IP Library › Granted Patent US 11,974,138
Granted Patent B2
US 11,974,138 · App. 17/478,669 · Granted Apr 30, 2024

Multi-stage Markov decision process (MDP) model for a spectrum sharing system

Inventors: Akash Sandeep Doshi (Austin, TX); Srinivas Yerramalli (San Diego, CA); Lorenzo Ferrari (Oakland, CA); Taesang Yoo (San Diego, CA)
Assignee: QUALCOMM Incorporated
H04W16/14G06N3/08H04W24/08
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,974,138
App. No.
17/478,669
Granted
Apr 30, 2024
Kind
B2
Abstract

A method of wireless communication performed by a first transmission device incudes receiving a first set of observations from a first receiving device based on transmissions from the first transmission device to the first receiving device at a previous time slot of a fixed contention-based spectrum sharing system. The first transmission device shares a spectrum with a second transmission device. The method also includes measuring an energy level from an ongoing data transmission of the second transmission device at a current time slot. The method further includes generating, at an artificial neural network of the first transmission device, at least one of a transmission determination, a set of transmission parameters, or a combination thereof, based on a second set of observations, the energy level, and a counter. The method also includes transmitting, to the first receiving device, during the current time slot based on at least one of the transmission determination, the set of transmission parameters, or the combination thereof.

Claims (73)

1. A method for wireless communication performed by a first transmission device, comprising:

receiving a first set of observations from a first receiving device based on transmissions from the first transmission device to the first receiving device at a previous time slot of a fixed contention-based spectrum sharing system, the first transmission device sharing a spectrum with a second transmission device;

measuring an energy level from an ongoing data transmission of the second transmission device at a current time slot;

generating, at an artificial neural network of the first transmission device, at least one of a transmission determination, a set of transmission parameters, or a combination thereof, based on a second set of observations, the energy level, and a counter; and

transmitting, to the first receiving device, during the current time slot based on at least one of the transmission determination, the set of transmission parameters, or the combination thereof.

2. The method of claim 1 , in which the second set of observations comprise the first set of observations or a subset of the first set of observations.

3. The method of claim 1 , in which:

the transmission device action comprises a determination to transmit to the first receiving device during the data transmission phase or a determination to refrain from transmitting to the first receiving device; and

the set of transmission parameters comprises at least one of an energy detection threshold for determining whether to transmit, a modulation and coding scheme, a transmission power, a precoding, an identity of the second transmission device, an estimated state of the second transmission device, a value associated with each potential transmission device action, a mean value of all potential transmission device actions and a difference for each potential action.

4. The method of claim 1 , in which generating the transmission determination comprises maximizing a reward aggregated across the first transmission device and the second transmission device based on a reward function of the first artificial neural network.

5. The method of claim 4 , in which the reward comprises a served rate for each receiving device served by the first transmission device and the second transmission device.

6. The method of claim 4 , further comprising training the artificial neural network to maximize the reward.

7. The method of claim 1 , in which the first set of observations comprises an average serving rate, a signal strength, and an interference power.

8. The method of claim 7 , further comprising initializing the average serving rate, the signal strength, and the interference power to a pre-determined value based on the previous time slot being an initial time slot of a sequence of time slots.

9. The method of claim 1 , in which the energy level is zero when the first transmission device is scheduled to transmit before the second transmission device.

10. The method of claim 1 , further comprising randomly drawing the counter at the current time slot, the counter determining a first transmission order of the first transmission device in relation to a second transmission order of the second transmission device.

11. The method of claim 1 , in which the first transmission device is a base station and the first receiving device is a user equipment (UE).

12. The method of claim 1 , in which the first transmission device is a user equipment (UE) and the first receiving device is a base station.

13. The method of claim 1 , further comprising determining a set of serving parameters for a subsequent time slot based on the transmission at the current time slot.

14. A first transmission device for wireless communication, comprising:

at least one processor;

memory coupled with the at least one processor; and

instructions stored in the memory and operable, when executed by the at least one processor, to cause the first transmission device:

to receive a first set of observations from a first receiving device based on transmissions from the first transmission device to the first receiving device at a previous time slot of a fixed contention-based spectrum sharing system, the first transmission device sharing a spectrum with a second transmission device;

to measure an energy level from an ongoing data transmission of the second transmission device at a current time slot;

to generate, at an artificial neural network, at least one of a transmission determination, a set of transmission parameters, or a combination thereof, based on a second set of observations, the energy level, and a counter; and

to transmit, to the first receiving device, during the current time slot based on at least one of the transmission determination, the set of transmission parameters, or the combination thereof.

15. The first transmission device of claim 14 , in which the second set of observations comprise the first set of observations or a subset of the first set of observations.

16. The first transmission device of claim 14 , in which:

the transmission device action comprises a determination to transmit to the first receiving device during the data transmission phase or a determination to refrain from transmitting to the first receiving device; and

the set of transmission parameters comprises at least one of an energy detection threshold to determine whether to transmit, a modulation and coding scheme, a transmission power, a precoding, an identity of the second transmission device, an estimated state of the second transmission device, a value associated with each potential transmission device action, a mean value of all potential transmission device actions and a difference for each potential action.

17. The first transmission device of claim 14 , in which execution of the instructions further cause the first transmission device to generate the transmission determination and maximize a reward aggregated across the first transmission device and the second transmission device based on a reward function of the first artificial neural network.

18. The first transmission device of claim 17 , in which the reward comprises a served rate for each receiving device served by the first transmission device and the second transmission device.

19. The first transmission device of claim 17 , in which execution of the instructions further cause the first transmission device to train the artificial neural network to maximize the reward.

20. The first transmission device of claim 14 , in which the first set of observations comprises an average serving rate, a signal strength, and an interference power.

