial cell. The terrestrial cell can be accessed via a relay node on the ground or an access node 104 located on the ground or in a satellite.

[0066] It is obvious to those skilled in the art that the depicted system is only an example of a part of a radio authority system, and in practice, the system may include multiple access nodes, the user equipment may have access to multiple radio cells, and the system may also include other devices, such as physical layer relay nodes or other network elements, etc., and at least one of the access nodes may be a Home eNodeB or a Home gNodeB.

[0067] Furthermore, in a geographical area of ​​a radio communication system, a plurality of different kinds of radio cells as well as a plurality of radio cells may be provided. A radio cell may be a macro cell (or umbrella cell), which may be a large cell with a diameter of up to tens of kilometres, or a smaller cell such as a micro cell, a femto cell or a pico cell. Figure 1 An access node can provide any of these cells. A cellular radio system can be implemented as a multi-layer network comprising several types of radio cells. In a multi-layer network, one access node can provide one or more types of radio cells, so multiple access nodes may be required to provide such a network structure.

[0068] In order to meet the need to improve the deployment and performance of communication systems, the concept of "plug and play" access nodes can be introduced. Networks that may be able to use "plug and play" access nodes may include, in addition to Home eNodeBs or Home gNodeBs, Home NodeB Gateways or HNB-GWs ( Figure 1 (not shown). The HNB-GW, which may be installed in an operator's network, may aggregate traffic from a large number of Home eNodeBs or Home gNodeBs back to the core network.

[0069] A user equipment can perform a random access procedure to access the network. The random access procedure may also be referred to as an initial access procedure. The purpose of performing the random access procedure may be, for example, initial access, handover, scheduling request, or timing synchronization. There are currently two types of random access procedures: contention-based random access (CBRA) and contention-free random access (CFRA). CFRA may also be referred to as non-contention-based random access. In CFRA, a given user equipment has a dedicated (i.e., UE-specific) random access preamble assigned by the gNB, while in CBRA, the user equipment randomly selects a preamble from a pool of preambles shared with other user equipment in the cell. In CBRA, if two or more user equipment attempt a random access procedure by using the same random access procedure on the same resource, the user equipment will not be able to access the network.

[0070] 5GNR supports two different CBRA procedures: four-step RACH and two-step RACH. RACH is the abbreviation for Random Access Channel. The four-step RACH procedure is described below.

[0071] Before preamble transmission, there may be a preliminary step of sending and receiving synchronization signal blocks (SSBs), known as DL beam scanning, which may not be a formal part of the random access procedure. As a result of this preliminary step, the user equipment can select the index of the preferred SSB beam and decode the relevant physical broadcast channel (PBCH) from the master information block (MIB), system information block (SIB), etc. The user equipment can also use this index to identify the appropriate RACH occasion for the preamble transmission, based on the SSB-to-RACH-occasion mapping indicated by SIB 1. The system information provided in the SSB may include the information required to create the same preamble set configured by the gNB.

[0072] The four-step RACH procedure begins with the user equipment sending a random access preamble to the gNB via the Physical Random Access Channel (PRACH) using a dedicated radio resource called a RACH instance. The message carrying the random access preamble is referred to as Message 1 (Msg 1). The user equipment can uniquely identify a set of possible preambles using information in the system information broadcast by the gNB. The user equipment randomly selects a random access preamble from this set.

[0073] The gNB decodes Msg 1 received from the user equipment and extracts its preamble. The gNB replies to the user equipment with a Random Access Response (RAR). The message carrying the Random Access Response is referred to as Message 2 (Msg2). Msg2 may include the detected preamble identifier, the Timing Advance Command, the Temporary Cell Radio Network Temporary Identifier (TC-RNTI), and an UL grant for transmitting Message 3 (Msg3) on the Physical Uplink Shared Channel (PUSCH). In other words, in Msg2, the gNB schedules resources for the user equipment to transmit Msg3 via the PUSCH.

[0074] The UE responds to Msg2 by scheduling PUSCH by sending Msg3 to the gNB using the corresponding timing advance information on its uplink beam. Msg3 may include an RRC connection request and an identifier of the UE.

[0075] The gNB responds to Msg3 by sending Message 4 (Msg4) to the UE, including RRC setup information and the UE identifier extracted from Msg3. This completes the random access procedure.

[0076] The two-step RACH procedure is otherwise similar to the four-step RACH procedure described above, but Msg1 and Msg3 are combined into a single message (denoted as MsgA) and sent by the UE without waiting for intermediate feedback (e.g., Msg2). Similarly, the gNB combines Msg2 and Msg4 into a single message (denoted as MsgB).

[0077] As described above, during the random access procedure, a user equipment can select a random access preamble from a predefined set called a "preamble set," which can include up to 64 preambles. The user equipment then transmits the selected preamble to the gNB. This preamble set is also referred to as a PRACH preamble set. The gNB can create a specific preamble set by broadcasting minimal system information within its coverage area (e.g., via SSB). The minimal system information can include the information required to create the same preamble set configured by the gNB. A given user equipment receiving this message can create the same preamble set configured by the gNB. By selecting a random preamble set from this set, the risk of collisions—that is, the risk of two user equipment selecting the same preamble—is reduced. The selected preamble can be used as a temporary identifier for the user equipment and to extract timing advance information.

[0078] The leader set can be composed of the first logical root sequence to be used in the leader set and a sequence called N CS The cyclic shift parameter value of N is uniquely determined. The continuous root sequences to be used starting from the first root sequence can be uniquely defined in the standard. CSThe value can also be uniquely defined in the standard for various sequence lengths. Once the first root sequence index and cyclic shift are known, all 64 preambles in the preamble set can be uniquely determined. As an example, for the first root sequence index 0 and N CS =2, the physical root sequences 1 and 138 can be used in the preamble set by cyclically shifting each sequence by 2.

[0079] If two adjacent gNBs use preamble sets with overlapping preambles, their uplink transmissions may interfere with each other if a user equipment transmits them. This can cause ambiguity when a given gNB determines which user equipment wants to connect to it. Therefore, it is desirable for gNBs to use orthogonal preamble sets whenever possible.

[0080] Operators can configure the leader set for each cell. To avoid non-orthogonal leader sets, they should carefully select the leader set configuration for each cell and distribute the root sequence among them. This fixed allocation scheme needs to be repeated whenever a new cell is added or a cell is reconfigured. In the standard, a finite number of leader sets are available, allowing the leader set to be identified using only two variables. This rigid leader set constraint can lead to suboptimal and frequent root sequence assignments, which is undesirable from the operator's perspective.

[0081] 6G networks are expected to utilize flexible decentralized and / or distributed computing systems and architectures, as well as ubiquitous computing. Based on mobile edge computing, artificial intelligence, short packet communications, distributed ledgers, and blockchain technologies, they will enable local spectrum licensing, spectrum sharing, infrastructure sharing, and intelligent automated management. Key features of 6G will include intelligent connectivity management and control capabilities, programmability, integrated sensing and communications, a reduced energy footprint, trustworthy infrastructure, scalability, and affordability. Furthermore, 6G targets new use cases, including integrating localization and sensing capabilities into system definitions to unify the user experience across the physical and digital worlds.

