BINNING-BASED TRANSMIT BEAMFORMING FOR WIRELESS COMMUNICATION SYSTEMS

Abstract

A method for transmit beamforming, the method comprising: Determining a time correlation and a frequency correlation of a beamforming vector estimated from a previous transmission on a current frequency hopping channel and one or more adjacent channels; Determining one or more beamforming weights for the current frequency hopping channel; and Providing a signal using transmit beamforming based on one or more beamforming weights for a transmit operation.

Description

[0001] The present disclosure relates generally to wireless communication devices and, more particularly, to binning-based transmit beamforming for wireless communication systems.

[0002] In wireless communications, such as Bluetooth, each connection event begins with a master device performing a transmit operation and a slave device performing a receive operation. It then continues with the slave device performing a transmit operation and the master device performing a receive operation. Because of this standardized sequence of transmitting and receiving for the master and slave devices, transmit beamforming (Tx beamforming) is hampered by the difficulty of obtaining an actual beamforming vector at the master device. In particular, the master device can estimate the beamforming vector while receiving a packet with a known preamble from the slave device.Because the master device first performs a transmit operation followed by a receive operation, the latest beamforming vector is not available at the master device during a Tx beamforming operation in a transmit time slot. Indeed, the latest estimate of the beamforming vector is available at the master device in the receive time slot following the completion of the Tx beamforming operation.

[0003] US 9 749 889 B2 describes a method for communicating with a station in a wireless network, the method comprising forming a plurality of narrowband beams, each having a different angular direction from an antenna of a base station and distributed together over a beam space to form a pseudo-omnidirectional beam pattern.

[0004] According to one aspect, a method of transmit beamforming is provided, the method comprising: Determining a time correlation and a frequency correlation of a beamforming vector estimated from a previous transmission on a current frequency hopping channel and one or more adjacent channels; Determining one or more beamforming weights for the current frequency hopping channel; and Providing a signal using transmit beamforming based on one or more beamforming weights for a transmit operation.

[0005] Conveniently, the one or more adjacent channels have a beamforming vector that corresponds to that of the current frequency hopping channel based on a channel coherence bandwidth.

[0006] Conveniently, the method further comprises the following: Combining the current frequency hopping channel and one or more neighboring channels into the same channel group (bin).

[0007] Conveniently, the method further comprises the following: Obtaining a latest estimate of the beamforming vector in a channel group, wherein the time correlation and the frequency correlation of the beamforming vector are determined from the latest estimate.

[0008] Conveniently, the method further comprises the following: Forming a channel group comprising a number of channels, wherein the number of channels combined in a channel group is determined by means of a channel group width parameter.

[0009] Conveniently, the method further comprises the following: Determining whether a new estimate of the beamforming vector is obtained from at least one of a plurality of channels grouped in a same channel group; and Updating a beamforming vector in each channel group of a plurality of channel groups when the new estimate of the beamforming vector is obtained.

[0010] Conveniently, the method further comprises the following: Combining consecutive channels into a plurality of channel groups for a first type of binning architecture, each of the plurality of channel groups excluding common channels.

[0011] Conveniently, the method further comprises the following: Combining consecutive channels into a plurality of channel groups for a second type of binning architecture, wherein each channel group of the plurality of channel groups is formed locally around a center indicated by a respective one of the consecutive channels.

[0012] It is useful to have overlapping channels among the multitude of channel groups.

[0013] Conveniently, the second type of binning architecture uses frequency correlation of beamforming vectors of the one or more adjacent channels.

[0014] Conveniently, phase estimates are updated in a channel group centered on a first channel and in one or more adjacent channel groups containing the first channel.

[0015] Conveniently, the method further comprises the following: Determining a first channel group centered on a channel having a new phase estimate; and Updating a phase estimate in the first channel group and in one or more adjacent channel groups of the first channel group with the new phase estimate.

[0016] Conveniently, the method further comprises the following: determining that a new beamforming vector estimate is available for one or more channel groups; adaptively calculating one or more objective functions based on a change in a channel condition associated with the one or more channel groups; and Determining an optimal channel group width using the adaptively calculated one or more objective functions.

