WO2018074766A1 - Method and apparatus for transmitting wakeup packet in wireless lan system - Google Patents

Abstract

A method and an apparatus for transmitting a wakeup packet in a wireless LAN system are provided. Specifically, a transmitting apparatus constitutes a wakeup packet and transmits the wakeup packet to a receiving apparatus. The wakeup packet includes a preamble sequence composed of an ON signal and an OFF signal to which an OOK scheme is applied. The ON signal is transmitted through an ON symbol generated by applying a first sequence to consecutive K subcarriers in the 20 MHz band and performing 64-point IFFT. The preamble sequence includes a period sequence and an OFF symbol indicating the end of the period sequence. The period sequence starts with the ON signal. The ON signal is set to a period in units of multiples of the ON symbols.

Description

Method and apparatus for transmitting wake-up packet in WLAN system

The present disclosure relates to a technique for performing low power communication in a WLAN system, and more particularly, to a method and apparatus for transmitting a wake-up packet by applying a OOK scheme or Manchester coding in a WLAN system.

Discussion is underway for the next generation wireless local area network (WLAN). In next-generation WLANs, 1) enhancements to the Institute of Electronics and Electronics Engineers (IEEE) 802.11 physical physical access (PHY) and medium access control (MAC) layers in the 2.4 GHz and 5 GHz bands, and 2) spectral efficiency and area throughput. aims to improve performance in real indoor and outdoor environments, such as in environments where interference sources exist, dense heterogeneous network environments, and high user loads.

The environment mainly considered in the next-generation WLAN is a dense environment having many access points (APs) and a station (STA), and improvements in spectral efficiency and area throughput are discussed in such a dense environment. In addition, in the next generation WLAN, there is an interest in improving practical performance not only in an indoor environment but also in an outdoor environment, which is not much considered in a conventional WLAN.

Specifically, in the next-generation WLAN, there is a great interest in scenarios such as wireless office, smart home, stadium, hotspot, building / apartment, and AP based on the scenario. And STA are discussing about improving system performance in a dense environment with many STAs.

In addition, in the next-generation WLAN, there will be more discussion about improving system performance in outdoor overlapping basic service set (OBSS) environment, improving outdoor environment performance, and cellular offloading, rather than improving single link performance in one basic service set (BSS). It is expected. The directionality of these next-generation WLANs means that next-generation WLANs will increasingly have a technology range similar to that of mobile communications. Considering the recent situation in which mobile communication and WLAN technology are discussed together in the small cell and direct-to-direct (D2D) communication area, the technical and business convergence of next-generation WLAN and mobile communication is expected to become more active.

The present specification proposes a method and apparatus for transmitting a wake-up packet by applying a OOK scheme or Manchester coding in a WLAN system.

An example of the present specification proposes a method and apparatus for transmitting a wake-up packet to a WLAN system.

This embodiment may be operated in a transmitter, the receiver may correspond to a low power wake-up receiver, and the transmitter may correspond to an AP.

First of all, the term “on signal” may correspond to a signal having an actual power value. The off signal may correspond to a signal that does not have an actual power value. The on signal may correspond to information 1, and the off signal may correspond to information 0.

The transmitter configures a wakeup packet.

The transmitter transmits the wakeup packet to the receiver.

How the wakeup packet is configured is as follows.

The preamble of the wakeup packet includes a synchronization sequence including an on signal and an off signal to which an On-Off Keying (OOK) scheme is applied.

The on signal is transmitted through an on symbol generated by applying a first sequence to K consecutive subcarriers in a 20 MHz band and performing a 64-point Inverse Fast Fourier Transform (IFFT). In other words, one bit may be transmitted through one symbol generated by performing an IFFT. In this case, K is a natural number.

The sync sequence includes an off symbol indicating the end of the sync sequence. In addition, the off symbol may be inserted at the end of the synchronization sequence. In other words, the synchronization sequence has a periodicity. The off symbol indicating the end of the synchronization sequence is overhead, but the end of the period can be known to help the time synchronization performance.

The sync sequence begins with the on signal. This is because it is advantageous to time synchronization at the same time as the notification that the signal is coming. In addition, the period of the on signal is set in units of multiples of the on symbol. That is, since information of each symbol itself has a periodicity, and one symbol has a length of 4 us, a synchronization sequence (or preamble sequence) having a periodicity of 4 us multiples can be configured.

For example, when the period is 1 and the length of the sync sequence is 6, the sync sequence may include {1, 1, 1, 1, 1, 0}. This is the case where the period is one times the symbol 4us.

As another example, when the period is 2 and the length of the sync sequence is 6, the sync sequence may consist of {1, 0, 1, 0, 1, 0}. This is the case where the period is twice the symbol 4us.

The transmitter may synchronize time with the receiver by calculating an auto-correlation value of the synchronization sequence. The transmitter calculates the correlation value of the two time signals that differ by the period in the synchronization sequence, and sets the case having the largest correlation value as the start point of the wakeup packet. This makes it fairly easy to time wake up packets.

In addition, the preamble of the wakeup packet may further include a symbol for channel estimation and measurement of received power, and a symbol for identifier.

However, when channel estimation and measurement of received power are performed using the synchronization sequence, the preamble of the wakeup packet may not include symbols for the channel estimation and the measurement of the received power.

In addition, when determining the identifier using the synchronization sequence, the preamble of the wakeup packet may not include a symbol for the identifier. The identifier may include identifier information of a transmitter or a receiver.

The wakeup packet may be applied only to the OOK scheme, or the OOK scheme and Manchester coding may be applied. If a OOK scheme and Manchester coding are applied to the wakeup payload included in the wakeup packet, a symbol for measuring the channel estimate and the received power may not be needed in the preamble of the wakeup packet.

According to an example of the present specification, a wakeup packet may be configured and transmitted by applying a OOK modulation scheme or Manchester coding in a transmitter, thereby reducing power consumption by using an envelope detector when decoding the wakeup packet in a receiver. . Therefore, the receiving device can decode the wakeup packet to the minimum power.

In addition, the transmitter may configure the preamble sequence of the wakeup packet to signal the start of the wakeup packet, and efficiently perform time synchronization using the periodicity of the preamble sequence.

1 is a conceptual diagram illustrating a structure of a wireless local area network (WLAN).

2 is a diagram illustrating an example of a PPDU used in the IEEE standard.

3 is a diagram illustrating an example of a HE PPDU.

4 illustrates a low power wake-up receiver in an environment in which data is not received.

5 illustrates a low power wake-up receiver in an environment in which data is received.

6 shows an example of a wakeup packet structure according to the present embodiment.

7 shows a signal waveform of a wakeup packet according to the present embodiment.

FIG. 8 is a diagram for describing a principle in which power consumption is determined according to a ratio of 1 and 0 of bit values constituting binary sequence information using the OOK method.

9 shows a method of designing a OOK pulse according to the present embodiment.

10 is an explanatory diagram of a Manchester coding scheme according to the present embodiment.

11 illustrates various examples of a symbol repetition technique of repeating n symbols according to the present embodiment.

12 illustrates various examples of a symbol reduction technique according to the present embodiment.

13 shows an example of time synchronization using autocorrelation according to the present embodiment.

14 shows another example of time synchronization using autocorrelation according to the present embodiment.

15 shows another example of time synchronization using autocorrelation according to the present embodiment.

16 shows another example of time synchronization using autocorrelation according to the present embodiment.

17 shows another example of time synchronization using autocorrelation according to the present embodiment.

18 shows another example of time synchronization using autocorrelation according to the present embodiment.

19 is a flowchart illustrating a procedure of transmitting a signal by applying the OOK method according to the present embodiment.

20 is a block diagram illustrating a wireless device to which the present embodiment can be applied.

1 is a conceptual diagram illustrating a structure of a wireless local area network (WLAN).

1 shows the structure of the infrastructure basic service set (BSS) of the Institute of Electrical and Electronic Engineers (IEEE) 802.11.

Referring to the top of FIG. 1, the WLAN system may include one or more infrastructure BSSs 100 and 105 (hereinafter, BSS). The BSSs 100 and 105 are a set of APs and STAs such as an access point 125 and a STA1 (station 100-1) capable of successfully synchronizing and communicating with each other, and do not indicate a specific area. The BSS 105 may include one or more STAs 103-1 and 105-2 that can be coupled to one AP 130.

The BSS may include at least one STA, APs 125 and 130 for providing a distribution service, and a distribution system (DS) 110 for connecting a plurality of APs.

The distributed system 110 may connect several BSSs 100 and 105 to implement an extended service set (ESS) 140 which is an extended service set. The ESS 140 may be used as a term indicating one network in which one or several APs 125 and 230 are connected through the distributed system 110. APs included in one ESS 140 may have the same service set identification (SSID).

The portal 120 may serve as a bridge for connecting the WLAN network (IEEE 802.11) with another network (for example, 802.X).

In the BSS as shown in the upper part of FIG. 1, a network between the APs 125 and 130 and a network between the APs 125 and 130 and the STAs 100-1, 105-1 and 105-2 may be implemented. However, it may be possible to perform communication by setting up a network even between STAs without the APs 125 and 130. A network that performs communication by establishing a network even between STAs without APs 125 and 130 is defined as an ad-hoc network or an independent basic service set (BSS).

1 is a conceptual diagram illustrating an IBSS.

Referring to the bottom of FIG. 1, the IBSS is a BSS operating in an ad-hoc mode. Since IBSS does not contain an AP, there is no centralized management entity. That is, in the IBSS, the STAs 150-1, 150-2, 150-3, 155-4, and 155-5 are managed in a distributed manner. In the IBSS, all STAs 150-1, 150-2, 150-3, 155-4, and 155-5 may be mobile STAs, and access to a distributed system is not allowed, thus making a self-contained network. network).

A STA is any functional medium that includes medium access control (MAC) conforming to the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standard and a physical layer interface to a wireless medium. May be used to mean both an AP and a non-AP STA (Non-AP Station).

The STA may include a mobile terminal, a wireless device, a wireless transmit / receive unit (WTRU), a user equipment (UE), a mobile station (MS), a mobile subscriber unit ( It may also be called various names such as a mobile subscriber unit or simply a user.

On the other hand, the term "user" may be used in various meanings, for example, may also be used to mean an STA participating in uplink MU MIMO and / or uplink OFDMA transmission in wireless LAN communication. It is not limited to this.

2 is a diagram illustrating an example of a PPDU used in the IEEE standard.

As shown, various types of PHY protocol data units (PPDUs) have been used in the IEEE a / g / n / ac standard. Specifically, the LTF and STF fields included training signals, the SIG-A and SIG-B included control information for the receiving station, and the data fields included user data corresponding to the PSDU.

This embodiment proposes an improved technique for the signal (or control information field) used for the data field of the PPDU. The signal proposed in this embodiment may be applied on a high efficiency PPDU (HE PPDU) according to the IEEE 802.11ax standard. That is, the signals to be improved in the present embodiment may be HE-SIG-A and / or HE-SIG-B included in the HE PPDU. Each of HE-SIG-A and HE-SIG-B may also be represented as SIG-A or SIG-B. However, the improved signal proposed by this embodiment is not necessarily limited to the HE-SIG-A and / or HE-SIG-B standard, and controls / control of various names including control information in a wireless communication system for transmitting user data. Applicable to data fields.

3 is a diagram illustrating an example of a HE PPDU.

The control information field proposed in this embodiment may be HE-SIG-B included in the HE PPDU as shown in FIG. 3. The HE PPDU according to FIG. 3 is an example of a PPDU for multiple users. The HE-SIG-B may be included only for the multi-user, and the HE-SIG-B may be omitted in the PPDU for the single user.