[0082] As described above, a given preamble set can be uniquely identified by using a logical root sequence identifier and a parameter indicating a cyclic shift. This enables the preamble set to be easily transmitted to the user equipment. However, this may not be the most efficient way to create a preamble set. Due to the expected large number of user equipment, sub-cells and data rate support expectations, the sets enabled in the current standard may not be optimal for the needs of next-generation networks (such as 6G). In this case, the problems that may arise from rigid preamble sets may include insufficient PRACH capacity, interference and range limitations. These problems come from the traditional approach, which introduces a limited number of preamble sets and may not be able to handle the next generation requirements. One way to solve the above problems is to multiply the number of available preamble sets. Such heterogeneous preamble sets will solve the problems of insufficient PRACH capacity, interference and range limitations.

[0083] For example, to transmit a preamble set, the gNB may need to transmit a physical root sequence of the first root sequence, e.g., 0 to 137, and a cyclic shift parameter NCS, e.g., varying from 0 to 15. There are only 2208 different preamble sets that can be generated with these constraints (138*16=2208). CS If both the NCS and the first root sequence are transmitted to the UE, the gNB may need to send 12 bits. In addition, the UE may need to store a table to map the NCS and the first root sequence to a specific preamble set that uses consecutive root sequences.

[0084] To address the above limitations caused by current PRACH solutions, a preamble set can be synthesized where each preamble has an arbitrary root sequence and cyclic shift. To indicate a preamble set consisting of 64 preambles, each with an arbitrary root sequence and cyclic shift, we need to use bits, given the current PRACH configuration. 1024 bits may be too large to be included in Msg1. The overhead increases linearly with the number of preambles in the preamble set. The overhead can be much larger for the long format, where the sequence length is 838. This may be a limitation in future 6G systems, where we envision many IoT devices attempting to establish connections and gNBs with very many subcells.

[0085] Some exemplary embodiments may provide such scalability and address complexity issues. Some exemplary embodiments may enable the transmission of an arbitrary set of preambles to a user device, where a given preamble may have an arbitrary root sequence and cyclic shift. Such flexibility may be beneficial, for example, for 6G and beyond, where support for a large number of IoT devices and mobile access points may be required. In addition, increased quality requirements may be required, such as minimizing PRACH capacity shortages, interference, and range limitations. Other examples include, but are not limited to, remotely controlled vehicles, such as drones, and a large number of connected augmented reality (AR) devices.

[0086] Some example embodiments may enable the following operations: communicating with a user equipment using an arbitrary preamble set, compressing the arbitrary preamble set, and causing the user equipment to create a preamble set suitable for transmission in Msg1.

[0087] Figure 2 A signaling diagram according to an example embodiment is shown. As described above, a given random access preamble in a preamble set can be uniquely identified by a physical root sequence index and a cyclic shift. In this example embodiment, for example, a preamble set can be identified as the non-zero entries of a two-dimensional preamble matrix (denoted as M), where the rows indicate the cyclic shift (N CS), with the columns indicating the root sequence (or vice versa). In other words, a given non-zero entry in the matrix indicates a unique random access preamble in the preamble set. Alternatively, any other value or symbol can be used instead of a non-zero entry to indicate a random access preamble. If the gNB has multiple subcells, each subcell uses an orthogonal preamble set, and the same preamble matrix can be used to represent all preamble sets used by all subcells of the gNB.

[0088] Reference Figure 2 In block 201, a network element of a radio access network (e.g., a gNB or a radio unit) generates a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble matrix. For example, the preamble matrix may be randomly generated or based on a study or based on the number of neighboring gNBs or UEs.

[0089] The plurality of entries may include at least two sets of values ​​or symbols, and the subset of entries (indicating a random access preamble) may include a set of values ​​or symbols from the at least two sets of values ​​or symbols. For example, the plurality of entries may include binary entries (1 and 0), and a non-zero entry in the preamble matrix, i.e., a value of 1, may indicate a random access preamble, in which case a zero entry, i.e., a value of 0, does not indicate a random access preamble. Alternatively, a zero entry in the preamble matrix, i.e., a value of 0, may indicate a random access preamble, in which case a non-zero entry does not indicate a random access preamble. Alternatively, instead of 1 and 0, the preamble matrix may include any two sets of symbols (e.g., A and B, Y and Z, 1 and 5), one of which identifies a root sequence and cyclic shift for a given random access preamble and the other does not.

[0090] For example, the rows of the preamble matrix may indicate the cyclic shift of the random access preamble, and the columns of the preamble matrix may indicate the root sequence of the random access preamble. Alternatively, the columns of the preamble matrix may indicate the cyclic shift of the random access preamble, and the rows of the preamble matrix may indicate the root sequence of the random access preamble.

[0091] To generate the preamble matrix, the network element may select a parameter value, denoted NR, to indicate the maximum number of physical root sequences to be generated, as specified by the vendor and / or standard. As an example, the current NR standard has 138 such logical root sequences, ranging from 0...137, in which case there may be 138 columns in the preamble matrix (if columns are used to indicate root sequences). However, the number of root sequences may also be higher or lower than 138.

[0092] In addition, the network element can be represented as N CS The parameter value represents the number of possible cyclic shifts, for example, as specified by the vendor and / or standard.

[0093] Then based on N R and NCS The parameter value of the leading matrix is ​​generated as an N-dimensional matrix (for example, N R xN CS or N CS xN R or 1xN R N CS or N R N CS x1). For example, in an NCS xNR preamble matrix, the number of rows in the preamble matrix may be equal to the parameter value NCS, and the number of columns in the preamble matrix may be equal to the parameter value N R For example, in N R xN CS In the leading matrix, the number of rows in the leading matrix can be equal to the parameter value N R , the number of columns in the leading matrix can be equal to the parameter value N CS .

[0094] Table 1 below presents a non-limiting example of a leader matrix. However, it should be noted that the leader matrix may also be different from that shown in Table 1 (for example, the number of rows and columns may be different, and the entries may be different). The binary leader matrix includes multiple entries with a value of 0 or 1. In this example, a given non-zero entry, i.e., an entry with a value of 1, represents a unique random access leader in the leader set. In other words, the row and column of a given non-zero entry uniquely identify a random access leader. For example, the non-zero entry at row i and column j represents the use of sequence j after being cyclically shifted by i elements. The number of non-zero entries in the leader matrix may be equal to the number of random access leaders in the leader set.

[0095]

[0096] Table 1

[0097] In block 202, a network element is configured to utilize a preamble matrix as a preamble set for detecting a random access preamble, e.g., indicated by a non-zero or zero entry in the preamble matrix. For example, the configuration may be performed by the control plane. Alternatively, the configuration may be performed by the network element (e.g., gNB) during a software setup procedure or (re)boot, etc.

