KR-20260067999-A - ADAPTIVE MODEM FOR WIRELESS DEVICES

Abstract

Methods and systems for an adaptive modem for wireless devices include, among other things, receiving a wireless signal transmitted by an access point; identifying waveform characteristics of the wireless signal; determining whether the wireless signal contains beacon information based on the waveform characteristics; determining a wireless standard for the wireless signal based on the waveform characteristics; obtaining a first pre-trained set of weights for the first type of wireless standard in response to determining that the wireless standard implemented by the wireless signal is a first type of wireless standard and that the wireless signal contains beacon information; applying the first pre-trained set of weights to a first machine learning (ML) model to construct a first ML model for the first type of wireless standard; and recovering one or more bits of beacon information from the wireless signal using the first ML model constructed for the first type of wireless standard.

Inventors

• 산트라, 아비크
• 우른, 키란

Assignees

• 인피니언 테크놀로지스 아메리카스 코퍼레이션

Dates

Publication Date

20260513

Application Date

20251103

Priority Date

20250425

Claims (20)

1. As a method, A step of receiving a wireless signal transmitted by an access point; A step of identifying the waveform characteristics of the above wireless signal; A step of determining whether the wireless signal includes beacon information based on the waveform characteristics above; A step of determining a wireless standard for the wireless signal based on the waveform characteristics above; A step of obtaining a first pre-trained set of weights for the first type of wireless standard in response to determining that the wireless standard is a first type of wireless standard and that the wireless signal contains beacon information - the first type of wireless standard implemented by the wireless signal is a wireless local area network (WLAN) utilizing a spreading technique of orthogonal frequency division multiplexing (OFDM) -; A step of applying the first pre-trained weight set to the first machine learning (ML) model to construct the first ML model for the first type of wireless standard; and A step of recovering one or more bits of the beacon information from the wireless signal using the first ML model configured for the first type of wireless standard. A method including
2. In paragraph 1, The step of recovering one or more bits of the beacon information from the wireless signal using the first ML model configured for the first type of wireless standard is: A step of converting the time domain signal of the above wireless signal into a frequency domain signal; A step of extracting OFDM symbols from the above frequency domain signal; and A step of recovering one or more bits of the beacon information from the above OFDM symbols A method including
3. In paragraph 1, In response to determining that the wireless standard is a second type of wireless standard and that the wireless signal contains beacon information, a step of obtaining a second pre-trained set of weights for the second type of wireless standard—the second type of wireless standard implemented by the wireless signal is a WLAN utilizing a direct sequence spread spectrum (DSSS) spreading technique—; A step of constructing the first ML model for the second type of wireless standard by applying the second pre-trained weight set to the first ML model; and A step of recovering one or more bits of the beacon information from the wireless signal using the first ML model configured for the second type of wireless standard. A method that further includes.
4. In paragraph 3, The step of recovering one or more bits from the wireless signal using the first ML model configured for the second type of wireless standard is: A step of performing channel equalization on the above wireless signal; A step of extracting symbols from the equalized wireless signal; and Step of recovering one or more bits from the above symbols A method including
5. In paragraph 1, In response to determining that the above wireless standard is a third type of wireless standard, a step of obtaining a third pre-trained set of weights for the said third type of wireless standard—the said third type of wireless standard implemented by the said wireless signal is a Wireless Personal Area Network (WPAN)—; A step of constructing the first ML model for the third type of wireless standard by applying the third pre-trained weight set to the first ML model; and A step of recovering one or more bits from the wireless signal using the first ML model configured for the third type of wireless standard. A method that further includes.
6. In paragraph 5, The step of recovering one or more bits from the wireless signal using the first ML model configured for the third type of wireless standard is: A step of obtaining a channel estimate of the wireless signal by performing channel estimation for a reference signal; A step of obtaining an equalized wireless signal by performing channel equalization on the data of the wireless signal using the above channel estimate; and A step of recovering one or more bits from the equalized radio signal using the first ML model configured for the third type of radio standard. A method including
7. In paragraph 6, A method for recovering one or more bits from a wireless signal based on the channel estimate, wherein the channel estimation is performed by a second ML model and the first ML model configured for the third type of wireless standard is used.
8. As a station device, Physical layer (PHY) for performing operations Includes, and the above operations are: The operation of receiving a wireless signal transmitted by an access point; An operation to identify the waveform characteristics of the above wireless signal; An operation to determine whether the wireless signal includes beacon information based on the waveform characteristics above; An operation to determine a wireless standard for the wireless signal based on the waveform characteristics above; An operation to acquire a first pre-trained set of weights for the first type of wireless standard in response to determining that the wireless standard implemented by the wireless signal is a first type of wireless standard and that the wireless signal contains beacon information - the first type of wireless standard implemented by the wireless signal is a wireless local area network (WLAN) utilizing a spreading technique of orthogonal frequency division multiplexing (OFDM) -; The operation of applying the first pre-trained weight set to the first machine learning (ML) model to construct the first ML model for the first type of wireless standard; and An operation to recover one or more bits of the beacon information from the wireless signal using the first ML model configured for the first type of wireless standard. A station device including
9. In paragraph 8, The operation of recovering one or more bits of the beacon information from the wireless signal using the first ML model configured for the first type of wireless standard is: An operation to convert the time domain signal of the above wireless signal into a frequency domain signal; The operation of extracting OFDM symbols from the above frequency domain signal; and Operation of recovering one or more bits of the beacon information from the above OFDM symbols A station device including