21. The first transmission device of claim 20 , in which execution of the instructions further cause the first transmission device to initialize the average serving rate, the signal strength, and the interference power to a pre-determined value based on the previous time slot being an initial time slot of a sequence of time slots.

22. The first transmission device of claim 14 , in which the energy level is zero when the first transmission device is scheduled to transmit before the second transmission device.

23. The first transmission device of claim 14 , in which execution of the instructions further cause the first transmission device to randomly draw the counter at the current time slot, the counter to determine a first transmission order of the first transmission device in relation to a second transmission order of the second transmission device.

24. The first transmission device of claim 14 , in which the first transmission device is a base station and the first receiving device is a user equipment (UE).

25. The first transmission device of claim 14 , in which the first transmission device is a user equipment (UE) and the first receiving device is a base station.

26. The first transmission device of claim 14 , in which execution of the instructions further cause the first transmission device to determine a set of serving parameters for a subsequent time slot based on the transmission at the current time slot.

27. An apparatus for wireless communication performed at a first transmission device, comprising:

means for receiving a first set of observations from a first receiving device based on transmissions from the first transmission device to the first receiving device at a previous time slot of a fixed contention-based spectrum sharing system, the first transmission device sharing a spectrum with a second transmission device;

means for measuring an energy level from an ongoing data transmission of the second transmission device at a current time slot;

means for generating, at an artificial neural network of the first transmission device, at least one of a transmission determination, a set of transmission parameters, or a combination thereof, based on a second set of observations, the energy level, and a counter; and

means for transmitting, to the first receiving device, during the current time slot based on at least one of the transmission determination, the set of transmission parameters, or the combination thereof.

28. The apparatus of claim 27 , in which the second set of observations comprise the first set of observations or a subset of the first set of observations.

29. The apparatus of claim 27 , in which:

the transmission device action comprises a determination to transmit to the first receiving device during the data transmission phase or a determination to refrain from transmitting to the first receiving device; and

the set of transmission parameters comprises at least one of an energy detection threshold for determining whether to transmit, a modulation and coding scheme, a transmission power, a precoding, an identity of the second transmission device, an estimated state of the second transmission device, a value associated with each potential transmission device action, a mean value of all potential transmission device actions and a difference for each potential action.

30. The apparatus of claim 27 , in which the means for generating the transmission determination comprises means for maximizing a reward aggregated across the first transmission device and the second transmission device based on a reward function of the first artificial neural network.

31. The apparatus of claim 27 , in which:

the first set of observations comprises an average serving rate, a signal strength, and an interference power; and

the apparatus further comprises means for initializing the average serving rate, the signal strength, and the interference power to a pre-determined value based on the previous time slot being an initial time slot of a sequence of time slots.

32. The apparatus of claim 27 , further comprising means for randomly drawing the counter at the current time slot, the counter determining a first transmission order of the first transmission device in relation to a second transmission order of the second transmission device.

33. A non-transitory computer-readable medium having program code recorded thereon for wireless communication at a first transmission device, the program code executed by at least one processor and comprising:

program code to receive a first set of observations from a first receiving device based on transmissions from the first transmission device to the first receiving device at a previous time slot of a fixed contention-based spectrum sharing system, the first transmission device sharing a spectrum with a second transmission device;

program code to measure an energy level from an ongoing data transmission of the second transmission device at a current time slot;

program code to generate, at an artificial neural network, at least one of a transmission determination, a set of transmission parameters, or a combination thereof, based on a second set of observations, the energy level, and a counter; and

program code to transmit, to the first receiving device, during the current time slot based on at least one of the transmission determination, the set of transmission parameters, or the combination thereof.

34. The non-transitory computer-readable medium of claim 33 , in which the second set of observations comprise the first set of observations or a subset of the first set of observations.

35. The non-transitory computer-readable medium of claim 33 , in which:

the transmission device action comprises a determination to transmit to the first receiving device during the data transmission phase or a determination to refrain from transmitting to the first receiving device; and

the set of transmission parameters comprises at least one of an energy detection threshold for determining whether to transmit, a modulation and coding scheme, a transmission power, a precoding, an identity of the second transmission device, an estimated state of the second transmission device, a value associated with each potential transmission device action, a mean value of all potential transmission device actions and a difference for each potential action.

36. The non-transitory computer-readable medium of claim 33 , in which the program code further comprises program code to generate the transmission determination and maximize a reward aggregated across the first transmission device and the second transmission device based on a reward function of the first artificial neural network.

37. The non-transitory computer-readable medium of claim 33 , in which:

the first set of observations comprises an average serving rate, a signal strength, and an interference power; and

the program code further comprises program code to initialize the average serving rate, the signal strength, and the interference power to a pre-determined value based on the previous time slot being an initial time slot of a sequence of time slots.

38. The non-transitory computer-readable medium of claim 33 , in which the energy level is zero when the first transmission device is scheduled to transmit before the second transmission device.

39. The non-transitory computer-readable medium of claim 33 , in which:

the program code further comprises program code to randomly draw the counter at the current time slot; and

the counter determines a first transmission order of the first transmission device in relation to a second transmission order of the second transmission device.

40. The non-transitory computer-readable medium of claim 33 , in which the program code further comprises program code to determine a set of serving parameters for a subsequent time slot based on the transmission at the current time slot.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 18, 2021
From: DOSHI, AKASH SANDEEP; YERRAMALLI, SRINIVAS; FERRARI, LORENZO; YOO, TAESANG
To: QUALCOMM INCORPORATED
Reel/Frame 057824/0942 →
Continuity (2)
Provisional Application 63080962 · Sep 21, 2020
Related Publication 20220095119A1 · Mar 24, 2022