[0098] In block 203, the network element encodes the leading matrix M into a sequence. The sequence can be denoted as s. For example, the network element can encode the leading matrix M into a binary sequence s of length L using an artificial neural network, a hash, or any other function, the purpose of which is to produce a smaller alternative representation of the leading matrix M. The sequence can be reduced to bits. The length L can refer to the number of bits in the sequence s. The sequence s can represent a compressed version of the leading matrix M. For example, the sequence s can be a hash or a binary vector. Alternatively, there can be no compression, or the sequence can even be expanded so that the output of the encoder is larger than its input.

[0099] In block 204, the network element sends the sequence s to one or more user devices. For example, the network element may broadcast the sequence s in a system information block (SIB) packet. The one or more user devices receive the sequence s from the network element. The sequence s may be used to indicate to the one or more user devices which preamble set is being used at the network element.

[0100] In block 205, one or more user devices decode the sequence s. For example, the one or more user devices may pass the sequence s through a decoder that outputs a preamble matrix or a specific root sequence and a cyclic shift.

[0101] In block 206, if the decoder outputs a preamble matrix, the one or more user devices may select an entry from at least a subset of entries included in the decoded preamble matrix, wherein the at least subset of entries refers to an entry indicating a random access preamble. For example, the one or more user devices may randomly select an entry.

[0102] In block 207, one or more user devices generate a root sequence and cyclic shift based at least in part on the decoding. For example, the root sequence and cyclic shift may be generated based on the row and column of the selected entry in the decoded preamble matrix. Alternatively, the decoder may output the root sequence and cyclic shift directly from the sequence s.

[0103] In block 208, one or more user equipments generate a random access preamble based on the generated root sequence and cyclic shift. The random access preamble may also be referred to as a PRACH preamble or a RACH preamble.

[0104] In block 209, one or more user devices may initiate a random access procedure by sending a generated random access preamble to the network element. The random access preamble may be applied to any random access procedure. For example, the random access preamble may be sent in Msg1 (in a four-step random access procedure) or MsgA (in a two-step random access procedure). The network element receives the random access preamble from one or more user devices.

[0105] In block 210, the network element determines which of the random access preambles in the preamble set has been sent by one or more user equipments.The network element may send a random access response or MsgB to the one or more user equipments in response to receiving the random access preamble.

[0106] Figure 3 A flow chart illustrating a method performed by a network element such as a radio access network or a device included therein according to an example embodiment is shown. For example, the network element may correspond to Figure 1 The access node 104 or wireless unit.

[0107] Reference Figure 3, in block 301, a preamble matrix comprising a plurality of entries is generated, wherein at least a subset of the plurality of entries indicates a random access preamble.

[0108] The rows of the preamble matrix may indicate the cyclic shift of the random access preamble, and the columns of the preamble matrix may indicate the root sequence of the random access preamble. Alternatively, the rows may indicate the root sequence, and the columns may indicate the cyclic shift.

[0109] The plurality of entries may include at least two sets of values ​​or symbols, and the subset may include a set of values ​​or symbols from the at least two sets of values ​​or symbols.

[0110] In block 302, a sequence encoded with a preamble matrix is ​​obtained. For example, the device itself may encode the preamble matrix into the sequence, or the encoding may be performed by another entity (e.g., in the cloud).

[0111] In block 303, the sequence is sent to one or more user equipments.

[0112] Figure 4 A flow chart of a method performed by an apparatus such as a user equipment, or included in a user equipment, according to an example embodiment is shown. A user equipment may also be referred to as a subscriber unit, a mobile station, a remote terminal, an access terminal, a user terminal, a terminal device, or a user equipment (UE). A user equipment may correspond to Figure 1 One of the user devices 100, 102.

[0113] Reference Figure 4 In block 401 , a sequence is received from a network element of a radio access network, wherein the sequence is encoded using a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble.

[0114] The rows of the preamble matrix may indicate the cyclic shift of the random access preamble, and the columns of the preamble matrix may indicate the root sequence of the random access preamble. Alternatively, the rows may indicate the root sequence, and the columns may indicate the cyclic shift.

[0115] The plurality of entries may include at least two sets of values ​​or symbols, and the subset may include a set of values ​​or symbols from the at least two sets of values ​​or symbols.

[0116] In block 402, a sequence is decoded.

[0117] In block 403, a root sequence and a cyclic shift are generated based at least in part on the decoding.

[0118] In block 404, a random access preamble is generated based on the root sequence and the cyclic shift.

[0119] In block 405, a random access preamble is sent to a network element.

[0120] The above-mentioned Figure 2-Figure 4 There is no absolute chronological order for the described blocks, related functions, and information exchanges (messages), and some of them may be executed simultaneously or in a different order than described. Other functions may also be executed between or within them, and other information may be sent and / or other rules may apply. Some blocks or parts of blocks or one or more pieces of information may also be omitted or replaced with corresponding blocks or parts of blocks or one or more pieces of information.

[0121] Figure 5 shows how a network element is configured to utilize the leading matrix as Figure 2 An example of a leader set in block 202 of FIG. Figure 5 , the control plane 501 configures the PRACH receiver to detect the random access preamble described by the preamble matrix M. A network element (e.g., gNB) or an encoder in a higher layer or the cloud can map the preamble matrix into a sequence s. In other words, the sequence s does not necessarily need to be calculated in the gNB or radio unit, as it can alternatively be calculated in the cloud, Layer 3, or Layer 2 software, for example. For preamble set initialization, the layer can signal the sequence s and the preamble matrix M to the gNB.

[0122] Figure 6 Shown in Figure 2 Box 203 or Figure 3 An example of encoding the leading matrix M into the sequence s at the network element in block 302 of FIG. Figure 6 , the preamble matrix M is provided as input to encoder 601. A sequence s is then received as output from encoder 601. For example, encoder 601 may run in a network element (e.g., a gNB) or a cloud server. If the encoder runs in the cloud, the gNB may store the associated sequence s indicating the preamble set. Encoder 601 may compress the preamble matrix M into a binary sequence s. Encoder 601 may include an artificial neural network, hashing, or any other function whose purpose is to produce a smaller, alternative representation of the preamble matrix M.

[0123] 7A shows the Figure 2 Box 205 or Figure 4In block 402 of FIGURE 4, a first decoder option is provided for decoding a leading matrix M from a sequence s at a user device. In the first decoder option, an artificial intelligence (AI) or machine learning (ML) model 701 at the user device maps a hash or sequence s to a leading matrix M. In other words, the sequence s may be provided as input to the AI / ML model 701, which may output the leading matrix M. For example, the AI / ML model 701 may include at least one convolutional neural network (CNN) layer, at least one fully connected (FC) layer, or at least one recurrent neural network (RNN) layer. Using the leading matrix M as input, the user device may then select non-zero entries (e.g., randomly) from the leading matrix and generate leading root sequences and cyclic shifts based on the rows and columns of the selected non-zero entries in the leading matrix M. Alternatively, the decoder may select non-zero entries from the leading matrix decoded by the AI / ML model 701. The user device may not need to store any leading matrix M. Its decoder function decodes the sequence s into the leading matrix M.