[0017] Conveniently, the method further comprises the following: determining that a new beamforming vector estimate is available for one or more channel groups; iteratively calculating correlation values ​​of one or more beamforming vectors in a time and a frequency dimension based on a previous correlation value; and Determine an optimal channel group width using the iteratively calculated correlation values.

[0018] Conveniently, the method further comprises the following: for each channel group width from a variety of possible channel group widths via a thorough offline search: Determine whether the channel group width reduces a packet error rate for a particular channel profile; Selecting the channel group width that reduces the packet error rate; and Apply the selected channel group width as a fixed channel group width during operation.

[0019] Conveniently, a phase estimate in a channel group is updated in response to a new phase estimate received from one or more channels in a same channel group.

[0020] Conveniently, the method further comprises the following: Determining whether a time for returning to a same frequency is greater than a channel coherence time; and Determine that a final estimate of the last phase of a current hop frequency is valid if the time to return to the same frequency is not longer than the channel coherence time.

[0021] Conveniently, the method further comprises the following: Selecting a channel group width that maximizes the correlation of phase values ​​averaged over a number of channels in both the time and frequency dimensions.

[0022] According to one aspect, there is provided a computer program product comprising instructions stored in a tangible, computer-readable storage medium, the instructions comprising: Instructions for determining a time correlation and a frequency correlation of a beamforming vector estimated from a previous transmission on a current frequency hopping channel and one or more adjacent channels; Instructions for combining the current frequency hopping channel and the one or more adjacent channels into a same channel group; Instructions for determining one or more beamforming weights in the same channel group for the current frequency hopping channel; and Instructions for providing a transmit signal using the one or more beamforming weights for a transmit operation.

[0023] The instructions shall also conveniently include the following: Instructions for dividing an allocation of non-overlapping channels into a plurality of channel groups based on a channel group width parameter.

[0024] Conveniently, successive channels of allocation of non-overlapping channels are grouped into respective channel groups of the plurality of channel groups.

[0025] For convenience, consecutive channels of allocation of non-overlapping channels are grouped together with intervals determined by means of the channel group width parameter.

[0026] The instructions shall also conveniently include the following: Instructions for using the same beamforming vector estimate for channels grouped in the same channel group.

[0027] According to one aspect, an apparatus comprises: a processing circuitry configured to: Determining a time correlation and a frequency correlation of a beamforming vector estimated from a previous transmission on a current frequency hopping channel and one or more adjacent channels; Combining the current frequency hopping channel and the one or more adjacent channels into a channel group; Determining one or more beamforming weights for the current frequency hopping channel; and Providing a transmit signal for a transmit operation using the one or more beamforming weights.

[0028] Conveniently, the processing circuitry is further configured to combine successive channels for a first type of binning architecture into a plurality of channel groups, each of the plurality of channel groups excluding common channels.

[0029] Conveniently, the processing circuitry is further configured to combine successive channels into a plurality of channel groups for a second type of binning architecture, each channel group of the plurality of channel groups being formed locally around a center indicated by a respective one of the successive channels.

[0030] Conveniently, channels grouped in the same channel group use the same phase estimation for transmit beamforming. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Certain features of the claimed technology are set forth in the appended claims. However, for illustrative purposes, several embodiments of the claimed technology are set forth in the following figures.