As shown, a HE-PPDU for a multiple user (MU) includes a legacy-short training field (L-STF), a legacy-long training field (L-LTF), a legacy-signal (L-SIG), High efficiency-signal A (HE-SIG-A), high efficiency-signal-B (HE-SIG-B), high efficiency-short training field (HE-STF), high efficiency-long training field (HE-LTF) It may include a data field (or MAC payload) and a PE (Packet Extension) field. Each field may be transmitted during the time period shown (ie, 4 or 8 ms, etc.).

The PPDU used in the IEEE standard is mainly described as a PPDU structure transmitted over a channel bandwidth of 20 MHz. The PPDU structure transmitted on a bandwidth wider than the channel bandwidth of 20 MHz (eg, 40 MHz and 80 MHz) may be a structure in which linear scaling of the PPDU structure used in the channel bandwidth of 20 MHz is applied.

The PPDU structure used in the IEEE standard is generated based on 64 Fast Fourier Tranforms (FTFs), and a CP portion (cyclic prefix portion) may be 1/4. In this case, the length of the effective symbol interval (or FFT interval) may be 3.2us, the CP length is 0.8us, and the symbol duration may be 4us (3.2us + 0.8us) plus the effective symbol interval and the CP length.

Wireless networks are ubiquitous, usually indoors and often installed outdoors. Wireless networks use various techniques to send and receive information. For example, but not limited to, two widely used technologies for communication are those that comply with IEEE 802.11 standards such as the IEEE 802.11n standard and the IEEE 802.11ac standard.

The IEEE 802.11 standard specifies a common Medium Access Control (MAC) layer that provides a variety of features to support the operation of IEEE 802.11-based wireless LANs (WLANs). The MAC layer utilizes protocols that coordinate access to shared radios and improve communications over wireless media, such as IEEE 802.11 stations (such as a PC's wireless network card (NIC) or other wireless device or station (STA) and access point ( Manage and maintain communication between APs).

IEEE 802.11ax is the successor to 802.11ac and has been proposed to improve the efficiency of WLAN networks, especially in high density areas such as public hotspots and other high density traffic areas. IEEE 802.11 can also use Orthogonal Frequency Division Multiple Access (OFDMA). The High Efficiency WLAN Research Group (HEW SG) within the IEEE 802.11 Work Group is dedicated to improving system throughput / area in high-density scenarios of APs (access points) and / or STAs (stations) in relation to the IEEE 802.11 standard. We are considering improving efficiency.

Wearable devices and small computing devices such as sensors and mobile devices are constrained by small battery capacities, but use wireless communication technologies such as Wi-Fi, Bluetooth®, and Bluetooth® Low Energy (BLE). Support, connect to and exchange data with other computing devices such as smartphones, tablets, and computers. Since these communications consume power, it is important to minimize the energy consumption of such communications in these devices. One ideal strategy to minimize energy consumption is to power off the communication block as frequently as possible while maintaining data transmission and reception without increasing delay too much. That is, the communication block is transmitted immediately before the data reception, and only when there is data to wake up, the communication block is turned on and the communication block is turned off for the remaining time.

Hereinafter, a low-power wake-up receiver (LP-WUR) will be described.

The communication system (or communication subsystem) described herein includes a main radio (802.11) and a low power wake up receiver.

The main radio is used for transmitting and receiving user data. The main radio is turned off if there are no data or packets to transmit. The low power wake-up receiver wakes up the main radio when there is a packet to receive. At this time, the user data is transmitted and received by the main radio.

The low power wake-up receiver is not for user data. It is simply a receiver to wake up the main radio. In other words, the transmitter is not included. The low power wake-up receiver is active while the main radio is off. Low power wake-up receivers target a target power consumption of less than 1 mW in an active state. In addition, low power wake-up receivers use a narrow bandwidth of less than 5 MHz. In addition, the target transmission range of the low power wake-up receiver is the same as that of the existing 802.11.

4 illustrates a low power wake-up receiver in an environment in which data is not received. 5 illustrates a low power wake-up receiver in an environment in which data is received.

As shown in Figures 4 and 5, if there is data to be transmitted and received, one way to implement an ideal transmission and reception strategy is a main radio such as Wi-Fi, Bluetooth® radio, or Bluetooth® Radio (BLE). Adding a low power wake-up receiver (LP-WUR) that can wake up.

Referring to FIG. 4, the Wi-Fi / BT / BLE 420 is turned off and the low power wake-up receiver 430 is turned on without receiving data. Some studies show that the low power wake-up receiver (LP-WUR) can consume less than 1mW.

However, as shown in FIG. 5, when a wakeup packet is received, the low power wakeup receiver 530 may receive the entire Wi-Fi / BT / BLE radio 520 so that the data packet following the wakeup packet can be correctly received. Wake up). In some cases, however, actual data or IEEE 802.11 MAC frames may be included in the wakeup packet. In this case, it is not necessary to wake up the entire Wi-Fi / BT / BLE radio 520, but only a part of the Wi-Fi / BT / BLE radio 520 to perform the necessary process. This can result in significant power savings.

One example technique disclosed herein defines a method for a granular wakeup mode for Wi-Fi / BT / BLE using a low power wakeup receiver. For example, the actual data contained in the wakeup packet can be passed directly to the device's memory block without waking up the Wi-Fi / BT / BLE radio.

As another example, if a wakeup packet contains an IEEE 802.11 MAC frame, only the MAC processor of the Wi-Fi / BT / BLE wireless device needs to wake up to process the IEEE 802.11 MAC frame included in the wakeup. That is, the PHY module of the Wi-Fi / BT / BLE radio can be turned off or kept in a low power mode.

A number of granular wakeup modes have been defined for Wi-Fi / BT / BLE radios that use low power wake-up receivers, requiring that the Wi-Fi / BT / BLE radio be powered on when a wake-up packet is received. However, according to the above embodiment, only necessary parts (or components) of the Wi-Fi / BT / BLE radio can be selectively woken up, thereby saving energy and reducing the waiting time. Many solutions that use low-power wake-up receivers to receive wake-up packets wake up the entire Wi-Fi / BT / BLE radio. One exemplary aspect discussed herein wakes up only the necessary portions of the Wi-Fi / BT / BLE radio required to process the received data, saving significant amounts of energy and reducing unnecessary latency in waking up the main radio. Can be.

In addition, in the above embodiment, the low power wake-up receiver 530 may wake up the main radio 520 based on the wake-up packet transmitted from the transmitter 500.

In addition, the transmitter 500 may be set to transmit a wakeup packet to the receiver 510. For example, the low power wake-up receiver 530 may be instructed to wake up the main radio 520.

6 shows an example of a wakeup packet structure according to the present embodiment.

The wakeup packet may include one or more legacy preambles. One or more legacy devices may decode or process the legacy preamble.

In addition, the wakeup packet may include a payload after the legacy preamble. The payload may be modulated by a simple modulation scheme, for example, an On-Off Keying (OOK) modulation scheme.

Referring to FIG. 6, the transmitter may be configured to generate and / or transmit a wakeup packet 600. The receiving device may be configured to process the received wakeup packet 600.

In addition, the wakeup packet 600 may include a legacy preamble or any other preamble 610 as defined by the IEEE 802.11 specification. In addition, the wakeup packet 600 may include a payload 620.

Legacy preambles provide coexistence with legacy STAs. The legacy preamble 610 for coexistence uses the L-SIG field to protect the packet. The 802.11 STA may detect the start of a packet through the L-STF field in the legacy preamble 610. The 802.11 STA can know the end of the packet through the L-SIG field in the legacy preamble 610. In addition, by adding a BPSK modulated symbol after the L-SIG, a false alarm of an 802.11n terminal can be reduced. One symbol (4us) modulated with BPSK also has a 20MHz bandwidth like the legacy part. The legacy preamble 610 is a field for third party legacy STAs (STAs not including LP-WUR). The legacy preamble 610 is not decoded from the LP-WUR.

The payload 620 may include a wakeup preamble 622. Wake-up preamble 622 may include a sequence of bits configured to identify wake-up packet 600. The wakeup preamble 622 may include, for example, a PN sequence.

In addition, the payload 620 may include a MAC header 624 including address information of a receiver receiving the wakeup packet 600 or an identifier of the receiver.

In addition, the payload 620 may include a frame body 626 that may include other information of the wakeup packet. For example, the frame body 626 may include length or size information of the payload.

In addition, the payload 620 may include a Frame Check Sequence (FCS) field 628 that includes a Cyclic Redundancy Check (CRC) value. For example, it may include a CRC-8 value or a CRC-16 value of the MAC header 624 and the frame body 626.

7 shows a signal waveform of a wakeup packet according to the present embodiment.

Referring to FIG. 7, the wakeup packet 700 includes a legacy preamble (802.11 preamble, 710) and a payload modulated by OOK. That is, the legacy preamble and the new LP-WUR signal waveform coexist.

In addition, the legacy preamble 710 may be modulated according to the OFDM modulation scheme. That is, the legacy preamble 710 is not applied to the OOK method. In contrast, the payload may be modulated according to the OOK method. However, the wakeup preamble 722 in the payload may be modulated according to another modulation scheme.

If the legacy preamble 710 is transmitted on a channel bandwidth of 20 MHz to which 64 FFTs are applied, the payload may be transmitted on a channel bandwidth of about 4.06 MHz. This will be described later in the OOK pulse design method.

First, a modulation scheme using the OOK scheme and a Manchester coding scheme will be described.

FIG. 8 is a diagram for describing a principle in which power consumption is determined according to a ratio of 1 and 0 of bit values constituting binary sequence information using the OOK method.

Referring to FIG. 8, information in the form of a binary sequence having 1 or 0 as a bit value is represented. By using a bit value of 1 or 0 of the binary sequence information, OOK modulation can be performed. That is, in consideration of the bit values of the binary sequence information, it is possible to perform the communication of the OOK modulation method. For example, when the light emitting diode is used for visible light communication, the light emitting diode is turned on when the bit value constituting the binary sequence information is 1, and the light emitting diode is turned off when the bit value is 0. The light emitting diode can be made to blink. As the light-emitting diode blinks, the receiver receives and restores data transmitted in the form of visible light, thereby enabling communication using visible light. However, since the blinking of the light emitting diode cannot be perceived by the human eye, the person feels that the illumination is continuously maintained.

For convenience of description, as shown in FIG. 8, information in the form of a binary sequence having 10 bit values is used. Referring to FIG. 8, there is information in the form of a binary sequence having a value of '1001101011'. As described above, when the bit value is 1, the transmitter is turned on, and when the bit value is 0, the transmitter is turned off, the symbol is turned on at 6 bit values out of 10 bit values. ) do. Therefore, when the symbol is turned on in all 10 bit values, if the power consumption is 100%, the power consumption is 60% according to the duty cycle of FIG. 8.

That is, it can be said that the power consumption of the transmitter is determined according to the ratio of 1 and 0 constituting the binary sequence information. In other words, if there is a constraint that the power consumption of the transmitter must be kept at a certain value, the ratio of 1 and 0, which constitutes information in binary sequence form, must also be maintained. For example, in the case of lighting equipment, since the lighting must be maintained at a specific luminance value desired by people, the ratio of 1 and 0 constituting the information in the form of a binary sequence must also be maintained.

However, since the receiver is mainly a wake-up receiver (WUR), the transmission power is not important. The main reason for using OOK is that the power consumption is very low when decoding the received signal. Until the decoding is performed, there is no significant difference in power consumption in the main radio or WUR, but a large difference occurs in the decoding process. Below is the approximate power consumption.