[0124] Figure 7B Shown for Figure 2 Box 205 or Figure 4 A second decoder option is provided for decoding the preamble matrix M from the user equipment in block 402 of FIG. In the second decoder option, the network provider can provide a decoder solution 702 that generates a random preamble identifier (root sequence, cyclic shift) directly from the sequence s. In other words, the sequence s can be provided as input to the decoder 702, which can then output the root sequence and cyclic shift for use by the user equipment to generate the corresponding random access preamble.

[0125] Figure 8 Shown in Figure 2 Box 207 or Figure 4 Example of generating a random access preamble at a user equipment in block 404 of FIG. The user equipment creates a random access preamble using the selected root sequence and cyclic shift. Figure 8 , the physical root sequence ID and cyclic shift, or the identifier of the preamble matrix M, can be provided as input to the preamble generator function 801. The preamble generator function 801 can then output a random access preamble. This function can be replaced by an AI / ML model and integrated into the decoder function above.

[0126] For example, suppose a (N R xN CS )=(138x 138) preamble matrix to describe all possible preamble sets, whose length is 138 root sequences 139. The non-zero entries in the preamble matrix can uniquely identify the preamble used in the preamble set. 138x1There are as many as 19044 preambles in a given preamble set. Therefore, it may not be feasible to store all possible preamble sets on the UE and transmit the used preamble set to the UE. If the UE cannot store preamble sets, the gNB may need to send a 16-bit Therefore, a preamble set with 64 preambles requires a total of 16x64≈1024 bits. For a SIB data packet used for PRACH, the size may be large.

[0127] As a non-limiting example, in some exemplary embodiments, the 138x1 The 4000 possible preamble sets are selected, with 60 to 100 preambles in a given preamble set. The selected 138x138 preamble matrix M can then be encoded into a sequence s of only 128 bits. The user device can then decode the sequence s into the preamble matrix M and generate a random access preamble.

[0128] Some example embodiments may be used with any PRACH receiver, as long as a version of the preamble matrix described by M can be created that describes the preamble support. Therefore, some example embodiments may be deployed on any gNB for any set of preambles.

[0129] Furthermore, some example embodiments may be used to indicate any sequence based on a lookup table to a user equipment.As an example, the same approach may be used for uplink / downlink mode of dynamic time division duplexing (TDD).

[0130] Some exemplary embodiments may also be backward compatible with legacy PRACH preamble sets. To transmit a preamble set, the gNB may transmit a physical root sequence for the first root sequence, which may range, for example, from 0 to 137, and a cyclic shift parameter (NCS), which may range, for example, from 0 to 15. There may be 2208 (138*16=2208) different preamble sets that can be generated using these parameters. To transmit the cyclic shift parameter NCS and the first root sequence, the gNB may need to send 12 bits. In some exemplary embodiments, the preamble matrices M corresponding to the 2208 preamble sets may be encoded using a binary sequence of 8 bits per preamble matrix. This reduces the number of bits required to transmit the preamble sequence by 33%.

[0131] Figure 9A flowchart of a method for training an artificial neural network, such as an autoencoder, according to an example embodiment is shown, including the cascade of encoder 601 and decoders 701 and 702 described above. That is, the encoder and decoder can be trained together, but they can be independent neural networks. The trained encoder can run on the gNB or in the cloud, while the trained decoder can run on the user device. The autoencoder may only need to be trained once and can be used across many sites and user devices. Training can be performed on any computing device.

[0132] Reference Figure 9 In block 901, the number of different preamble sets to be used in training is selected. For example, the number of preamble sets may be selected based on user input indicating the number of preamble sets. Alternatively, the number of preamble sets may be adaptively selected by a computer program, for example, based on the number of neighboring gNBs or UEs. The selected number of preamble sets is denoted herein as NM.

[0133] As a non-limiting example, the number of preamble sets may be selected to be 4000. However, the number of preamble sets may also be higher or lower than 4000.

[0134] In block 902, a first matrix (denoted as S) is created, wherein the first matrix includes a number of random access preamble sets equal to the number of selected preamble sets. For example, the number of rows in the first matrix may be equal to the number of preamble sets, and the number of columns in the first matrix may be equal to the number of root sequences multiplied by the number of cyclic shifts. In other words, the shape of the first matrix S may be [N M , N R *N CS ]. The first matrix may include binary entries, i.e., values ​​of 0 or 1. Alternatively, any two symbols may be used instead of 1 and 0.

[0135] In block 903 , the training batch size is set equal to a value denoted as B. For example, the training batch size may be set based on user input indicating the value B. Alternatively, the training batch size may be set by a computer program.

[0136] In block 904, an encoder is created having at least one neural network layer. For example, the at least one neural network layer of the encoder may include at least one fully connected (FC) layer and / or at least one convolutional neural network (CNN) layer.

[0137] In block 905, a decoder having at least one neural network layer is created. For example, the at least one neural network layer of the decoder may include at least one FC layer and / or at least one CNN layer.

[0138] In block 906, a sample batch (denoted as X) is selected from the first matrix S, where the batch size of the sample batch is equal to the training batch size B. For example, the sample batch can be randomly selected. The sample batch can be a flattened version of the leading matrix M, which the encoder learns to encode and the decoder learns to decode based on the encoder output.

[0139] In block 907 , a batch X of samples is provided as input to the encoder.

[0140] In block 908, the sequence (denoted as Y EN ) as the output from the encoder, where the sequence is encoded using the sample batch. This sequence can represent a compressed or uncompressed version of the sample batch X. For example, the encoder can compress the matrix (sample batch) into a binary sequence of length 128 bits. In this case, the shape of the compressed sequence can be [1, 128].

[0141] In block 909, sequence Y EN is provided as input to the decoder.

[0142] In block 910, a second matrix (denoted as Y DE ) as the output from the decoder, the second matrix Y DE The shape can be, for example, [N M , N R *N CS ], that is, the number of rows in the second matrix can be equal to the number of preamble code sets, and the number of columns in the second matrix can be equal to the number of root sequences multiplied by the number of cyclic shifts.

[0143] In block 911, a binary cross entropy loss is determined. The binary cross entropy can be used as a loss function to compare the sample batch and the second matrix to measure the error in reconstructing the sample batch from the compressed sequence. The binary cross entropy loss is denoted herein as L BCE (Y DE , Y true ). Y true Indicates whether the reconstructed matrix matches the original matrix (batch of samples). The binary cross entropy loss is propagated to the decoder and encoder, for example, through the ADAM optimizer or any other optimizer.

[0144] In block 912, the binary cross entropy loss is compared to a threshold value denoted as e. For example, the threshold value e may be less than or equal to 1e ∧ -5 (i.e., e≤1e ∧ -5).

[0145] If the binary cross entropy loss is not lower than the threshold (block 912: No), blocks 906-912 are repeated until the binary cross entropy loss is lower than the threshold. In other words, the apparatus may repeatedly select a sample batch, input a sample batch, receive a compressed sequence, input a compressed sequence, receive a second matrix, determine, and propagate until the binary cross entropy loss is lower than the threshold.