[0032] Certain features of the claimed technology are set forth in the appended claims. However, for illustrative purposes, one or more implementations of the claimed technology are set forth in the following figures. Fig. 1 is a diagram illustrating a wireless communication system according to one or more implementations. Fig. 2 is a diagram illustrating a wireless communication device including a host device and an associated radio. Fig. 3 schematically illustrates an example of a communication environment utilizing transmit beamforming according to one or more implementations of the claimed technology. Fig. 4 illustrates an example of a Bluetooth communication system according to one or more implementations of the claimed technology. Fig. 5A and Fig. 5B illustrate examples of a frame exchange between a master device and a slave device according to one or more implementations of the claimed technology. Fig. 6A illustrates a graphical representation of an example histogram depicting the time for a signal to return to a same frequency according to one or more implementations of the claimed technology. Fig. 6B illustrates a graphical representation of example curves representing packet error rates of various transmit signals using single and multiple antennas according to one or more implementations of the claimed technology. Fig. 7A schematically illustrates an example of a global binning structure according to one or more implementations of the claimed technology. Fig. 7B illustrates a graphical representation of an example histogram depicting the time for a signal to return to a same frequency based on global binning according to one or more implementations of the claimed technology. Fig. Figure 7C illustrates a graphical representation of example curves representing packet error rates of various transmit signals using single and multiple antennas based on global binning according to one or more implementations of the claimed technology. Fig. 8A schematically illustrates an example of a local binning structure according to one or more implementations of the claimed technology. Fig. 8B schematically illustrates an example of a phase estimate update for the local binning structure according to one or more implementations of the claimed technology. Fig. 8C illustrates a graphical representation of example curves representing packet error rates of various transmit signals using single and multiple antennas based on local binning according to one or more implementations of the claimed technology. Fig. 9A and Fig. 9B illustrate graphical representations depicting examples of packet error rates for various channel group widths according to one or more implementations of the claimed technology. Fig. 10 illustrates a graphical representation of exemplary curves representing various packet error rates of transmit signals with different channel group widths according to one or more implementations of the claimed technology. Fig. 11A schematically illustrates an example of an adaptive binning structure according to one or more implementations of the claimed technology. Fig. 11B schematically illustrates an example of an adaptive binning structure having a first predetermined channel group width according to one or more implementations of the claimed technology. Fig. 11C schematically illustrates an example of an adaptive binning structure with a second predetermined channel group width according to one or more implementations of the claimed technology. Fig. 11D illustrates a graphical representation of an exemplary curve representing the average correlation of phase values ​​for different channel group widths according to one or more implementations of the claimed technology. Fig. 12 illustrates a block diagram of a process for binning-based transmit beamforming according to one or more implementations of the claimed technology. Fig. Figure 13 schematically illustrates an electronic system for implementing any implementation of the claimed technology. DETAILED DESCRIPTION

[0033] The detailed description set forth below is intended to describe various configurations of the claimed technology, and is not intended to represent the only configurations in which the claimed technology may be practiced. The accompanying drawings are incorporated herein by reference and form a part of the detailed description. The detailed description contains specific details intended to provide a better understanding of the claimed technology. However, the claimed technology is not limited to the specific details set forth herein and may be practiced using one or more implementations. In one or more instances, structures and components are shown in block diagram form to avoid obscuring the concepts of the claimed technology.

[0034] In a conventional approach to Tx beamforming, a receiver estimates a beamforming vector and sends it back to the transmitter. However, this approach incurs high communication overhead. In another conventional approach, a transmitter assumes channel reciprocity and uses preambles received from a receiver to calculate beamforming weights on the same channel. In yet another approach, a transmitter applies a latest beamforming vector by reusing the beamforming weights estimated from a last reception on the same frequency-hopping channel. However, the channel coherence time, during which the channel is considered to be unchanged, is inherently a limiting factor for using previously estimated beamforming vectors on a current frequency-hopping channel. Bluetooth, for example, is a frequency-hopping-based system.If the time to return to the same frequency-hopping channel is greater than the channel coherence time, it is likely that a previous beamforming vector estimate is outdated for use on the current frequency-hopping channel. Using outdated beamforming weights can cause significant performance degradation compared to applying the latest beamforming vector.

[0035] In Bluetooth, the master device always performs a transmit operation before receiving a packet from the slave device. The master device cannot receive a training sequence / preamble from the slave device on the current frequency-hopping channel to estimate the beamforming vector before starting transmit beamforming. Therefore, the conventional approaches discussed above are not directly applicable if the master device needs to obtain the latest beamforming vector estimate before the beamforming operation.