-The existing Wi-Fi power consumption is about 100mW. Specifically, power consumption of Resonator + Oscillator + PLL (1500uW)-> LPF (300uW)-> ADC (63uW)-> decoding processing (OFDM receiver) (100mW) may occur.

-WUR power consumption is about 1mW. Specifically, power consumption of Resonator + Oscillator (600uW)-> LPF (300uW)-> ADC (20uW)-> decoding processing (Envelope detector) (1uW) may occur.

9 shows a method of designing a OOK pulse according to the present embodiment.

The OFDM transmitter of 802.11 can be reused to generate OOK pulses. The transmitter can generate a sequence having 64 bits by applying a 64-point IFFT as in 802.11.

The transmitter should generate the payload of the wakeup packet by modulating the OOK method. However, since the wakeup packet is for low power communication, the OOK method is applied to the ON-signal. The on signal is a signal having an actual power value, and the off signal (OFF-signal) corresponds to a signal having no actual power value. The off signal is also applied to the OOK method, but the signal is not generated using the transmitter, and since no signal is actually transmitted, it is not considered in the configuration of the wakeup packet.

In the OOK method, information (bit) 1 may be an on signal and information (bit) 0 may be an off signal. Alternatively, when the Manchester coding scheme is applied, information 1 may indicate a transition from an off signal to an on signal, and information 0 may indicate a transition from an on signal to an off signal. Alternatively, on the contrary, the information 1 may indicate the transition from the on signal to the off signal, and the information 0 may indicate the transition from the off signal to the on signal. Manchester coding scheme will be described later.

Referring to FIG. 9, as shown in the right frequency domain graph 920, the transmitter applies a sequence by selecting 13 consecutive subcarriers of a 20 MHz band as a reference band as a sample. In FIG. 9, 13 subcarriers located among the subcarriers in the 20 MHz band are selected as samples. That is, a subcarrier whose subcarrier index is from -6 to +6 is selected from the 64 subcarriers. In this case, the subcarrier index 0 may be nulled to 0 as the DC subcarrier. Set a specific sequence only to the 13 subcarriers selected as samples, and set all subcarriers except the subcarriers (subcarrier indexes -32 to -7 and subcarrier indexes +7 to +31) to 0. .

In addition, since subcarrier spacing is 312.5 KHz, 13 subcarriers have a channel bandwidth of about 4.06 MHz. That is, it can be said that power is provided only for 4.06MHz in the 20MHz band in the frequency domain. By moving the power to the center, the signal to noise ratio (SNR) can be increased and the power consumption of the AC / DC converter of the receiver can be reduced. In addition, the power consumption can be reduced by reducing the sampling frequency band to 4.06MHz.

In addition, as shown in the left time domain graph 910 of FIG. 9, the transmitter may generate one on-signal in the time domain by performing a 64-point IFFT on 13 subcarriers. One on-signal has a size of 1 bit. That is, a sequence composed of 13 subcarriers may correspond to 1 bit. On the other hand, the transmitter may not transmit the off signal at all. When performing IFFT, a 3.2us symbol may be generated, and if a CP (Cyclic Prefix, 0.8us) is included, one symbol having a length of 4us may be generated. That is, one bit indicating one on-signal may be loaded in one symbol.

The reason for configuring and sending the bits as in the above-described embodiment is to reduce power consumption by using an envelope detector in the receiver. As a result, the receiving device can decode the packet with the minimum power.

However, the basic data rate for one information may be 125 Kbps (8us) or 62.5Kbps (16us).

Generalizing the above, the signal transmitted in the frequency domain is as follows. That is, each signal having a length of K in the 20 MHz band may be transmitted on K consecutive subcarriers of a total of 64 subcarriers. That is, K may correspond to the bandwidth of the OOK pulse by the number of subcarriers used to transmit a signal. All other coefficients of the K subcarriers are zero. In this case, the indexes of the K subcarriers used by the signal corresponding to the information 0 and the information 1 are the same. For example, the subcarrier index used may be represented as 33-floor (K / 2): 33 + ceil (K / 2) -1.

In this case, the information 1 and the information 0 may have the following values.

Information 0 = zeros (1, K)

Information 1 = alpha * ones (1, K)

The alpha is a power normalization factor and may be, for example, 1 / sqrt (K).

10 is an explanatory diagram of a Manchester coding scheme according to the present embodiment.

Manchester coding is a type of line coding, and may indicate information as shown in the following table in a manner in which a transition of a magnitude value occurs in the middle of one bit period.

That is, Manchester coding means a method of converting data from 1 to 01, 0 to 10, 1 to 10, and 0 to 01. Table 1 shows an example in which data is converted from 1 to 10 and 0 to 01 using Manchester coding.

As shown in Fig. 10, the bit string to be transmitted, the Manchester coded signal, the clock reproduced on the receiving side, and the data reproduced on the clock are shown in order from top to bottom.

When the transmitting side transmits data using the Manchester coding scheme, the receiving side reads the data a little later on the basis of the transition point transitioning from 1 → 0 or 0 → 1 and recovers the data, and then transitions to 1 → 0 or 0 → 1. The clock is recovered by recognizing the transition point as the clock transition point. Alternatively, when the symbol is divided based on the transition point, it can be simply decoded by comparing the power at the front and the back at the center of the symbol.

As shown in FIG. 10, the bit string to be transmitted is 10011101, the Manchester coded signal is 0110100101011001, and the clock reproduced on the receiving side recognizes the transition point of the Manchester coded signal as the transition point of the clock. Then, the data is recovered by using the reproduced clock.

By using the Manchester coding scheme, it is possible to communicate in a synchronous manner using only a data transmission channel without using a separate clock.

In addition, this method can use the TXD pin for data transmission and the RXD pin for reception by using only the data transmission channel. Therefore, synchronized bidirectional transmission is possible.

This specification proposes various symbol types that can be used in the WUR and thus data rates.

Since STAs requiring robust performance and STAs receiving strong signals from APs are mixed, it is necessary to support efficient data rates in some situations. In order to obtain reliable and robust performance, a symbol coding based symbol coding technique and a symbol repetition technique may be used. In addition, a symbol reduction technique may be used to obtain a high data rate.

In this case, each symbol may be generated using an existing 802.11 OFDM transmitter. In addition, the number of subcarriers used to generate each symbol may be thirteen. However, it is not limited thereto.

In addition, each symbol may use OOK modulation formed of an ON-signal and an OFF-signal.

One symbol generated for the WUR may be composed of a CP (Cyclic Prefix or Guard Interval) and a signal part representing actual information. Symbols having various data rates may be designed by variously setting or repeating the lengths of the CP and the actual information signal.

The following are various examples of symbol types.

For example, the basic WUR symbol may be represented as CP + 3.2us. That is, one bit is represented using a symbol having the same length as the existing Wi-Fi. Specifically, the transmitting apparatus applies a specific sequence to all available subcarriers (for example, 13 subcarriers) and then performs IFFT to form an information signal portion of 3.2 us. In this case, a coefficient of 0 may be loaded on the DC subcarrier or the middle subcarrier index among all available subcarriers.

Different sequences may be applied to the available subcarriers according to the 3.2us on signal and the 3.2us off signal. A 3.2us off signal can be generated by applying all coefficients to zero.

CP may be used by adopting a specific length from the rear of the information signal 3.2us immediately behind. At this time, CP may be 0.4us or 0.8us. This length is equal to the guard interval of 802.11ac.

Accordingly, one bit information corresponding to one basic WUR symbol may be represented as shown in the following table.

Information ‘0’	Information ‘1’
3.2us OFF-signal	3.2us ON-signal

Table 2 does not indicate CP separately. In fact, CP + 3.2us including CP may point to one 1-bit information. That is, the 3.2us on signal can be viewed as a (CP + 3.2us) on signal. A 3.2us off signal can be seen as a (CP + 3.2us) off signal.

As another example, a symbol to which Manchester coding is applied may be represented as CP + 1.6us + CP + 1.6us or CP + 1.6us + 1.6us. The symbol to which the Manchester coding is applied may be generated as follows.

In the OOK transmission using the Wi-Fi transmitter, the time used for transmitting one bit (or symbol) except for the guard interval of the transmission signal is 3.2 us. In this case, if Manchester coding is applied, a signal size transition should occur at 1.6us. That is, each sub-information having a length of 1.6us should have a value of 0 or 1, and may configure a signal in the following manner.

* Information 0-> 1 0 (each can be referred to as sub information 1 0 or sub symbol 1 (ON) 0 (OFF))

First 1.6us (sub information 1 or sub symbol 1): Sub information 1 may have a value of beta * ones (1, K). Beta is a power normalization factor and may be, for example, 1 / sqrt (ceil (K / 2)).

In addition, a specific sequence is applied in units of two squares to all available subcarriers (eg, 13 subcarriers) to generate a symbol to which Manchester coding is applied. That is, even-numbered subcarriers of a particular sequence are nulled to zero. That is, in a particular sequence, coefficients may exist at intervals of two cells. For example, suppose that 13 subcarriers are used to construct an on-signal, a particular sequence with coefficients spaced two spaces apart is {a 0 b 0 c 0 d 0 e 0 f 0 g}, {0 a 0 b 0 c 0 d 0 e 0 f 0} or {a 0 b 0 c 0 0 0 d 0 e 0 method. At this time, a, b, c, d, e, f, g is 1 or -1.

That is, the transmitter maps a specific sequence to K consecutive subcarriers of 64 subcarriers (for example, 33-floor (K / 2): 33 + ceil (K / 2) -1) and the remaining subcarriers. IFFT is performed by setting the coefficient to 0. In this way, signals in the time domain can be generated. The signal in the time domain is a 3.2us long signal having a 1.6us period because coefficients exist at intervals of two spaces in the frequency domain. One of the first or second 1.6us period signals can be selected and used as sub information 1.

Second 1.6us (sub information 0 or sub symbol 0): The sub information 0 may have a value of zeros (1, K). Similarly, the transmitter maps a specific sequence to K consecutive subcarriers of 64 subcarriers (eg, 33-floor (K / 2): 33 + ceil (K / 2) -1) and performs IFFT. The signal in the time domain can be generated. The sub information 0 may correspond to a 1.6us off signal. The 1.6us off signal can be generated by setting all coefficients to zero.

One of the first or second 1.6us periodic signals of the signal in the time domain may be selected and used as the sub information 0. You can simply use the zeros (1,32) signal as subinformation zero.

* Information 1-> 0 1 (each can be referred to as sub information '0', '1' or sub symbol 0 (OFF) 1 (ON))

Since information 1 is also divided into the first 1.6us (sub information 0) and the second 1.6us (sub information 1), a signal corresponding to each sub information may be configured in the same manner as the information 0 is generated.

By using a technique of generating information 0 and information 1 using Manchester coding, it is possible to prevent the off symbol from continuing as compared to the conventional method. Therefore, a coexistence problem with an existing Wi-Fi device may not occur. The coexistence problem is a problem caused by transmitting a signal by determining that another device is a channel idle state due to a continuous off symbol. If only OOK modulation is used, for example, the off-symbol may be contiguous with the sequence 100001 or the like, but if Manchester coding is used, the off-symbol cannot be contiguous with the sequence 100101010110.

According to the above description, the sub information may be referred to as a 1.6us information signal. The 1.6us information signal may be a 1.6us on signal or a 1.6 off signal. The 1.6us on signal and the 1.6 off signal may have different sequences applied to each subcarrier.

CP can be used by adopting a specific length from the back of the 1.6us of the information signal immediately after. At this time, CP may be 0.4us or 0.8us. This length is equal to the guard interval of 802.11ac.

Accordingly, one bit information corresponding to one Manchester coded symbol may be represented as shown in the following table.