[0146] In this way, the artificial neural network can be trained to minimize the reconstruction error, for example by performing gradient descent on the loss function via an ADAM optimizer.

[0147] In block 913, if the binary cross entropy loss is below a threshold, training is terminated.

[0148] Figure 10 The diagram shows Figure 9 A flowchart of an example embodiment of a method for training inference using an artificial neural network.

[0149] See also Figure 10 , in block 1001, the leading matrix M is flattened to obtain the resulting first leading matrix X, which has a shape of [1, N R *N CS ]. Flattening can be used to reduce the number of parameters, kernel size, etc.

[0150] In block 1002, a first leading matrix X is fed as input to an encoder having at least one neural network layer. For example, the at least one neural network layer of the encoder may include at least one FC layer and / or at least one CNN layer.

[0151] In block 1003, a compressed sequence Y of a first preamble matrix X is received. EN As the output of the encoder, the shape of the compressed sequence can be, for example, [1, 128].

[0152] In block 1004, the compressed sequence Y EN It is fed as input to a decoder which has at least one neural network layer.

[0153] In block 1005, the second leading matrix Y DE is received as output from the decoder, where the second leading matrix Y DE The shape is [1, N R *N CS ].

[0154] In block 1006, the second leading matrix Y DE Select the pair of root sequence and cyclic shift (n R , n CS ), where n R <N R , and n CS <NCS , for example, the pair may be selected randomly.

[0155] In block 1007, the selected pair (n R , n CS ) is provided as an output from an artificial neural network.

[0156] Alternatively, a tensor can be created as input and a tensor can be created as output. In this case, the options can be: N: batch size, C: channels, H: height, W: width -> (N, C, H, W) -> (1, 1, N R , N CS ).

[0157] Figure 11 An example scenario is described in which some example embodiments may be used to dynamically allocate root sequences between gNBs. Some example embodiments may enable this allocation to be dynamically changed between different gNBs and between different preamble sets on the same gNB. Some example embodiments may also enable the use of arbitrary preamble set sizes, where preamble sets can have up to 64 preamble sets, compared to current networks. Some example embodiments may address the issue of limited, less-orthogonal PRACH sequences by allowing preamble sets of arbitrary size and with arbitrary cyclic shifts.

[0158] Reference Figure 11 The control plane 1100 can determine the preamble matrix for gNBs 1101, 1102, and 1103. The gNBs can have the ability to identify UE density and collisions. For example, the compressed sequence S for a given preamble matrix M can be calculated in the cloud, Layer 3, or Layer 2 software.

[0159] exist Figure 11 In this example, the first gNB (gNB1) 1101 can be located in a sparse area, so the preamble matrix M for gNB1 can be very sparse and selectively changed. In other words, if there are fewer UEs, the number of preambles can be lower because the probability of collision (i.e., two UEs selecting the same preamble) is lower in this case.

[0160] The second gNB (gNB2) 1102 may be located in a densely populated area, so the preamble matrix M for gNB2 may change more frequently and be less sparse compared to the preamble matrix for gNB1.

[0161] The third gNB (gNB3) 1103 may be located in the problematic cell acquisition area, so the leading matrix M of gNB3 may be changed based on the problem resolution metric (e.g., the sum of collisions within a 10-minute period).

[0162] Figure 12An example of an apparatus 1200 is shown, which includes a method for performing Figure 4 The method or apparatus of any other example embodiment described above. For example, apparatus 1200 may be an apparatus such as a user device, or may be included in a user device, or may be included in a user device. The user device may correspond to Figure 1 100, 102. A user device may also be referred to as a subscriber unit, a mobile station, a remote terminal, an access terminal, a user terminal, a terminal device, or a user equipment (UE).

[0163] The apparatus 1200 includes at least one processor 1210. The at least one processor 1210 interprets computer program instructions and processes data. The at least one processor 1210 may include one or more programmable processors. The at least one processor 1210 may include programmable hardware with embedded firmware, and may alternatively or additionally include one or more application-specific integrated circuits (ASICs).

[0164] The at least one processor 1210 is coupled to at least one memory 1220. The at least one processor is configured to read data from and write data to the at least one memory 1220. The at least one memory 1220 may include one or more memory cells. The memory cells may be volatile or non-volatile. It should be noted that there may be one or more non-volatile memory cells and one or more volatile memory cells, or alternatively, one or more non-volatile memory cells. Volatile memory may be, for example, random access memory (RAM), dynamic random access memory (DRAM), or synchronous dynamic random access memory (SDRAM). Non-volatile memory may be, for example, read-only memory (ROM), programmable read-only memory (PROM), electronically erasable programmable read-only memory (EEPROM), flash memory, optical storage, or magnetic storage. In general, memory may be referred to as non-transitory computer-readable media. The term "non-transitory" as used herein is a limitation on the medium itself (i.e., tangible, not a signal), rather than a limitation on the persistence of data storage (e.g., RAM versus ROM). At least one memory 1220 stores computer-readable instructions that are executed by at least one processor 1210 to perform one or more of the above-described exemplary embodiments. For example, non-volatile memory stores computer-readable instructions, and at least one processor 1210 uses volatile memory to execute instructions for temporarily storing data and / or instructions. Computer-readable instructions may refer to computer program code.

[0165] The computer-readable instructions may have been pre-stored in at least one memory 1220, or alternatively or additionally, they may be received by the device via an electromagnetic carrier signal and / or may be copied from a physical entity such as a computer program product. Execution of the computer-readable instructions by at least one processor 1210 causes the device 1200 to perform one or more of the above-described example embodiments. That is, at least one processor and at least one memory storing instructions may provide a means for providing or causing the performance of any of the methods and / or blocks described above.

[0166] In the context of this document, "memory" or "computer-readable medium" or "computer-readable medium" can be any non-transitory medium or media or device that can contain, store, communicate, propagate or transport instructions for use by or in connection with an instruction execution system, device or apparatus such as a computer. The term "non-transitory" as used herein is a limitation on the medium itself (i.e., tangible, as opposed to a signal), not on the persistence of the data storage (e.g., RAM vs. ROM).

[0167] The device 1200 may also include or be connected to an input unit 1230. The input unit 1230 may include one or more interfaces for receiving input. The one or more interfaces may include, for example, one or more temperature, motion, and / or orientation sensors, one or more cameras, one or more accelerometers, one or more microphones, one or more buttons, and / or one or more touch detection units. In addition, the input unit 1230 may include an interface to which external devices may be connected.

[0168] Device 1200 may also include an output unit 1240, which may include or be connected to one or more displays capable of presenting visual content, such as a light emitting diode (LED) display, a liquid crystal display (LCD), and / or a liquid crystal on silicon (LCoS) display. Output unit 1240 may also include one or more audio outputs, which may be, for example, speakers.