[0036] The claimed technology provides for an evaluation of the time correlation and the frequency correlation of beamforming vectors estimated from a previous transmission on a current frequency hopping channel and its neighboring channels in order to derive the beamforming weights for the current frequency hopping channel.

[0037] Adjacent channels have a similar beamforming vector due to the channel coherence bandwidth. As a result, grouping nearby channels into channel groups and using the most recent beamforming vector estimate in a channel group reduces the age of the beamforming weights to be used for transmit beamforming.

[0038] Binning determines the number of channels to be grouped into channel groups (bins) using a channel group width parameter. Channels belonging to the same channel group use the same beamforming vector estimate. A beamforming vector in each channel group is updated whenever a new beamforming vector estimate is received from any of the channels belonging to the same channel group. Thus, binning evaluates channel correlation in both the frequency and time dimensions.

[0039] In some implementations, the claimed technology includes global binning, local binning, and an adaptive version of global and local binning. In global binning, consecutive channels are grouped into channel groups, and consequently, channel groups have no shared channels. In local binning, channel groups are formed around a center specified by each channel such that there are overlapping channels between channel groups. In the adaptive versions, an optimal channel group width can be determined with respect to changing channel conditions by maximizing penalty- and reward-based objective functions, which can be adaptively computed when a new beamforming vector is available.

[0040] In some implementations, the claimed technology includes an iterative version of adaptive binning, which provides for lower computational complexity. Once a new estimate of the beamforming vector is available in the iterative version, correlation values ​​of beamforming vectors in the time and frequency dimensions can be iteratively calculated based on a previous correlation value, rather than calculating a correlation matrix in its entirety.

[0041] In some implementations, the claimed technology includes fixed binning, which can be used where a thorough search is performed to select a channel group width that minimizes the PER (Packet Error Rate). In some manifestations, the channel group width selection for a given channel profile is performed offline, and the selected channel group width remains unchanged during operation.

[0042] The claimed technology provides beamforming gain in the form of increased receive signal-to-noise ratio (SNR), improved error rate performance, and reduced number of transmissions for any wireless device using Bluetooth with multiple antennas compared to a single antenna system.

[0043] Fig. Figure 1 is a diagram illustrating a wireless communication system 100 according to one or more implementations. However, not all of the depicted components may be required, and one or more implementations may include additional components not shown in the figure. Variations in the arrangement and type of components are possible without departing from the spirit or scope of the claims set forth herein. Additional components, different components, or fewer components may be provided.

[0044] The wireless communication system 100 includes the base stations (BS) and / or access points (AP) 111 to 113 (an AP may be a personal control point), the wireless communication devices 120 to 127, and a network hardware component 114. The wireless communication devices 120 to 127 include the laptop computers 120 and 124, the PDAs 121 and 127, the PCs 123 and 126, the mobile phones 122 and 125, and / or any other type of device that supports wireless communication.

[0045] Base stations or access points 111 to 113 are operably coupled to network hardware 114 via local area network (LAN) connections 115 to 117. Network hardware 114, which may be a router, switch, bridge, modem, or system controller, may provide a wide area network (WAN) connection 118 for wireless communication system 100. Base stations or access points 111 to 113 have an associated antenna or antenna array for individually communicating with wireless communication devices within their range. The wireless communication devices register with a particular base station or access point 111 to 113 to receive services within wireless communication system 100.In direct connections (for example, point-to-point communication), the wireless communication devices can communicate directly over an assigned channel.

[0046] Base stations can be used for cellular phone systems (including LTE and 5G systems) and similar systems, while access points can be used for wireless home or building networks. Regardless of the specific type of communication system, any wireless communication device may include and / or be coupled to a radio unit. The radio unit includes a linear amplifier and / or a programmable multi-stage amplifier to improve performance, reduce cost, reduce size, and / or enhance broadband applications. The radio unit may also include or be coupled to an antenna or antenna array with a specific antenna coverage pattern for propagating outgoing RF (radio frequency) signals and / or receiving incoming RF signals.In some forms, the antenna array may be directional antennas (for example, with beamforming).