10. In paragraph 8, The above PHY is for performing operations, and the operations are: An operation to acquire a second pre-trained set of weights for the second type of wireless standard in response to determining that the wireless standard implemented by the wireless signal is a second type of wireless standard and that the wireless signal contains beacon information - the second type of wireless standard implemented by the wireless signal is a WLAN utilizing a direct sequence spread spectrum (DSSS) spreading technique -; The operation of applying the second pre-trained weight set to the first ML model to construct the first ML model for the second type of wireless standard; and An operation to recover one or more bits of the beacon information from the wireless signal using the first ML model configured for the second type of wireless standard. A station device including additionally.
11. In Paragraph 10, The operation of recovering one or more bits from the wireless signal using the first ML model configured for the second type of wireless standard is: Operation of performing channel equalization for the above wireless signal; The operation of extracting symbols from the above-mentioned equalized wireless signal; and The operation of recovering one or more bits from the above symbols A station device including
12. In paragraph 8, The above PHY is for performing operations, and the operations are: An operation to acquire a third pre-trained set of weights for the third type of wireless standard in response to determining that the wireless standard implemented by the wireless signal is a third type of wireless standard - the third type of wireless standard implemented by the wireless signal is a Wireless Personal Area Network (WPAN) -; The operation of applying the third pre-trained weight set to the first ML model to construct the first ML model for the third type of wireless standard; and An operation to recover one or more bits from the wireless signal using the first ML model configured for the third type of wireless standard. A station device including additionally.
13. In Paragraph 12, The operation of recovering one or more bits from the wireless signal using the first ML model configured for the third type of wireless standard is: An operation to obtain a channel estimate of the wireless signal by performing channel estimation for a reference signal; An operation to obtain an equalized wireless signal by performing channel equalization on the data of the wireless signal using the above channel estimate; and An operation to recover one or more bits from the equalized radio signal using the first ML model configured for the third type of radio standard. A station device including
14. In Paragraph 13, A station device that performs the channel estimation using a second ML model and recovers one or more bits from the wireless signal based on the channel estimate using the first ML model configured for the third type of wireless standard.
15. As a wireless network, Access Point (AP); and Station device It includes, and the physical layer (PHY) of the station device is for performing operations, and said operations are: The operation of receiving a wireless signal transmitted by the above AP; An operation to identify the waveform characteristics of the above wireless signal; An operation to determine whether the wireless signal includes beacon information based on the waveform characteristics above; An operation to determine a wireless standard for the wireless signal based on the waveform characteristics above; An operation to acquire a first pre-trained set of weights for the first type of wireless standard in response to determining that the wireless standard implemented by the wireless signal is a first type of wireless standard and that the wireless signal contains beacon information - the first type of wireless standard implemented by the wireless signal is a wireless local area network (WLAN) utilizing a spreading technique of orthogonal frequency division multiplexing (OFDM) -; The operation of applying the first pre-trained weight set to the first machine learning (ML) model to construct the first ML model for the first type of wireless standard; and An operation to recover one or more bits of the beacon information from the wireless signal using the first ML model configured for the first type of wireless standard. A wireless network including
16. In paragraph 15, The operation of recovering one or more bits of the beacon information from the wireless signal using the first ML model configured for the first type of wireless standard is: An operation to convert the time domain signal of the above wireless signal into a frequency domain signal; The operation of extracting OFDM symbols from the above frequency domain signal; and Operation of recovering one or more bits of the beacon information from the above OFDM symbols A wireless network including
17. In paragraph 15, The PHY of the above station device is for performing operations, and said operations are: An operation to acquire a second pre-trained set of weights for the second type of wireless standard in response to determining that the wireless standard implemented by the wireless signal is a second type of wireless standard and that the wireless signal contains beacon information - the second type of wireless standard implemented by the wireless signal is a WLAN utilizing a direct sequence spread spectrum (DSSS) spreading technique -; The operation of applying the second pre-trained weight set to the first ML model to construct the first ML model for the second type of wireless standard; and An operation to recover one or more bits of the beacon information from the wireless signal using the first ML model configured for the second type of wireless standard. A wireless network that further includes
18. In Paragraph 17, The operation of recovering one or more bits from the wireless signal using the first ML model configured for the second type of wireless standard is: Operation of performing channel equalization for the above wireless signal; The operation of extracting symbols from the above-mentioned equalized wireless signal; and The operation of recovering one or more bits from the above symbols A wireless network including
19. In paragraph 15, The PHY of the above station device is for performing operations, and said operations are: An operation to acquire a third pre-trained set of weights for the third type of wireless standard in response to determining that the wireless standard implemented by the wireless signal is a third type of wireless standard - the third type of wireless standard implemented by the wireless signal is a Wireless Personal Area Network (WPAN) -; The operation of applying the third pre-trained weight set to the first ML model to construct the first ML model for the third type of wireless standard; and An operation to recover one or more bits from the wireless signal using the first ML model configured for the third type of wireless standard. A wireless network that further includes
20. In Paragraph 19, The operation of recovering one or more bits from the wireless signal using the first ML model configured for the third type of wireless standard is: An operation to obtain a channel estimate of the wireless signal by performing channel estimation for a reference signal; An operation to obtain an equalized wireless signal by performing channel equalization on the data of the wireless signal using the above channel estimate; and An operation to recover one or more bits from the equalized radio signal using the first ML model configured for the third type of radio standard. A wireless network including