[0169] Device 1200 also includes a connection unit 1250. Connection unit 1250 enables wireless connection with one or more external devices. Connection unit 1250 includes at least one transmitter and at least one receiver, which can be integrated into device 1200 or to which device 1200 can be connected. The at least one transmitter includes at least one transmitting antenna, and the at least one receiver includes at least one receiving antenna. Connection unit 1250 can include an integrated circuit or a group of integrated circuits that provide wireless communication capabilities for device 1200. Alternatively, the wireless connection can be a hardwired application-specific integrated circuit (ASIC). Connection unit 1250 can include one or more components, such as a power amplifier, a digital front end (DFE), an analog-to-digital converter (ADC), a digital-to-analog converter (DAC), a frequency converter, a (de)modulator, and / or an encoder / decoder circuit, which are controlled by corresponding control units.

[0170] It should be noted that the device 1200 may also include Figure 12 Various components not shown in the figure may be hardware components and / or software components.

[0171] Figure 13 An example of a device 1300 is shown, which includes a device for performing Figure 3 The method or apparatus of any other exemplary embodiment described above, for example, the device 1300 may be a network element such as a wireless access network or a device included therein, for example, the network element may correspond to Figure 1 A network element may also be referred to as, for example, a network node, a radio access network (RAN) node, a next generation radio access network (NG-RAN) node, a NodeB, an eNB, a gNB, a base transceiver station (BTS), a base station, a NR base station, a 5G base station, an access node, an access point (AP), a relay node, a repeater, an integrated access and backhaul (IAB) node, an IAB donor node, a distributed unit (DU), a central unit (CU), a baseband unit (BBU), a radio unit (RU), a radio head, a remote radio head (RRH), or a transmit and receive point (TRP).

[0172] Device 1300 may include, for example, circuitry or a chipset suitable for implementing one or more of the above-described example embodiments. Device 1300 may be an electronic device including one or more electronic circuits. Device 1300 may include communication control circuitry 1310, such as at least one processor, and at least one memory 1320 storing instructions that, when executed by the at least one processor, cause device 1300 to perform one or more of the above-described example embodiments. For example, these instructions may include computer program code (software) 1322, wherein the at least one memory and computer program code (software) 1322 are configured to, together with the at least one processor, cause device 1300 to perform one or more of the above-described example embodiments. Computer program code herein may also refer to instructions that, when executed by the at least one processor, cause device 1300 to perform one or more of the above-described example embodiments. That is, the at least one processor and the at least one memory storing instructions may provide means for providing or causing the execution of any of the methods and / or blocks described above.

[0173] The processor is coupled to the memory 1320. The processor is configured to read data from and write data to the memory 1320. The memory 1320 may include one or more memory cells. The memory cells may be volatile or non-volatile. It should be noted that there may be one or more non-volatile memory cells and one or more volatile memory cells, alternatively, one or more non-volatile memory cells, alternatively, one or more volatile memory cells. Volatile memory may be, for example, random access memory (RAM), dynamic random access memory (DRAM), or synchronous dynamic random access memory (SDRAM). Non-volatile memory may be, for example, read-only memory (ROM), programmable read-only memory (PROM), electronically erasable programmable read-only memory (EEPROM), flash memory, optical storage, or magnetic storage. In general, memory may be referred to as non-memory transient computer-readable media. The term "non-transient" as used herein refers to the limitation of the medium itself (i.e., tangible, not a signal), rather than a limitation on the persistence of data storage (e.g., RAM vs. ROM). The memory 1320 stores computer-readable instructions executed by the processor. For example, a non-volatile memory stores computer-readable instructions, and the processor uses a volatile memory to execute instructions, and is used to temporarily store data and / or instructions.

[0174] The computer readable instructions may have been pre-stored in the memory 1320, or alternatively, alternatively, or additionally, they may be received by the device via an electromagnetic carrier signal and / or may be copied from a physical entity such as a computer program product. Execution of the computer readable instructions causes the device 1300 to perform one or more of the functions described above.

[0175] Memory 1320 may be implemented using any suitable data storage technology, such as semiconductor-based memory devices, flash memory, magnetic storage devices and systems, optical storage devices and systems, fixed memory, and / or removable memory. The memory may include a configuration database for storing configuration data. For example, the configuration database may store a list of current neighbor cells and, in some exemplary embodiments, a structure of frames used in detected neighbor cells.

[0176] The device 1300 may also include a communication interface 1330, which includes hardware and / or software for implementing communication connectivity according to one or more communication protocols. The communication interface 1330 includes at least one transmitter (Tx) and at least one receiver (Rx), which may be integrated into the device 1300 or the device 1300 may be connected to the transmitter (Rx). The communication interface 1330 may include one or more components, such as a power amplifier, a digital front end (DFE), an analog-to-digital converter or an analog-to-digital converter (ADC), a digital-to-analog converter (DAC), a frequency converter, a (de)modulator, and / or an encoder / decoder circuit, which are controlled by corresponding control units.

[0177] The communication interface 1330 provides the apparatus with radio communication capabilities for communicating in a cellular communication system. The communication interface may, for example, provide a radio interface for one or more user equipment. The apparatus 1300 may also include another interface toward a core network such as a network coordinator or AMF and / or an access node of the cellular communication system.

[0178] The device 1300 may further include a scheduler 1340 configured to allocate radio resources. The scheduler 1340 may be configured together with the communication control circuit 1310 or may be configured separately.

[0179] It should be noted that the device 1300 may also include Figure 13 Various components not shown in the figure may be hardware components and / or software components.

[0180] Figure 14 An example of a device 1400 is shown, which includes a device for performing Figure 9 or apparatus of any other example embodiments described above. Figure 14 It can be shown that the method is configured to perform at least the above-mentioned Figure 9 The device 1400 may correspond to a receiver or a transceiver or a subunit thereof. The device 1400 may correspond to Figure 1 Any one of the elements 100, 102, 104, 108, 112, or Figure 1A (sub)element within any one of elements 100, 102, 104, 108, 112.

[0181] Device 1400 may include, for example, circuitry or a chipset suitable for implementing one or more of the exemplary embodiments described above. Device 1400 may be an electronic device including one or more electronic circuits. Device 1400 may include training circuitry 1410, such as at least one processor, and at least one memory 1420 storing instructions that, when executed by the at least one processor, cause device 1400 to perform one or more of the exemplary embodiments described above, such as Figure 9 Such instructions may include, for example, computer program code (software) 1422, wherein the at least one memory and the computer program code (software) 1422 are configured to, together with the at least one processor, enable the device 1400 to perform one or more of the above exemplary embodiments. The computer program code herein may also refer to instructions that, when executed by the at least one processor, enable the apparatus 1400 to perform one or more of the above exemplary embodiments, that is, the at least one processor and the at least one memory storing the instructions may provide a means for providing or causing the execution of any of the above methods and / or blocks.