[0047] According to some implementations, base stations are used for cellular phone systems (for example, AMPS (Advanced Mobile Phone Services), digital AMPS, GSM (Global System for Mobile Communications), CDMA (Code Division Multiple Access), LMDS (Local Multi-Point Distribution Systems), MMDS (Multi-Channel Multi-Point Distribution Systems), EDGE (Enhanced Data Rates for GSM Evolution), GPRS (General Packet Radio Service), HSDPA (High-Speed ​​Downlink Packet Access), HSUPA (High-Speed ​​Uplink Packet Access and / or variations thereof) and similar systems), while access points are used for wireless home or building networks (for example, IEEE 802.11, Bluetooth, ZigBee, any other type of RF-based network protocol and / or variations thereof). Regardless of the specific type of communication system, each wireless communication device includes a built-in radio and / or is coupled to a radio.

[0048] One or more of the devices shown may include circuitry and / or software that allows the particular device to communicate with each other or with nearby BT devices 150-159 using a Bluetooth (BT)-based communication system technology. The communication range using BT is smaller than that of typical WLAN (Wireless Local Area Network) network connections. A BT communication connection may utilize various versions of a BT specification, including Bluetooth Core Specification Version 4.0, Volume 6 (Low Energy Controller Volume), which applies to Bluetooth™ Low Energy (BLE) radio technology.Although BLE can be operated in conjunction with classic BT radio technology, BLE has a functional difference in the application of the protocol to establish a communication link between one or more BLE-compatible devices.

[0049] Fig. Figure 2 illustrates a schematic block diagram of a wireless communication portion 200 of a wireless device according to one or more implementations of the claimed technology. However, not all of the depicted components may be required, and one or more implementations may include additional components not shown in the figure. Variations in the arrangement and type of components are possible without departing from the spirit or scope of the claims set forth herein. Additional components, different components, or fewer components may be provided.

[0050] The wireless communication portion 200 includes a transmitter (TX) 201, a receiver (RX) 202, a local oscillator (LO) 207, and a baseband module 205. The baseband module 205 may be configured to provide baseband processing operations. In some implementations, the baseband module 205 includes a digital signal processing (DSP) unit. The baseband module 205 is coupled to a host unit (e.g., host 210), an application processing unit, or one or more other units that provide functional Bluetooth processing for the device and / or the interface to a user.

[0051] As in Fig. 2, a host 210 is provided. The host 210 may represent a host module of a Bluetooth device, while the wireless communication part 200 is used to provide the radio function (e.g., RF front-end) and the baseband functions. The radio part of the wireless communication part 200 may be implemented to support one or more Bluetooth modes of operation, or it may include other wireless systems, such as WLAN (e.g., WiFi) and / or cellular or satellite communications. Any or all of the Fig. 2 shown hardware components can be integrated into one or more of the Fig. 1 shown wireless communication devices may be installed.

[0052] Memory 206 is coupled to baseband module 205. Memory 206 can be used to store data, including program instructions that affect baseband module 205. Various types of storage devices can be used for memory 206. Memory 206 can be located anywhere in wireless communication portion 200.

[0053] The transmitter 201 and the receiver 202 are coupled to an antenna 204 via the transmit / receive (T / R) switching module 203. The T / R switching module 203 is configured to switch the antenna between the transmitter and receiver depending on the operating mode. In some embodiments, separate antennas are used for the transmitter 201 and the receiver 202. In some implementations, multiple antennas or antenna arrays are used with the wireless communication portion 200 to provide antenna diversity or multiple inputs / multiple outputs (MIMO) capabilities.

[0054] At frequencies in the gigahertz range (for example, 2.4 GHz to 5 GHz), omnidirectional antennas can provide adequate coverage for communication between wireless devices. At higher frequencies, however, directional antennas with beamforming capabilities are provided to direct the beam to concentrate the transmitted energy due to the limited range of the signal. In these cases, antenna arrays enable the beam to be directed in a specific direction or toward a specific target. Beamforming allows a pair of stations (STAs) or an access point (AP) and an STA to train and align their directional antennas to achieve an optimal wireless link for communication with each other. Beamforming is accomplished after the two devices have successfully completed a training sequence, as noted above. One feature of beamforming is beam refinement.Beam refinement is a process in which a STA improves its antenna configuration (or antenna weight and vector) for transmission and / or reception.