Information ‘0’	Information ‘1’
1.6us ON-signal + 1.6us OFF-signal	1.6us OFF-signal + 1.6us ON-signal
Or 1.6us OFF-signal + 1.6us ON-signal	Or 1.6us ON-signal + 1.6us OFF-signal

Table 3 does not indicate CP separately. In fact, CP + 1.6us + CP + 1.6us or CP + 1.6us + 1.6us including CP may indicate one 1-bit information. That is, in the former case, the 1.6us on signal and the 1.6us off signal may be regarded as the (CP + 1.6us) on signal and the (CP + 1.6us) off signal.

As another example, a method of constructing a wake-up packet by repeating symbols for performance improvement is proposed.

The symbol repetition technique is applied to the wakeup payload 724. The symbol repetition technique means repetition of a time signal after insertion of an IFFT and a cyclic prefix (CP) of each symbol. Thus, the length (time) of the wakeup payload 724 is doubled.

That is, it is proposed to apply a symbol representing information such as information 0 or information 1 to a specific sequence and to repeat the configuration as follows.

Option 1: Information 0 and Information 1 can be repeatedly represented by the same symbol.

Info 0-> 0 0 (repeat Info 0 twice)

-Info 1-> 1 1 (repeat Info 1 twice)

Option 2: Information 0 and Information 1 can be repeatedly represented by different symbols.

Information 0-> 0 1 or 1 0 (repeat info 0 and info 1)

Info 1-> 1 0 or 0 1 (repeat Info 1 and Info 0)

Hereinafter, a method of decoding by a receiving apparatus a signal transmitted by applying a symbol repetition technique to the transmitting apparatus will be described.

The transmitted signal may correspond to a wakeup packet, and a method of decoding the wakeup packet can be largely divided into two types. The first is non-coherent detection and the second is coherent detection. In non-coherent detection, the phase relationship between the transmitter and receiver signals is not fixed. Thus, the receiver does not need to measure and adjust the phase of the received signal. In contrast, the coherent detection method requires that the phase of the signal between the transmitter and the receiver be aligned.

The receiver includes the low power wake-up receiver described above. The low power wake-up receiver may decode a packet (wake-up packet) transmitted using an OOK modulation scheme using an envelope detector to reduce power consumption.

The envelope detector measures and decodes the power or magnitude of the received signal. The receiver sets a threshold based on the power or magnitude measured by the envelope detector. When decoding the symbol to which the OOK is applied, it is determined as information 1 if it is greater than or equal to the threshold value, and as information 0 when it is smaller than the threshold value.

The method of decoding a symbol to which the symbol repetition technique is applied is as follows. In the option 1, the receiving apparatus may use the wake-up preamble 722 to calculate a power when symbol 1 (symbol including information 1) is transmitted and determine the threshold.

In detail, the average power of the two symbols may be determined to determine information 1 (1 1) if the value is equal to or greater than the threshold value, and to determine information 0 (0 0) if the value is less than the threshold value.

In addition, in option 2, information may be determined by comparing the power of two symbols without determining a threshold.

Specifically, if information 1 is composed of 0 1 and information 0 is composed of 1 0, it is determined as information 0 if the power of the first symbol is greater than the power of the second symbol. On the contrary, if the power of the first symbol is less than the power of the second symbol, it is determined as information 1.

This is because the order of symbols can be reconstructed by an interleaver. The interleaver may be applied in units of specific symbol numbers below the packet unit.

In addition, not only two symbols but also n can be extended as follows. 11 illustrates various examples of a symbol repetition technique of repeating n symbols according to the present embodiment.

Option 1: Information 0 and information 1 may be repeatedly represented by the same symbol n times as shown in FIG.

-Information 0-> 0 0 ... 0 (repeat info 0 n times)

-Information 1-> 1 1 ... 1 (repeat information 1 n times)

* Option 2: As shown in FIG. 11, information 0 and information 1 may be repeatedly represented by different symbols n times.

-Information 0-> 0 1 0 1 ... or 1 0 1 0 ... (repeat info 0 and info 1 n times with each other)

-Info 1-> 1 0 1 0 ... or 0 1 0 1 ... (repeat info 1 and info 0 n times with each other)

Option 3: As shown in FIG. 11, one half of a symbol may be configured as information 0 and the other half may be configured as information 1 to represent n symbols.

-Information 0-> 0 0 ... 1 1 ... or 1 1 ... 0 0 ... (n / 2 symbols consist of information 0, the remaining n / 2 symbols consist of information 1 do)

-Information 1-> 1 1 ... 0 0 ... or 0 0 ... 1 1 ... (n / 2 symbols consist of information 0, the remaining n / 2 symbols consist of information 1 do)

Option 4: When n is odd, as shown in FIG. 11, n symbols may be represented by dividing the number of symbols 1 (symbol including information 1) and the number of symbols 0 (symbol including information 0).

Information 0-> n symbols where the number of symbols 1 is odd and the number of symbols 0 is even, or n symbols where the number of symbols 1 is even and the number of symbols 0 is odd

Information 1-> n symbols where the number of symbols 0 is odd and the number of symbols 1 is even, or n symbols where the number of symbols 0 is even and the number of symbols 1 is odd

In addition, the order of symbols may be reconstructed by the interleaver. The interleaver may be applied in units of packets and specific symbols.

In addition, as described above, the receiving apparatus may determine whether the information is 0 or 1 by determining the threshold value and comparing the powers of the n symbols.

However, the use of consecutive symbol 0 (or off symbol) may cause a coexistence problem with an existing Wi-Fi device and / or another device. The coexistence problem is a problem caused by transmitting a signal by determining that another device is a channel idle state due to a continuous off symbol. Thus, the option 2 scheme may be preferred as it is desirable to avoid the use of consecutive off symbols to solve the leveling problem.

In addition, it can be extended in a manner of expressing m information using n symbols. In this case, the first or last m is represented by 0 (OFF) or 1 (ON) symbols depending on the information, and the nm or 0 (OFF) or 1 (ON) redundant symbols are formed consecutively before or after. can do.

For example, if a code rate of 3/4 is applied to the information 010, it may be 1,010 or 010,1 or 0,010 or 010,0. However, in order to prevent the use of consecutive off symbols, it may be desirable to apply a code rate of 1/2 or less.

In this embodiment as well, the order of symbols can be reconstructed by the interleaver. The interleaver may be applied in units of packets and specific symbols.

Hereinafter, various embodiments of a symbol to which the symbol repetition technique is applied will be described.

In general, a symbol to which the symbol repetition technique is applied may be represented by n (CP + 3.2us) or CP + n (1.6us).

As shown in FIG. 11, n (n> = 2) information signals (symbols) are used to represent 1 bit, and a specific sequence is applied to all available subcarriers (for example, 13). Form a signal (symbol).

Different sequences may be applied to the available subcarriers according to the 3.2us on signal and the 3.2us off signal. A 3.2us off signal can be generated by applying all coefficients to zero.

CP may be used by adopting a specific length from the rear of the information signal 3.2us immediately behind. At this time, CP may be 0.4us or 0.8us. This length is equal to the guard interval of 802.11ac.

Accordingly, 1 bit information corresponding to a symbol to which a general symbol repetition technique is applied may be represented as shown in the following table.

Information ‘0’	Information ‘1’
3.2us OFF-signal	3.2us ON-signal
Or two specific consecutive signals are 3.2us ON-signal + 3.2us OFF-signal, all other signals are ON or all OFF	Or two specific consecutive signals: 3.2us OFF-signal + 3.2us ON-signal, all other signals ON or all OFF
Or two specific consecutive signals: 3.2us OFF-signal + 3.2us ON-signal, all other signals ON or all OFF	Or two specific consecutive signals are 3.2us ON-signal + 3.2us OFF-signal, all other signals are ON or all OFF
Or a certain number (or ceil (n / 2) or floor (n / 2)) placed in a specific position is 3.2us OFF-signal, 3.2us ON-signalEx) ON + OFF + ON + OFF…	Or a certain number (or ceil (n / 2) or floor (n / 2)) placed in a specific position is 3.2us ON-signal, 3.2us OFF-signalEx) OFF + ON + OFF + ON + OFF…

Table 4 does not indicate CP separately. In fact, n pieces (CP + 3.2us) including CPs or CP + n pieces (3.2us) may indicate one 1-bit information. That is, in the case of n (CP + 3.2us), the 3.2us on signal may be viewed as a (CP + 3.2us) on signal, and the 3.2us off signal may be viewed as a (CP + 3.2us) off signal.

As another example, a symbol to which the symbol repetition technique is applied may be represented as CP + 3.2us + CP + 3.2us or CP + 3.2us + 3.2us.

According to the above embodiment, two information signals (symbols) are used to represent one bit and a specific sequence is applied to all available subcarriers (for example, thirteen), and then IFFT is taken to generate an information signal (symbol) of 3.2us. Form.

Different sequences may be applied to the available subcarriers according to the 3.2us on signal and the 3.2us off signal. A 3.2us off signal can be generated by applying all coefficients to zero.

CP may be used by adopting a specific length from the rear of the information signal 3.2us immediately behind. At this time, CP may be 0.4us or 0.8us. This length is equal to the guard interval of 802.11ac.

Accordingly, one bit information corresponding to a symbol to which the symbol repetition technique is applied may be represented as shown in the following table.

Information ‘0’	Information ‘1’
3.2us OFF-signal	3.2us ON-signal + 3.2us ON-signal
Or 3.2us ON-signal + 3.2us OFF-signal	Or 3.2us OFF-signal + 3.2us ON-signal
Or 3.2us OFF-signal + 3.2us ON-signal	Or 3.2us ON-signal + 3.2us OFF-signal

Table 5 does not indicate CP separately. Indeed, CP + 3.2us + CP + 3.2us or CP + 3.2us + 3.2us, including CP, may point to one 1-bit information. That is, in the case of CP + 3.2us + CP + 3.2us, the 3.2us on signal can be viewed as a (CP + 3.2us) on signal, and the 3.2us off signal can be viewed as a (CP + 3.2us) off signal. .

As another example, a symbol to which the symbol repetition technique is applied may be represented as CP + 3.2us + CP + 3.2us + CP + 3.2us or CP + 3.2us + 3.2us + 3.2us.

According to the above embodiment, three information signals (symbols) are used to represent one bit and a specific sequence is applied to all available subcarriers (eg, thirteen), and then IFFT is taken to generate an information signal (symbol) of 3.2us. Form.

Different sequences may be applied to the available subcarriers according to the 3.2us on signal and the 3.2us off signal. A 3.2us off signal can be generated by applying all coefficients to zero.

CP may be used by adopting a specific length from the rear of the information signal 3.2us immediately behind. At this time, CP may be 0.4us or 0.8us. This length is equal to the guard interval of 802.11ac.

Accordingly, one bit information corresponding to a symbol to which the symbol repetition technique is applied may be represented as shown in the following table.

Information ‘0’	Information ‘1’
3.2us OFF-signal + 3.2us OFF-signal + 3.2us OFF-signal	3.2us ON-signal + 3.2us ON-signal + 3.2us ON-signal
Or 3.2us ON-signal + 3.2us ON-signal + 3.2us OFF-signal	Or 3.2us ON-signal + 3.2us OFF-signal + 3.2us ON-signal
Or 3.2us ON-signal + 3.2us OFF-signal + 3.2us ON-signal	Or 3.2us ON-signal + 3.2us ON-signal + 3.2us OFF-signal
Or 3.2us ON-signal + 3.2us OFF-signal + 3.2us ON-signal	Or 3.2us OFF-signal + 3.2us ON-signal + 3.2us OFF-signal

Table 6 does not indicate CP separately. In fact, CP + 3.2us + CP + 3.2us + CP + 3.2us or CP + 3.2us + 3.2us + 3.2us, including CP, may point to one 1-bit information. That is, in the case of CP + 3.2us + CP + 3.2us + CP + 3.2us, the 3.2us on signal can be viewed as a (CP + 3.2us) on signal, and the 3.2us off signal is a (CP + 3.2us) off It can be seen as a signal.