[0182] The processor is coupled to the memory 1420. The processor is configured to read data from and write data to the memory 1420. The memory 1420 may include one or more memory cells. The memory cells may be volatile or non-volatile. It should be noted that there may be one or more non-volatile memory cells and one or more volatile memory cells, alternatively, one or more non-volatile memory cells, alternatively, one or more volatile memory cells. Volatile memory may be, for example, random access memory (RAM), dynamic random access memory (DRAM), or synchronous dynamic random access memory (SDRAM). Non-volatile memory may be, for example, read-only memory (ROM), programmable read-only memory (PROM), electronically erasable programmable read-only memory (EEPROM), flash memory, optical storage, or magnetic storage. In general, memory may be referred to as non-memory transient computer-readable media. The term "non-transient" as used herein refers to the limitation of the medium itself (i.e., tangible, not a signal), rather than a limitation on the persistence of data storage (e.g., RAM vs. ROM). The memory 1420 stores computer-readable instructions executed by the processor. For example, a non-volatile memory stores computer-readable instructions, and the processor uses a volatile memory to execute instructions, and is used to temporarily store data and / or instructions.

[0183] The computer readable instructions may have been pre-stored in the memory 1420, or alternatively or additionally, they may be received by the device via an electromagnetic carrier signal and / or may be copied from a physical entity such as a computer program product. Execution of the computer readable instructions causes the device 1400 to perform one or more of the functions described above.

[0184] Memory 1420 may be implemented using any suitable data storage technology, such as semiconductor-based memory devices, flash memory, magnetic storage devices and systems, optical storage devices and systems, fixed memory, and / or removable memory. The memory may include a configuration database for storing configuration data. For example, the configuration database may store a list of current neighbor cells and, in some exemplary embodiments, a structure of frames used in detected neighbor cells.

[0185] The device 1400 may also include a communication interface 1430, which includes hardware and / or software for implementing communication connectivity according to one or more communication protocols. The communication interface 1430 includes at least one transmitter (Tx) and at least one receiver (Rx), which may be integrated into the device 1400 or the device 1400 may be connected to the at least one receiver (Rx). The communication interface 1430 may include one or more components, such as a power amplifier, a digital front end (DFE), an analog-to-digital converter or an analog-to-digital converter (ADC), a digital-to-analog converter (DAC), a frequency converter, a (de)modulator, and / or an encoder / decoder circuit, which are controlled by corresponding control units.

[0186] The communication interface 1430 provides a device with radio communication capabilities to communicate in a wireless communication system and enables communication with one or more access nodes, one or more user equipment (possibly via the multiple access nodes), and / or one or more other network nodes or elements. The communication interface 1430 can enable the device 1400 to transmit a fully trained artificial neural network or machine learning algorithm to another device, such as the device 1200 or the device 1300.

[0187] It should be noted that the device 1400 may also include Figure 14 Various components not shown in the figure may be hardware components and / or software components.

[0188] The term "circuitry" as used in this application may refer to one or more or all of the following: a) a hardware circuit implementation only (e.g. an implementation in analog and / or digital circuitry only); and b) a combination of hardware circuitry and software, such as (as applicable): i) a combination of analog and / or digital hardware circuitry and software / firmware; ii) any portion of a hardware processor working together with software (including a digital signal processor, software and memory) to enable a device (e.g. a mobile phone) to perform various functions; and c) hardware circuitry and / or processors, such as a microprocessor or part of a microprocessor, that requires software (e.g. firmware) to operate, but where software is not required for operation, the software may not be present.

[0189] This definition of circuitry applies to all uses of this term in this application, including in any claims. As a further example, as used in this application, the term circuitry also covers an implementation of merely a hardware circuit or processor (or multiple processors) or a portion of a hardware circuit or processor and its (or its) accompanying software and / or firmware. The term circuitry also covers, for example and if applicable to the particular claim element, a baseband integrated circuit or processor integrated circuit for a mobile device or a similar integrated circuit in a server, cellular network device, or other computing or networking device.

[0190] Figure 15 An example of an artificial neural network 1530 with one hidden layer is shown. Figure 16 An example of a computational node is shown, and the artificial neural network 1530 may correspond to the encoder 601 or the decoders 701 , 702 .

[0191] Artificial neural networks (ANNs) 1530 consist of a set of rules designed to perform tasks such as regression, classification, clustering, and pattern recognition. ANNs achieve these goals through a learning process in which they are presented with examples of various input data along with desired outputs. With this, they learn to identify the correct output for any input in the training data manifold. Learning with labels is called supervised learning, while learning without labels is called unsupervised learning. Deep learning typically requires large amounts of input data.

[0192] Deep learning (also known as deep structured learning or layered learning) is part of a broader family of machine learning methods based on layers used in artificial neural networks. A deep neural network (DNN) 1530 is an artificial neural network that includes multiple hidden layers 1502 between an input layer 1500 and an output layer 1514. Training a DNN enables it to find the correct mathematical operations to transform inputs into appropriate outputs, even when the relationships are highly nonlinear and / or complex.

[0193] Each hidden layer 1502 includes nodes 1504, 1506, 1508, 1510, 1512 where computation occurs. Figure 16 As shown, each node 1504 combines input data 1500 with a set of coefficients or weights 1600 that amplify or attenuate the input 1500, thereby assigning it importance for the task the algorithm is trying to learn. The input-weighted products are added 1602, and the sum is passed through an activation function 1604 to determine whether and to what extent the signal should further pass through the network 1530 to influence the final result, such as classification behavior. In this process, the neural network learns to identify the correlation between certain relevant features and the best results.

[0194] In the case of classification, the output of deep learning network 1530 can be considered the probability of a particular outcome, such as the probability of successful decoding of a sequence. In this case, the number of layers 1502 can vary proportionally to the amount of input data 1500 used. However, when the amount of input data 1500 is high, the accuracy of the results 1514 is more reliable. On the other hand, when there are fewer layers 1502, the calculation may take less time, thereby reducing latency. However, this is highly dependent on the specific DNN architecture and / or computing resources.

[0195] The model's initial weights 1600 can be set in a variety of alternative ways. During the training phase, they are adjusted to improve the accuracy of the process based on errors in the analyzed decisions. Training the model is essentially a trial-and-error exercise. In principle, each node 1504, 1506, 1508, 1510, 1512 of the neural network 1530 makes a decision (input * weight) and then compares this decision with the collected data to identify discrepancies with the collected data. In other words, it determines the error and adjusts the weights 1600 based on this error. Therefore, training the model can be considered a corrective feedback loop.

[0196] For example, a stochastic gradient descent optimization algorithm can be used to train a neural network model, where the gradient is calculated using a backpropagation algorithm. The gradient descent algorithm seeks to change the weights 1600 so that the next evaluation reduces the error, which means that the optimization algorithm is navigating along the gradient (or slope) of the error. Any other suitable optimization algorithm can also be used if sufficiently accurate weights 1600 are provided. Therefore, the training parameters of the neural network 1530 can include weights 1600.

[0197] In the context of optimization algorithms, the function used to evaluate a candidate solution (i.e., a set of weights) is called an objective function. For neural networks where the goal is to minimize error, the objective function can be called a cost function or loss function. In adjusting weights 1600, any suitable method can be used as a loss function, some examples of which are mean squared error (MSE), maximum likelihood (MLE), and cross entropy.