Description

Adaptive Modem for Wireless Devices This application claims the benefit of U.S. provisional application No. 63/716,898 filed November 6, 2024, which is incorporated herein by reference. Technology field The present disclosure relates to wireless devices, and more specifically, to an adaptive modem for wireless devices. Wireless devices typically implement multiple wireless standards that operate in shared frequency bands. For example, Wireless Local Area Network (WLAN) technologies, including Wi-Fi®, and Wireless Personal Area Network (WPAN) technologies, including Bluetooth® (BT), Bluetooth® Low Energy (BLE), and the Institute of Electrical and Electronics Engineers (IEEE) 802.15.4, share wireless communication bands. These multi-standard wireless devices require specialized processing for the specific requirements of each standard. The aspects and embodiments of the present disclosure will be more fully understood from the detailed description provided below and from the accompanying drawings of the various aspects and embodiments of the present disclosure, but these accompanying drawings are not to be construed as limiting the present disclosure to specific aspects or embodiments and are for illustrative and illustrative purposes only. FIG. 1 is a block diagram of an exemplary wireless network having a station device having an adaptive modem according to implementations of the present disclosure. FIG. 2 is an exemplary drawing of an adaptive modem component according to implementations of the present disclosure. FIG. 3a is an exemplary drawing of an adaptive modem for WPAN signals according to implementations of the present disclosure. FIG. 3b is an exemplary drawing of an adaptive modem for WPAN signals according to implementations of the present disclosure. FIG. 4 illustrates a flow diagram of an exemplary method for processing wireless signals using an adaptive modem according to embodiments of the present disclosure. Aspects of the present disclosure relate to adaptive modems for wireless devices. Wireless devices include modems that convert radio frequency (RF) signals into digital data by executing physical layer (PHY) processing chains. These chains follow a sequence of signal processing stages: front-end corrections for hardware defects, such as bias, imbalance, intermodulation, synchronization and timing recovery, conversion, and equalization, and finally, demodulation and decoding, respectively. The transform and equalization stages for processing wireless signals differ significantly between Wireless Local Area Network (WLAN) and Wireless Personal Area Network (WPAN) standards based on their waveform characteristics. WLAN devices utilize spreading techniques such as Orthogonal Frequency Division Multiplexing (OFDM), which divides data across multiple frequency subcarriers for parallel transmission, and Direct Sequence Spread Spectrum (DSSS), which spreads signals across a wider bandwidth to resist interference. WLAN devices also employ modulation schemes such as Binary Phase Shift Keying (BPSK), which represents binary data by modulating phase shifts; Differential Binary Phase Shift Keying (DBPSK), which modulates phase shifts between binary data; and Differential Quadrature Phase Shift Keying (DQPSK), which encodes multiple bits by modulating phase shifts. WPAN devices focused on power efficiency use simpler modulation schemes such as Gaussian frequency shift keying (GFSK), which modulates frequency using a Gaussian filter for stability, or 8-phase differential phase shift keying (8DPSK), which encodes data through 8 distinct phase changes. To support multiple standards, current implementations in wireless devices include separate modems for each standard. Each modem implementing a WLAN or WPAN standard processes the entire PHY chain independently. This involves front-end correction, demodulation, and decoding, even for simple tasks such as beacon detection. However, current implementations increase hardware redundancy and power consumption while decreasing energy efficiency, posing significant challenges for battery-powered devices. The aspects and embodiments of the present disclosure address these and other limitations of the prior art by implementing a machine learning (ML) model using a plurality of pre-trained weight sets that can be dynamically selected to construct an ML model for one of a plurality of wireless standards. For example, a radio frequency (RF) interface of a station device (STA) receives a wireless signal transmitted by an access point (AP). The received wireless signal is provided to the physical layer (PHY) of the STA, which includes an ML model. The STA's PHY identifies the waveform characteristics of the wireless signal, and these waveform characteristics include various attributes that define the structure and behavior of the signal. These characteristics can consist of the modulation scheme and spreading technique used for the wireless signal. The spreading techniq