As another example, a symbol to which the symbol repetition technique is applied may be represented as CP + 3.2us + CP + 3.2us + CP + 3.2us + CP + 3.2us or CP + 3.2us + 3.2us + 3.2us + 3.2us.

According to the above embodiment, four information signals (symbols) are used to represent one bit and a specific sequence is applied to all available subcarriers (for example, thirteen), and then IFFT is taken to generate an information signal (symbol) of 3.2us. Form.

Different sequences may be applied to the available subcarriers according to the 3.2us on signal and the 3.2us off signal. A 3.2us off signal can be generated by applying all coefficients to zero.

CP may be used by adopting a specific length from the rear of the information signal 3.2us immediately behind. At this time, CP may be 0.4us or 0.8us. This length is equal to the guard interval of 802.11ac.

Accordingly, one bit information corresponding to a symbol to which the symbol repetition technique is applied may be represented as shown in the following table.

Information ‘0’	Information ‘1’
3.2us OFF-signal + 3.2us OFF-signal + 3.2us OFF-signal + 3.2us OFF-signal	3.2us ON-signal + 3.2us ON-signal + 3.2us ON-signal + 3.2us ON-signal
Or 3.2us ON-signal + 3.2us ON-signal + 3.2us ON-signal + 3.2us OFF-signal	Or 3.2us ON-signal + 3.2us ON-signal + 3.2us OFF-signal + 3.2us ON-signal
Or 3.2us ON-signal + 3.2us ON-signal + 3.2us OFF-signal + 3.2us ON-signal	Or 3.2us ON-signal + 3.2us ON-signal + 3.2us ON-signal + 3.2us OFF-signal
Or 3.2us ON-signal + 3.2us ON-signal + 3.2us OFF-signal + 3.2us ON-signal	Or 3.2us ON-signal + 3.2us OFF-signal + 3.2us ON-signal + 3.2us ON-signal
Or 3.2us ON-signal + 3.2us OFF-signal + 3.2us ON-signal + 3.2us ON-signal	Or 3.2us ON-signal + 3.2us ON-signal + 3.2us OFF-signal + 3.2us ON-signal
Or 3.2us ON-signal + 3.2us OFF-signal + 3.2us ON-signal + 3.2us OFF-signal	Or 3.2us OFF-signal + 3.2us ON-signal + 3.2us OFF-signal + 3.2us ON-signal
Or 3.2us ON-signal + 3.2us ON-signal + 3.2us OFF-signal + 3.2us OFF-signal	Or 3.2us OFF-signal + 3.2us OFF-signal + 3.2us ON-signal + 3.2us ON-signal
Or 3.2us OFF-signal + 3.2us OFF-signal + 3.2us ON-signal + 3.2us ON-signal	Or 3.2us ON-signal + 3.2us ON-signal + 3.2us OFF-signal + 3.2us OFF-signal

Table 7 does not indicate CP separately. Indeed, CP + 3.2us + CP + 3.2us + CP + 3.2us + CP + 3.2us or CP + 3.2us + 3.2us + 3.2us + 3.2us, including CP, may point to one single bit of information. That is, in the case of CP + 3.2us + CP + 3.2us + CP + 3.2us + CP + 3.2us, the 3.2us on signal can be regarded as (CP + 3.2us) on signal and the 3.2us off signal is (CP + 3.2us) off signal.

As another example, a symbol to which Manchester coding is applied based on symbol repetition may be represented by n (CP + 1.6us + CP + 1.6us) or CP + n (1.6us + 1.6us).

According to the above embodiment, a symbol is repeated using a symbol repeated n (> = 2) times, and a specific sequence is applied to all available subcarriers (for example, 13), and the remaining coefficients of 0 are applied. By setting IFFT, 3.2us of signal with 1.6us period is generated. Take one of these and set it as a 1.6us information signal (symbol).

The sub information may be called a 1.6us information signal. The 1.6us information signal may be a 1.6us on signal or a 1.6 off signal. The 1.6us on signal and the 1.6 off signal may have different sequences applied to each subcarrier. The 1.6us off signal can be generated by applying all coefficients to zero.

CP can be used by adopting a specific length from the back of the 1.6us of the information signal immediately after. At this time, CP may be 0.4us or 0.8us. This length is equal to the guard interval of 802.11ac.

Accordingly, 1 bit information corresponding to a symbol to which Manchester coding is applied based on the symbol repetition may be represented as shown in the following table.

Information ‘0’	Information ‘1’
(1.6us ON-signal + 1.6us OFF-signal) Repeat n times	(1.6us OFF-signal + 1.6us ON-signal) Repeat n times
Or (1.6us OFF-signal + 1.6us ON-signal) n times	Or (1.6us ON-signal + 1.6us OFF-signal) n times
(1.6us ON-signal + 1.6us OFF-signal) + (1.6us OFF-signal + 1.6us ON-signal) floor (n / 2) Repeat + if necessary (1.6us ON-signal + 1.6us OFF-signal)	(1.6us OFF-signal + 1.6us ON-signal) + (1.6us ON-signal + 1.6us OFF-signal) floor (n / 2) Repeat + if necessary (1.6us OFF-signal + 1.6us ON-signal)
(1.6us OFF-signal + 1.6us ON-signal) + (1.6us ON-signal + 1.6us OFF-signal) floor (n / 2) Repeat + if necessary (1.6us OFF-signal + 1.6us ON-signal)	(1.6us ON-signal + 1.6us OFF-signal) + (1.6us OFF-signal + 1.6us ON-signal) floor (n / 2) Repeat + if necessary (1.6us ON-signal + 1.6us OFF-signal)

Table 8 does not indicate CP separately. In fact, n (CP + 1.6us + CP + 1.6us) or CP + n (1.6us + 1.6us) including CP may indicate one 1-bit information. That is, in the case of n (CP + 1.6us + CP + 1.6us), the 1.6us on signal can be viewed as a (CP + 1.6us) on signal, and the 1.6us off signal is a (CP + 1.6us) off signal. Can be seen as.

Like the embodiments described above, the symbol repetition technique can satisfy the range requirement of low power wake-up communication. When only the OOK method is applied, the data rate for one symbol is 250 Kbps (4us). At this time, if the symbol is repeated twice using the symbol repetition technique, the data rate may be 125 Kbps (8us), and if the fourth repetition is performed, the data rate may be 62.5 Kbps (16us), and if the eight times are repeated, the data rate may be 31.25Kbps (32us). have. In the case of low-power wake-up communication, if there is no BCC, the symbol needs to be repeated eight times to satisfy the range requirement.

Hereinafter, various embodiments of a symbol to which a symbol reduction technique is applied among symbol types that can be used in the WUR will be described.

12 illustrates various examples of a symbol reduction technique according to the present embodiment.

According to the embodiment of FIG. 12, as the m value increases, the symbol is further reduced to reduce the length of a symbol carrying one piece of information. When m = 2, the length of a symbol carrying one piece of information becomes CP + 1.6us. When m = 4, the length of a symbol carrying one piece of information becomes CP + 0.8us. When m = 8, the length of a symbol carrying one piece of information becomes CP + 0.4us.

As the length of the symbol decreases, a high data rate can be ensured. If only the OOK scheme is applied, the data rate for one symbol is 250 Kbps (4us). In this case, using m = 2, the data rate is 500 Kbps (2us), if m = 4, the data rate is 1Mbps (1us), if m = 8 data rate can be 2Mbps (0.5us). .

For example, a symbol to which a symbol reduction technique is generally applied may be represented by CP + 3.2us / m (m = 2,4,8,16,32, ...) (option 1).

As shown in option 1 of FIG. 12, a symbol using a symbol reduction technique is used to represent one bit, and a specific sequence is applied to every available subcarrier (for example, 13) in m units, and the remaining coefficients are set to zero. do. Subsequently, if IFFT is applied to the subcarrier to which the specific sequence is applied, a 3.2us signal having a 3.2us / m period is generated. Take one of these and map it to the 3.2us / m information signal (information 1).

For example, if a specific sequence is applied in units of two columns (m = 2) to 13 subcarriers, the on signal may be configured as follows.

On signal (information 1): {a 0 b 0 c 0 d 0 e 0 f 0 g} or {0 a 0 b 0 c 0 d 0 e 0 f 0}, where a, b, c, d, e, f, g is 1 or -1.

As another example, if a specific sequence is applied to 13 subcarriers in units of four squares (m = 4), the on signal may be configured as follows.

On signal (Info 1): {a 0 0 0 b 0 0 0 c 0 0 0 d} or {0 a 0 0 0 b 0 0 0 c 0 0 0} or {0 0 a 0 0 0 b 0 0 0 c 0 0} or {0 0 0 a 0 0 0 b 0 0 0 c 0} or {0 0 a 0 0 0 0 0 0 0 b 0 0}, where a, b, c, d is 1 or -1.

As another example, if a specific sequence is applied to 13 subcarriers by 8 columns (m = 8), the on signal may be configured as follows.

On signal (information 1): {a 0 0 0 0 0 0 0 b 0 0 0 0} or {0 a 0 0 0 0 0 0 0 b 0 0 0} or {0 0 a 0 0 0 0 0 0 0 0 b 0 0} or {0 0 0 a 0 0 0 0 0 0 0 b 0}, or {0 0 0 0 a 0 0 0 0 0 0 0 b}, where a and b are 1 or -1 .

The 3.2us / m information signal is divided into a 3.2us / m on signal and a 3.2us / m off signal. In addition, different sequences may be applied to the (usable) subcarriers for the 3.2us / m on signal and the 3.2us / m off signal, respectively. A 3.2us / m off signal can be generated by applying all coefficients to zero.

CP can be used by adopting a specific length from the back of the information signal 3.2us / m immediately behind. At this time, CP may be 0.4us or 0.8us. This length is equal to the guard interval of 802.11ac. However, when m = 8, the CP cannot be 0.8us. Alternatively, CP may be 0.1us or 0.2us, or another value.

Accordingly, 1 bit information corresponding to a symbol to which a general symbol reduction technique is applied may be represented as shown in the following table.

Information ‘0’	Information ‘1’
3.2us / m OFF-signal	3.2us / m ON-signal

CP is not shown separately in Table 9. In fact, CP + 3.2us / m including CP may indicate one 1-bit information. That is, the 3.2us / m on signal may be viewed as a CP + 3.2us / m on signal, and the 3.2us / m off signal may be viewed as a CP + 3.2us / m off signal.

As another example, a symbol to which the symbol reduction technique is applied may be represented by CP + 3.2us / m + CP + 3.2us / m (m = 2,4,8) (option 2).

In the OOK transmission using the Wi-Fi transmitter, the time used for transmitting one bit (or symbol) except for the guard interval of the transmission signal is 3.2 us. At this time, if the symbol reduction technique is applied, the time used for one bit transmission is 3.2us / m. However, in this embodiment, the time used for transmitting one bit is repeated as 3.2us / m + 3.2us / m by repeating a symbol to which the symbol reduction technique is applied, and the signal between 3.2us / m signals is also used by using the characteristics of Manchester coding. A transition in size was allowed to occur. That is, each sub-information having a length of 3.2us / m should have a value of 0 or 1, and may configure a signal in the following manner.