[0198] As for the activation function 1604 of node 1504, it defines the output 1514 of that node 1504 given an input or set of inputs 1500. Node 1504 calculates a weighted sum of the inputs, possibly adding a bias, and then makes a "activate" or "not activate" decision based on a decision threshold as a binary activation or using an activation function 1604 that gives a nonlinear decision function. Any suitable activation function 1604 can be used, such as sigmoid, rectified linear unit (ReLU), normalized exponential function (softmax), sotfplus, tanh, etc. In deep learning, the activation function 1604 is typically set at the layer level and applied to all neurons in that layer. The output 1514 is then used as the input to the next node, and so on, until the desired solution to the original problem is found.

[0199] The techniques and methods described herein can be implemented in various ways. For example, these techniques can be implemented in hardware (one or more devices), firmware (one or more devices), software (one or more modules), or a combination thereof. For hardware implementation, the (multiple) devices of the exemplary embodiments can be implemented in one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), graphics processing units (GPUs), processors, controllers, microcontrollers, microprocessors, other electronic units designed to perform the functions described herein, or a combination thereof. For firmware or software, the modules of at least one chipset (e.g., programs, software codes can be stored in a memory unit and executed by a processor. The memory unit can be implemented inside the processor or outside the processor. In the latter case, it can be communicatively coupled to the processor by various means, as known in the art. In addition, the components of the system described herein can be rearranged and / or supplemented by additional components to facilitate the implementation of various aspects thereof, etc., and they are not limited to the precise configurations set forth in the given figures, as will be understood by those skilled in the art.

[0200] It is obvious to those skilled in the art that, as technology advances, the concepts of the present invention may be implemented in various ways. The embodiments are not limited to the exemplary embodiments described above, but may vary within the scope of the claims. Therefore, all words and expressions should be interpreted broadly, and they are intended to illustrate, not to restrict, the exemplary embodiments.

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 to at least: generating a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble, a row of the preamble matrix indicates a cyclic shift of the random access preamble and a column of the preamble matrix indicates a root sequence of the random access preamble, or the row indicates the root sequence and the column indicates the cyclic shift; Obtaining a sequence encoded using the leading matrix; as well as The sequence is transmitted to one or more user devices. 2 . The apparatus of claim 1 , wherein the plurality of entries comprises at least two sets of values ​​or symbols, and the subset comprises a set of values ​​or symbols from the at least two sets of values ​​or symbols.

3. The apparatus according to claim 1 , further caused to: The preamble matrix is ​​configured as a preamble set for detecting a random access preamble indicated by at least a subset of the plurality of entries.

4. 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 to at least: receiving a sequence from a network element of a wireless access network, wherein the sequence is encoded using a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble, a row of the preamble matrix indicates a cyclic shift of the random access preamble and a column of the preamble matrix indicates a root sequence of the random access preamble, or the row indicates the root sequence and the column indicates the cyclic shift; decoding the sequence; generating a root sequence and a cyclic shift based at least in part on the decoding; generating a random access preamble based on the root sequence and the cyclic shift; as well as The random access preamble is sent to the network element. 5 . The apparatus of claim 4 , wherein the plurality of entries comprises at least two sets of values ​​or symbols, and the subset comprises a set of values ​​or symbols from the at least two sets of values ​​or symbols.

6. The apparatus of claim 4, wherein the decoding comprises mapping the sequence to the preamble matrix, wherein the device is further caused to: selecting an entry from said at least a subset of the plurality of entries included in the predecessor matrix, The root sequence and the cyclic shift are generated based on columns and rows of entries in the leading matrix.

7. The apparatus of claim 4, wherein the decoding comprises generating the root sequence and the cyclic shift from the sequence.

8. The apparatus of claim 4, wherein the sequence is decoded using a decoder trained by: creating a first matrix including a plurality of random access preamble sets; selecting a sample batch from the first matrix; Inputting the sample batch into an encoder having at least one neural network layer; receiving a sequence as output from the encoder, wherein the sequence is encoded using the batch of samples; inputting the sequence into the decoder having at least one neural network layer; receiving a second matrix as an output from the decoder; determining a binary cross entropy loss that compares the batch of samples to the second matrix; Propagating the binary cross entropy loss to the decoder and the encoder through an optimizer; and The selecting of the sample batch, the inputting of the sample batch, the receiving of the sequence, the inputting of the sequence, the receiving of the second matrix, the determining, and the propagating are repeated until the binary cross entropy loss is lower than a threshold.

9. A method comprising: generating a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble, a row of the preamble matrix indicates a cyclic shift of the random access preamble and a column of the preamble matrix indicates a root sequence of the random access preamble, or the row indicates the root sequence and the column indicates the cyclic shift; Obtaining a sequence encoded using the leading matrix; as well as The sequence is transmitted to one or more user devices.

10. A method comprising: receiving a sequence from a network element of a wireless access network, wherein the sequence is encoded using a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble, a row of the preamble matrix indicates a cyclic shift of the random access preamble and a column of the preamble matrix indicates a root sequence of the random access preamble, or the row indicates the root sequence and the column indicates the cyclic shift; decoding the sequence; generating a root sequence and a cyclic shift based at least in part on the decoding; generating a random access preamble based on the root sequence and the cyclic shift; as well as The random access preamble is sent to the network element.

11. A non-transitory computer-readable medium comprising program instructions that, when executed by a device, cause the device to at least: generating a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble, a row of the preamble matrix indicates a cyclic shift of the random access preamble and a column of the preamble matrix indicates a root sequence of the random access preamble, or the row indicates the root sequence and the column indicates the cyclic shift; Obtaining a sequence encoded using the leading matrix; as well as The sequence is transmitted to one or more user devices.

12. A non-transitory computer-readable medium comprising program instructions that, when executed by a device, cause the device to at least: receiving a sequence from a network element of a wireless access network, wherein the sequence is encoded using a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble, a row of the preamble matrix indicates a cyclic shift of the random access preamble and a column of the preamble matrix indicates a root sequence of the random access preamble, or the row indicates the root sequence and the column indicates the cyclic shift; decoding the sequence; generating a root sequence and a cyclic shift based at least in part on the decoding; generating a random access preamble based on the root sequence and the cyclic shift; as well as The random access preamble is sent to the network element.

13. A system comprising at least one or more user equipment and a network element of a wireless access network; The network element is configured as follows: generating a preamble matrix comprising a plurality of entries, wherein at least a subset of the plurality of entries indicates a random access preamble, a row of the preamble matrix indicates a cyclic shift of the random access preamble and a column of the preamble matrix indicates a root sequence of the random access preamble, or the row indicates the root sequence and the column indicates the cyclic shift; Obtaining a sequence encoded using the leading matrix; as well as transmitting the sequence to the one or more user devices; The one or more user equipments are configured to: receiving the sequence from the network element; decoding the sequence; generating a root sequence and a cyclic shift based at least in part on the decoding; generating a random access preamble based on the root sequence and the cyclic shift; as well as The random access preamble is sent to the network element.

Images