AI-Based Selection of Network and Initial Access in 5G/6G
An unsolved problem in 5G and 6G is how a mobile user device can select the most appropriate base station or access point for communication. Currently, the user device is forced to do a tedious and time-consuming blind search, and even then is not guaranteed to have found the best one. Therefore, to assist user devices (especially reduced-capability IoT devices) in network discovery, disclosed herein is a “hailing” message that the user device can broadcast, and determine from the base station reply messages which one is closest, or has the best signal, or other criterion. The base stations, upon detecting the hailing message, can use an AI model to determine whether they can accommodate another user device (optionally with coordination among multiple access points for load-leveling), and if so, can transmit a reply message after a delay related to the received power in the hailing message.
1 . A base station of a wireless network, the base station containing non-transitory computer-readable media containing instructions that, when executed by a computing environment, cause a method to be performed, the method comprising;
a) receiving a broadcast hailing message from a prospective user device on a predetermined hailing frequency, wherein the broadcast hailing message indicates that a reply message is requested from each base station that can accept a new user;
b) measuring an amplitude of the broadcast hailing message as received by the base station;
c) determining a current traffic density and a current message fault rate in the wireless network;
d) providing, as input to an artificial intelligence model, the amplitude of the broadcast hailing message, the current traffic density, and the current message fault rate;
e) determining, as output from the artificial intelligence model, whether the base station can accept the prospective user device;
f) upon determining that the base station can accept the prospective user device, calculating a delay time inversely related to the amplitude, the delay time not exceeding a predetermined maximum delay time; and
g) after the delay time, transmitting, according to a predetermined transmission power level, a reply message to the prospective user device on the predetermined hailing frequency.
2 . The base station of claim 1 , wherein the reply message is transmitted according to 5G or 6G technology.
3 . The base station of claim 1 , the method further comprising:
a) after the base station detects a communication from another base station to the prospective user device during the delay time, waiting a predetermined listen-before-talk (LBT) interval, and then, after no further transmissions are detected during the LBT interval, transmitting the reply message to the prospective user device on the predetermined hailing frequency.
4 . The base station of claim 1 , wherein the reply message specifies a particular frequency, or an offset indicating the particular frequency.
5 . The base station of claim 4 , wherein the particular frequency is related to a physical broadcast channel or a random access channel of the base station.
6 . The base station of claim 1 , wherein the reply message indicates a first frequency and a second frequency different from the first frequency.
7 . The base station of claim 1 , wherein the reply message specifies system information of the base station.
8 . A wireless network comprising a first base station in signal communication with a plurality of user devices, the first base station configured to:
a) receive a hailing message from a prospective user device, and measure a value related to the hailing message, the hailing message received on a hailing frequency allocated for user devices to transmit discovery messages to base stations;
b) determine a current traffic density and a current message fault rate of the first base station;
c) provide, as input to an artificial intelligence model, a maximum traffic capacity of the first base station, the current traffic density, the current message fault rate, and the value related to the hailing message;
d) determine, as output from the artificial intelligence model, whether the first base station can accommodate the prospective user device;
e) upon determining that the first base station can accommodate the prospective user device, communicate the value related to the hailing message to a network administrative entity; and
f) upon receiving, from the network administrative entity, an instruction to accept the prospective user device, transmit a reply message to the prospective user device.
9 . The wireless network of claim 8 , wherein the value related to the hailing message comprises at least one of:
a) an amplitude of the as-received hailing message;
b) a power level of the as-received hailing message;
c) a signal quality of the as-received hailing message; and
d) a signal-to-noise ratio of the as-received hailing message.
10 . The wireless network of claim 8 , wherein the value related to the hailing message comprises a distance between the prospective user device and the first base station.
11 . The wireless network of claim 8 , further comprising:
a) determining whether the hailing message is an emergency message;
b) when the hailing message is not an emergency message, ignoring the hailing message;
c) when the hailing message is an emergency message, determining whether another base station has responded to the emergency message; and
d) after no other base station has responded to the emergency message, transmitting a reply message to the prospective user device.
12 . The wireless network of claim 8 , further comprising:
a) when the hailing message indicates that the prospective user device has not received a synchronization signal block (SSB) message of the first base station, transmitting, in the reply message, system information of the first base station.
13 . A wireless network base station containing non-transitory computer-readable media containing instructions that, when executed by a computing environment, cause a method to be performed, the method comprising:
a) receiving a broadcast hailing message from a user device on a predetermined frequency, the broadcast hailing message indicating that a reply message is requested from one or more base stations;
b) measuring an amplitude of the broadcast hailing message as received;
c) providing, as input to an artificial intelligence model, the amplitude of the broadcast hailing message as received and a current message fault rate of the wireless network, and determining, as output from the artificial intelligence model, whether the wireless network can accommodate the user device;
d) upon determining that the wireless network can accommodate the user device, calculating a delay time inversely related to the amplitude;
e) monitoring the predetermined frequency during the delay time; and
f) after no communication from another base station to the user device is detected during the delay time, transmitting a reply message to the user device on the predetermined frequency.
14 . The base station of claim 13 , the method further comprising:
a) when a communication from another base station to the user device is detected during the delay time, waiting a predetermined listen-before-talk (LBT) interval, and then, after no further transmissions are detected, transmitting the reply message to the user device on the predetermined frequency.
15 . The base station of claim 13 , wherein the reply message specifies a particular frequency or an offset indicating the particular frequency.
16 . The base station of claim 15 , wherein the particular frequency is related to a physical broadcast channel or a random access channel of the base station.
17 . The base station of claim 13 , wherein the reply message indicates a first frequency and a second frequency different from the first frequency.
18 . The base station of claim 13 , wherein the reply message specifies system information of the base station.
19 . The base station of claim 13 , the method further comprising:
a) measuring a signal quality of the broadcast hailing message; and
b) providing, as further input to the artificial intelligence model, the measured signal quality.
20 . The base station of claim 13 , the method further comprising:
a) measuring a noise or interference level associated with the broadcast hailing message; and
b) providing, as further input to the artificial intelligence model, the measured noise or interference level.