* Information 0-> 1 0 (each can be referred to as sub information 1 0 or sub symbol 1 (ON) 0 (OFF))

First 3.2us / m signal (sub-information 1 or sub-symbol 1): A specific sequence in m-column for all available subcarriers (e.g. 13 subcarriers) to generate symbols with symbol reduction Apply. That is, in a particular sequence, coefficients may exist at intervals of m columns.

The transmitter maps a specific sequence to K consecutive subcarriers of 64 subcarriers and sets a coefficient to 0 for the remaining subcarriers to perform IFFT. In this way, signals in the time domain can be generated. Since the signal in the time domain has coefficients at intervals of m in the frequency domain, a 3.2us signal having a 3.2us / m period is generated. You can take one of these and use it as a 3.2us / m on signal (sub information 1).

Second 3.2us / m signal (sub information 0 or subsymbol 0): As with the first 3.2us / m signal, the transmitter maps a particular sequence to K consecutive subcarriers of 64 subcarriers, Can be generated to generate a time domain signal. The sub information 0 may correspond to a 3.2 us / m off signal. The 3.2us / m off signal can be generated by setting all coefficients to zero.

One of the first or second 3.2us / m periodic signals of the signal in the time domain may be selected and used as the sub information 0.

* Information 1-> 0 1 (each can be referred to as sub information '0', '1' or sub symbol 0 (OFF) 1 (ON))

-Since information 1 is also divided into the first 3.2us / m signal (sub information 0) and the second 3.2us / m signal (sub information 1), the signal corresponding to each sub information is generated in the same way as information 0 is generated. Can be configured.

In addition, information 0 may be configured as 01 and information 1 may be configured as 10.

As in option 2 of FIG. 12, 1-bit information corresponding to a symbol to which a symbol reduction technique is applied may be represented as shown in the following table.

Information ‘0’	Information ‘1’
3.2us / m OFF-signal + 3.2us / m ON-signal or 3.2us / m ON-signal + 3.2us / m OFF-signal	3.2us / m ON-signal + 3.2us / m OFF-signal or 3.2us / m OFF-signal + 3.2us / m ON-signal

In Table 10, CP is not separately indicated. In fact, CP + 3.2us / m including CP may indicate one 1-bit information. That is, the 3.2us / m on signal may be viewed as a CP + 3.2us / m on signal, and the 3.2us / m off signal may be viewed as a CP + 3.2us / m off signal.

Embodiments illustrated by option 1 and option 2 of FIG. 12 may be generalized as shown in the following table.

	Information ‘0’	Information ‘1’
Option 1 (m = 2,4,8)	2us OFF-signal	2us ON-signal
	1us OFF-signal	1us ON-signal
	0.5us OFF-signal	0.5us ON-signal
Option 2 (m = 4,8)	1us OFF-signal + 1us ON-signal or 1us ON-signal + 1us OFF-signal	1us ON-signal + 1us OFF-signal or 1us OFF-signal + 1us ON-signal
	0.5us OFF-signal + 0.5us ON-signal or 0.5us ON-signal + 0.5us OFF-signal	0.5us ON-signal + 0.5us OFF-signal or 0.5us OFF-signal + 0.5us ON-signal

Table 11 shows each signal in length including CP. That is, CP + 3.2us / m including the CP may indicate one 1-bit information.

For example, when m = 4 in Option 2, a symbol carrying one piece of information becomes CP + 0.8us, and thus a 1us off signal or 1us on signal is composed of a CP (0.2us) + 0.8us signal. In Option 2, since Manchester coding is applied and symbols are repeated, the data rate of one information can be 500 Kbps when m = 4.

As another example, when m = 8 in Option 2, the length of a symbol carrying one piece of information becomes CP + 0.4us, so that the 0.5us off signal or the 0.5us on signal consists of a CP (0.1us) + 0.4us signal. In Option 2, since Manchester coding is applied and symbols are repeated, the data rate of one information may be 1Mbps when m = 8.

In the table below, the data rates that can be secured through the above-described embodiments are compared with each other.

CP	Default symbol (Example 1) (CP + 3.2us)	Man. Symbol (Example 2) (CP + 1.6 + CP + 1.6)	Man. Symbol (Example 3) (CP + 1.6 + 1.6)
0.4us	277.8	250.0	277.8
0.8us	250.0	208.3	250.0

CP	Symbol rep.n (CP + 3.2us)			CP rep.CP + n (3.2us)			Man. symbol rep.n (CP + 1.6us + CP + 1.6us)
	n = 2 (Example 4)	n = 3 (Example 5)	n = 4 (Example 6)	n = 2 (Example 7)	n = 3 (Example 8)	n = 4 (Example 9)	n = 2 (Example 10)	n = 3 (Example 11)	n = 4 (Example 12)
0.4us	138.9	92.6	69.4	147.1	100.0	75.8	125.0	83.3	62.5
0.8us	125.0	83.3	62.5	138.9	96.2	73.5	104.2	69.4	52.1

CP	Man. CP + n symbols (1.6us + 1.6us)			Symbol reductionCP + 3.2us / m
	n = 2 (Example 13)	n = 3 (Example 14)	n = 4 (Example 15)	m = 2 (Example 16)	m = 4 (Example 17)	m = 8 (Example 18)
0.4us	147.1	100.0	75.8	500.0	833.3	1250.0
0.8us	138.9	96.2	73.5	416.7	625.0	NA

CP	Symbol reductionCP + 3.2us / m		Man. symbol rep. w / Man.CP + 3.2us / m + CP + 3.2us / m
	m = 4	m = 8	m = 4	m = 8
0.1us	1111.1	2000	555.6	1000
0.2us	1000	1666.7	500	833.3

Hereinafter, the present specification proposes a method of configuring a wakeup preamble in a wakeup packet.

The wakeup preamble signals the start of a wakeup packet. In addition, the wake-up preamble may be used for acquiring channel information to help with timing synchronization, signal detection, and the like, for receiving a received power and for identifying an identifier.

The transmitter can adjust time synchronization by applying auto-correlation using the periodicity of the wake-up preamble. Alternatively, the transmitting apparatus may have information about the sequence itself so that time synchronization may be applied by applying cross-correlation.

In this specification, a preamble sequence (or wake-up preamble sequence) having periodicity for time synchronization is proposed. In addition, the channel information acquisition, the received power measurement, and the identifier point of view together with the preamble sequence are considered.

The preamble sequence may use a general on (1) symbol, an off (0) symbol, or a symbol to which Manchester coding is applied. Regardless of whether coding is applied to the payload in which the actual MAC address is transmitted, the preamble may use a general OOK and a OOK to which Manchester coding is applied. When Manchester coding is applied, information 1 may be configured in an order of off sub-symbol and on sub-symbol, and information 0 may be configured in an order of on sub-symbol and off sub-symbol. . In this case, the length of the symbol may be 4us (3.2us except for CP), and the length of the sub-symbol may be 2us (1.6us except for CP).

The start of the wake up preamble starts with an on symbol or an on sub symbol. This is because it is advantageous to synchronize with the notification of the signal coming in.

The preamble sequence may consist of a sequence having periodicity to match time synchronization.

When the general OOK method is applied, the information of each symbol may be set to have a periodicity, thereby making a preamble having a periodicity of 4 us multiples. If the period is 1, all of them can be configured as a preamble sequence of 1. That is, when the period is 1, the preamble sequence may consist of {1,1,1, ..., 1}. If the period is 2 or more, a multiple of the period may be configured as a preamble sequence in which zero remainder is one. That is, when the period is 2, the preamble sequence may consist of {1,0,1,0, ..., 1,0}.

In addition, the transmitter may make the symbol itself having information 1 into a signal having periodicity. That is, the transmitter may generate a signal having a period of 1.6 us or 0.8 us except for CP by inserting coefficients in units of two or four spaces in the subcarrier. Considering the CP, it may be desirable to make a signal with a 0.8us period. A signal having a 1.6us or 0.8us period may generate 32 or 16 samples when sampling at a receiver, and each may be mapped to a Golay sequence of 802.11ad and a new Golay sequence or Barker sequence. In the transmitting apparatus, a time signal of a Golay sequence or a Barker sequence may be simply transmitted without using OFDM.

When Manchester coding is applied together with the OOK method, periodicity may be made in units of multiples of symbols in a manner similar to that of the general OOK method described above.

In addition, the transmitter may make a signal having a periodicity in the on sub-symbol itself. That is, the transmitter can generate a signal having a 0.4us or 0.8us period except for the CP. If the CP is 0.8us before the symbol, it may be desirable to make a signal with a 0.4us or 0.8us period. If the CP is 0.4us each located before the sub-symbol, it may be desirable to make a signal having a 0.4us period. The 0.4us or 0.8us signal can generate 8 or 16 samples when sampling at the receiver and can be mapped to specific Golay sequences and Barker sequences, respectively. The transmitter may transmit a time signal of a Golay sequence or a Barker sequence without using OFDM.

Hereinafter, it is assumed that the period is a multiple unit of a symbol to propose a preamble sequence. However, the preamble sequence proposed in the above assumption may also be applied to a case having a period less than a symbol unit.

The preamble sequence consists of a periodic sequence having a periodicity and an additional symbol indicating the end of the periodic sequence (which may be referred to as a sync sequence), a channel estimation and symbols for a received power measurement symbol and an identifier. Can be.

In the synchronization sequence, the symbol for the end of the period is overhead, but can help with time synchronization performance. By simply repeating the off symbol several times, the end of the wake-up preamble can be signaled, or a single off symbol can be placed to achieve a performance gain when timing synchronization with autocorrelation calculations.

When the general OOK method is applied, if the period is p symbols (P * 4us) and there are m consecutive off symbols in the p symbol, the symbol indicating the end of the periodic sequence may consist of m + 1 off symbols or 1 off symbol. Can be. For example, if the period is one symbol, the synchronization sequence may be {1,1,1,1, ..., 1,0}. If the end of the sequence having periodicity is composed of off symbols, overhead may be reduced by configuring m + 1 off symbols including the off symbols. Or, no additional off symbol may be needed.

When Manchester coding is applied together with the OOK scheme, when the period of the preamble sequence is one symbol, an end signal may be configured as one off symbol, and when the period is 2 or more, one or two off symbols may be used. can do. When the end of the periodic sequence is announced simply by repetition of the off symbol, the off period in the sequence having periodicity is only 4us at most when two or more symbols are present. Therefore, the off symbol to be inserted can reduce the overhead compared to the general OOK method if only two off symbols are required.

When a general OOK scheme is applied to not only a wakeup preamble but also a wakeup packet, time synchronization, channel estimation, and received signal power calculation may be simultaneously performed using a periodic sequence. In this case, in consideration of overhead or performance, it may be advantageous to construct a synchronization sequence using a sequence having a period of 1 and a symbol indicating the end thereof as shown below.

Sync sequence: {1,1,1,1,1,1, ..., 1,0}

When the general OOK scheme is applied to the wakeup packet as well as the wakeup preamble, there may be a case that a symbol indicating the end of the periodic sequence is not needed. This may be the case that the p th (p: period) symbol is off (0) after the periodic sequence in view of autocorrelation calculation to match time synchronization. Alternatively, when the time signal for information 1 used in the signal following the periodic sequence and the synchronization sequence is different, a symbol indicating the end may not be needed.

When the general OOK scheme is applied to the wake-up preamble, it may be desirable to avoid the following cases from the viewpoint of autocorrelation calculation for time synchronization.

If there is no symbol for the end of the periodic sequence: the n-pth symbol is off (0)

If there is a symbol for the end of the periodic sequence: n-p-1th symbol is off (0)

n: (symbol) length of sync sequence

Considering various periods, lengths of synchronization sequences, and synchronization performances, it may be advantageous to construct a synchronization sequence as follows.

Sync sequence: {1,1,1,1,1,1, ..., 1,0}

When simultaneously performing time synchronization, channel estimation, and received power calculation using the synchronization sequence, Manchester coding may be applied to not only the wakeup preamble but also the wakeup packet. At this time, the channel estimation and the received power calculation are not affected according to the period (compared to the general OOK method consisting of all symbols only, the OOK with Manchester coding may be disadvantageous in the channel estimation and the received signal power calculation. However, due to signal characteristics, this estimation may not be necessary). However, when the end of the cycle sequence is simply announced by repetition of off symbols, it may be preferable to use a sequence having a period of 1 as in the general OOK scheme in consideration of the overhead.

In addition, the wake-up preamble is applied to the general OOK scheme (except that the identifier symbol is Manchester coding), and when Manchester coding is applied to the wake-up payload, channel estimation and calculation of received signal power are required in the wake-up preamble. It may not be.

In contrast, when the wake-up preamble is applied to the Manchester coding scheme and the wake-up payload is applied to the general OOK, performance degradation may occur due to disadvantages such as channel estimation and received signal power calculation in the wake-up preamble.

In the following, a method of constructing a synchronization sequence is proposed. First, it is assumed that all packets as well as the wake-up preamble apply a general OOK. It is also assumed that time synchronization is performed by applying autocorrelation. In addition, it is assumed that all time signals corresponding to each information in all packets are the same. However, even in the situation where the above assumption does not apply, the following method of configuring a synchronization sequence may be used.

Step 1: Determine the length and period of the cycle sequence (determine n and p).

Step 2: Determine the cycle sequence.

Step 3: Determine whether the pth symbol is off at the end of the cycle sequence, and insert an off symbol to indicate the end of the cycle (if the last symbol of the cycle sequence is off, replace it with a symbol to end the cycle).

Step 4: If the n-pth symbol (if no symbol indicates the end of the cycle sequence) or the n-p-1th symbol (if there is a symbol indicating the end of the cycle sequence) goes off to step 2 to determine a new sequence. If a new sequence cannot be set, go to step 1 and re-determine the period sequence length and period (n is the length of the period sequence).

Hereinafter, various embodiments of time synchronization using autocorrelation will be described.

In this embodiment, it is assumed that the general OOK is applied to all packets as well as the wake-up preamble. In addition, it is assumed that all time signals corresponding to each information in all packets are the same.

13 shows an example of time synchronization using autocorrelation according to the present embodiment.

13 shows a case where the period is 1 and the length of the synchronization sequence is 6. If there is no off symbol indicating the end of the periodic sequence as shown in Figure 13, the synchronization sequence may be {1, 1, 1, 1, 1, 1}. In addition, the window size is 24 us (6 symbols = length of the synchronization sequence). The magnitude of the two autocorrelated signals is 20us (5 symbols = length of sync sequence minus period). In this case, the autocorrelation of two time signals having a difference of 4 us (one symbol and the same period) is obtained, and the case having the largest value is used as the start point of the packet.

Referring to FIG. 13, the auto-correlation value is obtained by varying a starting point of a signal A and a signal B having a size of 20 us within a predetermined period. In this case, if the first symbol of the following signal (wake-up payload) is 1, the start point of signal A may have the same correlation value from 0us to 4us (the start point of signal B is 4us to 8us), which may be difficult to synchronize.

14 shows another example of time synchronization using autocorrelation according to the present embodiment.

14 shows the case where the period is 1 and the length of the synchronization sequence is 6. If there is an off symbol indicating the end of the periodic sequence as shown in Figure 14, the synchronization sequence may be {1, 1, 1, 1, 1, 0}. In addition, the window size is 20 us (5 symbols = length of the sync sequence minus 1 symbol). The magnitude of the two autocorrelated signals is 16us (4 symbols = length of sync sequence minus period-1). In this case, the autocorrelation of two time signals having a difference of 4 us (one symbol and the same period) is obtained, and the case having the largest value is used as the start point of the packet.

Referring to FIG. 14, auto-correlation values are obtained by varying a starting point of a signal A and a signal B having a size of 16 us within a predetermined period. In this case, since the last symbol of the wake-up preamble is off, the maximum correlation value is obtained at the start point of the signal A at 0us (the start point of the signal B is 4us) regardless of the first symbol of the following signal (wakeup payload). Therefore, time synchronization is good.

15 shows another example of time synchronization using autocorrelation according to the present embodiment.

15 shows the case where the period is 1 and the length of the synchronization sequence is 6 (p = 1, n = 6). If there is no off symbol indicating the end of the periodic sequence as shown in Figure 15, the synchronization sequence may be {1, 1, 1, 1, 1, 1}. In addition, the window size is 24 us (6 symbols = length of the synchronization sequence). The magnitude of the two autocorrelated signals is 20us (5 symbols = length of sync sequence minus period). In this case, the autocorrelation of two time signals having a difference of 4 us (one symbol and the same period) is obtained, and the case having the largest value is used as the start point of the packet.

Referring to FIG. 15, auto-correlation values are obtained by varying a starting point of a signal A and a signal B having a size of 20 us within a predetermined period. In this case, the autocorrelation of signal A '' and signal B '' should be the maximum value. However, the same autocorrelation results from signal A '' 'and signal B' ''. Furthermore, considering the noise and channel conditions, it may be difficult to synchronize because the timing offset may be pushed down by one symbol.

16 shows another example of time synchronization using autocorrelation according to the present embodiment.

FIG. 16 shows the case where the period is 1 and the length of the synchronization sequence is 6 (p = 1, n = 6). If there is an off symbol indicating the end of the periodic sequence as shown in Figure 16, the synchronization sequence may be {1, 1, 1, 1, 1, 0}. In addition, the window size is 20 us (5 symbols = length of the sync sequence minus 1 symbol). The magnitude of the two autocorrelated signals is 16us (4 symbols = length of sync sequence minus period-1). In this case, the autocorrelation of two time signals having a difference of 4 us (one symbol and the same period) is obtained, and the case having the largest value is used as the start point of the packet.

Referring to FIG. 16, auto-correlation values are obtained by varying a starting point of a signal A and a signal B having a size of 16 us within a predetermined period. In this case, the last symbol of the wake-up preamble is off (additional off symbol), so the autocorrelation between signal A '' and signal B '' has a maximum value (at the start of signal A '' at 0us, The starting point will have a maximum correlation at 4us). Therefore, it is good to synchronize time using this signal.

17 shows another example of time synchronization using autocorrelation according to the present embodiment.

17 shows the case where the period is 2 and the length of the synchronization sequence is 7 (p = 2, n = 7). 17 shows an example in which there is no need to have an off symbol indicating the end of a periodic sequence. In this case, the synchronization sequence may be {1,0,1,0,1, 0, 1}. That is, the p th symbol from the back is an off symbol (p = 2). In addition, the window size is 28 us (7 symbols = length of the synchronization sequence). The magnitude of the two autocorrelated signals is 20us (5 symbols = length of sync sequence minus period). In this case, the autocorrelation of two time signals having a difference of 8 us (two symbols and the same period) is obtained, and the case having the largest value is used as the start point of the packet.

Referring to FIG. 17, a start point of a signal A and a signal B having a size of 20 us is varied within a predetermined period to obtain an autocorrelation value. In this case, the autocorrelation of signal A '' and signal B '' has a maximum value (the starting point of signal A '' has a maximum correlation value at 4us and the starting point of signal B '' has a maximum correlation value at 8us). In this case, even if there is no additional off symbol, the symbol corresponding to the first sequence following the signal A '' at the time of autocorrelation is always smaller than the maximum value of the autocorrelation because the p-th (p = 2) off symbol. Therefore, it is good to synchronize time using this signal.

18 shows another example of time synchronization using autocorrelation according to the present embodiment.

18 shows the case where the period is 3 and the length of the synchronization sequence is 8 (p = 3, n = 8). 18 is an example of a sequence to be avoided when there is no off symbol indicating the end of a periodic sequence. In this case, the synchronization sequence may be {1,0,0,1, 0, 0,1,0}. The preceding npth symbol is an off symbol (np = 5). In addition, since the p th symbol from the back is an off symbol (p = 3), no additional off symbol is required. The window size is also 32 us (8 symbols = length of the sync sequence). The magnitude of the two autocorrelated signals is 20us (5 symbols = length of sync sequence minus period). In this case, the autocorrelation of two time signals having a difference of 12us (3 symbols and the same period) is obtained, and the case having the largest value is set as the start point of the packet.

Referring to FIG. 18, auto-correlation values are obtained by varying a starting point of a signal A and a signal B having a size of 20 us within a predetermined period. In this case, the maximum value of the autocorrelation of signal A '' and signal B '' should be obtained (the starting point of signal A '' has the maximum correlation value at 4us and the starting point of signal B '' at 12us). However, since the fifth symbol (p = 2) is an off symbol at the time of autocorrelation, the maximum value of autocorrelation between signal A 'and signal B' is generated. Therefore, it is difficult to synchronize time using the signal.

An embodiment of time synchronization using autocorrelation is shown as follows.

First, if there is no off symbol indicating the end of the preamble sequence, the number of symbols of the synchronization sequence is n, the period (symbol number) of the periodic sequence is p, the window size (number of symbols) is n, and the two taking autocorrelation The magnitude of the signal is np and the time difference between the two signals taking autocorrelation is p.

If there is an off symbol at the end of the preamble sequence, the number of symbols in the synchronization sequence is n, the period (symbol number) of the periodic sequence is p, the window size (number of symbols) is n-1, and the two taking autocorrelation The magnitude of the signal is np-1 and the time difference between the two signals taking autocorrelation is p.

In the wake-up preamble, when the time synchronization, channel and received power estimation, and the role of identifier are performed at the same time, the computational burden may increase, and thus, each configuration may be performed as follows.

A synchronization sequence for time synchronization (period sequence + symbol for end of period sequence) + symbols for channel and received power estimation + symbols for identifier

The synchronization sequence may be configured in the same manner as described above.

The symbols for channel and received power estimation are advantageously composed of all on symbols in a typical OOK.

The symbols for the identifier may be composed of BSS color or group ID, or other specific information.

In the general OOK scheme, the preamble sequence is {1,1,1,... , 1,0, 1,1,… , 1, identifier symbols}.

In a OOK with Manchester coding applied to the payload, the symbols for the channel and the received power estimate may be composed and symbols for the channel and the received power estimate may not be needed. Accordingly, the preamble sequence to which the Manchester coding is applied to the payload may be composed of {period sequence + symbol for identifying the end of the periodic sequence + symbol for the identifier}. In this case, the additional overhead may be smaller than that of the general OOK.

Alternatively, the channel and the received power estimation may be simultaneously performed using the synchronization sequence, and only symbols for the identifier may be separately configured. At this time, the preamble sequence is, for example, {1,1,1,... , 1, 0, identifier symbols}.

Alternatively, the identifier may be simultaneously determined using the synchronization sequence, and only the channel and the received power estimation may be performed separately. At this time, the preamble sequence is, for example, {1,1,1,... , 1,0, 1,1,… , 1}.

Alternatively, both the channel and the received power estimation and the identifier determination may be performed simultaneously using the synchronization sequence. At this time, the preamble sequence is, for example, {1,1,1,... , 1,0}.

19 is a flowchart illustrating a procedure of transmitting a signal by applying the OOK method according to the present embodiment.

An example of FIG. 19 is performed in a transmitter, the receiver may correspond to a low power wake-up receiver, and the transmitter may correspond to an AP.

First of all, the term “on signal” may correspond to a signal having an actual power value. The off signal may correspond to a signal that does not have an actual power value. The on signal may correspond to information 1, and the off signal may correspond to information 0.

In step S1910, the transmitter configures a wakeup packet.

In operation S1920, the transmitter transmits the wakeup packet to the receiver.

How the wakeup packet is configured is as follows.

The preamble of the wakeup packet includes a synchronization sequence including an on signal and an off signal to which an On-Off Keying (OOK) scheme is applied.

The on signal is transmitted through an on symbol generated by applying a first sequence to K consecutive subcarriers in a 20 MHz band and performing a 64-point Inverse Fast Fourier Transform (IFFT). In other words, one bit may be transmitted through one symbol generated by performing an IFFT. In this case, K is a natural number.

The sync sequence includes an off symbol indicating the end of the sync sequence. In addition, the off symbol may be inserted at the end of the synchronization sequence. In other words, the synchronization sequence has a periodicity. The off symbol indicating the end of the synchronization sequence is overhead, but the end of the period can be known to help the time synchronization performance.

The sync sequence begins with the on signal. This is because it is advantageous to time synchronization at the same time as the notification that the signal is coming. In addition, the period of the on signal is set in units of multiples of the on symbol. That is, since information of each symbol itself has a periodicity, and one symbol has a length of 4 us, a synchronization sequence (or preamble sequence) having a periodicity of 4 us multiples can be configured.

For example, when the period is 1 and the length of the sync sequence is 6, the sync sequence may include {1, 1, 1, 1, 1, 0}. This is the case where the period is one times the symbol 4us.

As another example, when the period is 2 and the length of the sync sequence is 6, the sync sequence may consist of {1, 0, 1, 0, 1, 0}. This is the case where the period is twice the symbol 4us.

The transmitter may synchronize time with the receiver by calculating an auto-correlation value of the synchronization sequence. The transmitter calculates the correlation value of the two time signals that differ by the period in the synchronization sequence, and sets the case having the largest correlation value as the start point of the wakeup packet. This makes it fairly easy to time wake up packets.

In addition, the preamble of the wakeup packet may further include a symbol for channel estimation and measurement of received power, and a symbol for identifier.

However, when channel estimation and measurement of received power are performed using the synchronization sequence, the preamble of the wakeup packet may not include symbols for the channel estimation and the measurement of the received power.

In addition, when determining the identifier using the synchronization sequence, the preamble of the wakeup packet may not include a symbol for the identifier. The identifier may include identifier information of a transmitter or a receiver.

The wakeup packet may be applied only to the OOK scheme, or the OOK scheme and Manchester coding may be applied. If a OOK scheme and Manchester coding are applied to the wakeup payload included in the wakeup packet, a symbol for measuring the channel estimate and the received power may not be needed in the preamble of the wakeup packet.

In the K subcarriers to which the first sequence is applied, the coefficient of the subcarrier is set to 1 or -1 in units of m subcarriers, and the coefficients of the remaining subcarriers are set to 0. In other words, if m = 2, the first sequence is applied to a subcarrier with coefficients in two spaces; if m = 4, the first sequence is applied to a subcarrier with coefficients in four spaces; The first sequence may be applied to a subcarrier having coefficients in units of eight columns. M is a natural number and may correspond to an even number as a value for decreasing a symbol.

This is to generate an information signal (or symbol) applying a symbol reduction technique. When the first sequence is applied to K subcarriers having coefficients of a subcarrier in units of m subcarriers and IFFT is performed, a 3.2us signal having a 3.2us / m period is generated. You can take one of these to construct a 3.2us / m on signal. The 3.2us / m on signal may be transmitted through the first symbol. Accordingly, the first symbol may have a length of 3.2 us / m except for CP. As a result, the length of the symbol may be reduced and the data rate may be increased instead of the conventional OOK method.

In addition, if m = 1 can be seen that the simple OOK method is applied, if m = 2 can be seen that the Manchester coding method is applied. That is, the first information and the second information may also be considered to have characteristics of an information signal to which Manchester coding is applied and an information signal to which a symbol reduction technique is applied.

In addition, the off signal may be transmitted through an off symbol generated by applying a second sequence to K consecutive subcarriers in the 20 MHz band and performing a 64-point IFFT. The first sequence and the second sequence may be different from each other. In the K subcarriers to which the second sequence is applied, the coefficients of all subcarriers may be set to zero.

Even in the case of the off signal, when the second sequence is applied to K subcarriers in which the coefficients of all subcarriers are set to 0 and IFFT is performed, a 3.2us signal having a 3.2us / m period is generated. You can take one of these to construct a 3.2us / m off signal. The 3.2us / m off signal may be transmitted through the second symbol. Therefore, the second symbol may have a length of 3.2 us / m except for CP.

The K subcarriers may correspond to the partial band of the 20MHz band. For example, assuming K = 13 and 20 MHz as a reference band, since 13 subcarriers are sampled and IFFT is performed even though 64 subcarriers (or bit sequences) can be used, 13 subcarriers are approximately 4.06 MHz. It can correspond to a band. That is, a specific sequence (first sequence or second sequence) is set only to 13 subcarriers selected as samples, and all other subcarriers except 13 subcarriers are set to 0. That is, it can be said that power is provided only for 4.06MHz in the 20MHz band in the frequency domain.

In addition, the subcarrier spacing of each of the K subcarriers may be 312.5 KHz. When m is 2, the on symbol and the off symbol may have a length of 1.6us. When m is 4, the on symbol and the off symbol may have a length of 0.8us.

In addition, by inserting a cyclic prefix (CP) in front of each symbol can reduce or prevent the occurrence of Inter Symbol Interference (ISI).

For example, a cyclic prefix (CP) may be inserted in front of the on symbol and the off symbol, respectively. When m is 2, the CP may have a length of 0.4 us. When m is 4, the CP may have a length of 0.2us. This embodiment may be effective when there is an influence of ISI in the middle of the signal since the CP is inserted in the middle of the signal.

Also, the transmitter may first configure the information 0 and the information 1 by knowing the power values of the on signal and the off signal. The receiver decodes information 0 and information 1 by using an envelope detector, thereby reducing power consumed in decoding.

20 is a block diagram illustrating a wireless device to which the present embodiment can be applied.

Referring to FIG. 20, the wireless device may be an AP or a non-AP station (STA) that may implement the above-described embodiment. The wireless device may correspond to the above-described user or may correspond to a transmission device for transmitting a signal to the user.

The AP 2000 includes a processor 2010, a memory 2020, and an RF unit 2030.

The RF unit 2030 may be connected to the processor 2010 to transmit / receive a radio signal.

The processor 2010 may implement the functions, procedures, and / or methods proposed herein. For example, the processor 2010 may perform an operation according to the present embodiment described above. That is, the processor 2010 may perform an operation that may be performed by the AP during the operations disclosed in the embodiments of FIGS. 1 to 19.

The non-AP STA 2050 includes a processor 2060, a memory 2070, and a radio frequency unit 2080.

The RF unit 2080 may be connected to the processor 2060 to transmit / receive a radio signal.

The processor 2060 may implement the functions, processes, and / or methods proposed in this embodiment. For example, the processor 2060 may be implemented to perform the non-AP STA operation according to the present embodiment described above. The processor may perform the operation of the non-AP STA in the embodiment of FIGS. 1 to 19.

The processors 2010 and 2060 may include application-specific integrated circuits (ASICs), other chipsets, logic circuits, data processing devices, and / or converters for interconverting baseband signals and wireless signals. The memories 2020 and 2070 may include read-only memory (ROM), random access memory (RAM), flash memory, memory cards, storage media and / or other storage devices. The RF unit 2030 and 2080 may include one or more antennas for transmitting and / or receiving a radio signal.

When the embodiment is implemented in software, the above-described technique may be implemented as a module (process, function, etc.) for performing the above-described function. Modules may be stored in memories 2020 and 2070 and executed by processors 2010 and 2060. The memories 2020 and 2070 may be inside or outside the processors 2010 and 2060, and may be connected to the processors 2010 and 2060 by various well-known means.

Claims (13)

1. In a method for transmitting a wake-up packet in a wireless LAN system,

   Configuring, by the transmitting device, a wakeup packet; And

   Transmitting, by the transmitter, the wakeup packet to a receiver;

   The preamble of the wakeup packet includes a synchronization sequence consisting of an on signal and an off signal to which an on-off keying (OOK) scheme is applied.

   The on signal is transmitted through an on symbol generated by applying a first sequence to K consecutive subcarriers in a 20 MHz band and performing a 64-point Inverse Fast Fourier Transform (IFFT).

   The sync sequence includes an off symbol indicating an end of the sync sequence,

   The synchronization sequence starts with the on signal,

   The on signal has a period set in units of multiples of the on symbol, and

   K is a natural number

   Way.
2. The method of claim 1,

   The off symbol is inserted at the end of the sync sequence

   Way.
3. The method of claim 2,

   If the period is 1 and the length of the synchronization sequence is 6,

   The sync sequence is composed of {1, 1, 1, 1, 1, 0}

   Way.
4. The method of claim 2,

   If the period is 2 and the length of the synchronization sequence is 6,

   The sync sequence is composed of {1, 0, 1, 0, 1, 0}

   Way.
5. The method of claim 2,

   The transmitting device calculates an auto-correlation value of the synchronization sequence to synchronize time with the receiving device.

   Way.
6. The method of claim 2,

   The preamble of the wakeup packet further includes symbols for channel estimation and measurement of received power, and symbols for identifiers.

   Way.
7. The method of claim 6,

   When performing channel estimation and measurement of received power using the synchronization sequence, the preamble of the wakeup packet does not include symbols for the channel estimation and measurement of received power.

   Way.
8. The method of claim 6,

   When determining the identifier using the sync sequence, the preamble of the wakeup packet does not include a symbol for the identifier.

   Way.
9. The method of claim 1,

   The wake-up packet is applied only to the OOK scheme, or to the OOK scheme and Manchester coding.

   Way.
10. The method of claim 1,

    In the K subcarriers to which the first sequence is applied, the coefficient of the subcarrier is set to 1 or -1 in units of m subcarriers, and the coefficients of the remaining subcarriers are set to 0,

    M is a natural number

    Way.
11. The method of claim 10,

    The K subcarriers correspond to the partial band of the 20 MHz band,

    Each subcarrier spacing of the K subcarriers is 312.5 KHz,

    When m is 2, the on symbol and the off symbol have a length of 1.6us,

    When m is 4, the on symbol and the off symbol have a length of 0.8us.

    Way.
12. The method of claim 11,

    Cyclic prefix (CP) is inserted in front of the on symbol and the off symbol, respectively.

    When m is 2, the CP has a length of 0.4us,

    When m is 4, the CP has a length of 0.2us.

    Way.
13. A transmitter for transmitting a wake-up packet in a wireless LAN system,

    RF unit for transmitting or receiving a wireless signal; And

    Including a processor for controlling the RF unit,

    The processor is:

    Construct a wakeup packet, and

    The wakeup packet is transmitted to a receiving device,

    The preamble of the wakeup packet includes a synchronization sequence consisting of an on signal and an off signal to which an on-off keying (OOK) scheme is applied.

    The on signal is transmitted through an on symbol generated by applying a first sequence to K consecutive subcarriers in a 20 MHz band and performing a 64-point Inverse Fast Fourier Transform (IFFT).

    The sync sequence includes an off symbol indicating an end of the sync sequence,

    The synchronization sequence starts with the on signal,

    The on signal has a period set in units of multiples of the on symbol, and

    K is a natural number

    Transmitter.

Images