[0055] Outgoing data for transmission from host 210 is forwarded to baseband module 205 and converted into baseband signals, then upconverted for transmission via transmitter 201. For example, transmitter 201 converts the baseband signals into outgoing radio frequency (RF) signals for transmission from wireless communication portion 200 via antenna 204. Transmitter 201 may use a variety of upconversion or modulation techniques to convert the outgoing baseband signals into outgoing RF signals. The conversion process depends on the particular communication standard or protocol in use.

[0056] In a similar manner, incoming RF signals are received by antenna 204 and coupled to receiver 202. Receiver 202 then converts the incoming RF signals into incoming baseband signals, which are then coupled to baseband module 205. Receiver 202 may utilize a variety of downconversion or demodulation techniques to convert the incoming RF signals into incoming baseband signals. The incoming baseband signals are processed by baseband module 205, and incoming data is output from baseband module 205 to host 210.

[0057] The LO 207 provides local oscillation signals to the transmitter 201 for upconversion and to the receiver 202 for downconversion. In some embodiments, separate LO signals may be used for the transmitter 201 and the receiver 202. While a variety of LO circuitry may be used, some implementations utilize a PLL (phase-locked loop) to lock the LO to output a frequency-stable LO signal based on a selected channel frequency.

[0058] The baseband module 205, the LO 207, the transmitter 201, and the receiver 202 may be integrated on a same integrated circuit (IC) chip. The transmitter 201 and the receiver 202 may sometimes be referred to as RF front-end modules (or components) or radio devices. In some embodiments, one or more of the aforementioned components may be located on separate IC chips. Similarly, other Fig. 2, along with the baseband module 205, the LO 207, the transmitter 201, and the receiver 202, may be incorporated into the same integrated circuit chip. In some embodiments, the antenna 204 is incorporated into the same integrated circuit chip. With the advent of system-on-chip (SOC) integration, host devices, processing units for applications and / or user interfaces, such as the host 210, may be incorporated into the same integrated circuit chip along with the baseband module 205, the transmitter 201, and the receiver 202.

[0059] Any of the various embodiments of the wireless communication portion 200, which may be implemented within various communication systems, may include functionality for performing communication via multiple standards, multiple protocols, or multiple other predetermined communication means. For example, the wireless communication portion 200 implemented as a single communication device may include functionality for performing communication according to a first protocol, a second protocol, and / or a third protocol. These various protocols may be the WiMAX (Worldwide Interoperability for Microwave Access) protocol, a protocol compatible with a wireless local area network (e.g., WLAN / WiFi) (e.g., one of the IEEE 802.11 (Institute of Electrical and Electronics Engineers) protocols, such as 802.11a, 802.11b, 802.11g, 802.11n, 802.11ac, or 802.11b).11ax), a Bluetooth protocol, or any other predetermined means by which wireless communication can be effected.

[0060] Fig. Figure 3 schematically illustrates an example of a communications environment 300 utilizing transmit beamforming according to one or more implementations of the claimed technology. However, not all of the depicted components may be required, and one or more implementations may include additional components not shown in the figure. Variations in the arrangement and type of components are possible without departing from the spirit or scope of the claims set forth herein. Additional components, different components, or fewer components may be provided.

[0061] In some forms, a dual-core BT chip (or integrated circuit) provides beamforming / combining functions that are used to direct a beam to concentrate the transmitted energy. If, as in Fig. 3, wireless communication devices, such as mobile phone 125 and BT device 154, communicate in communication environment 300, a directed beam 302 is directed by mobile phone 125 toward BT device 154. In some embodiments, BT device 154 may be a wireless headset. The illustration of Fig. Figure 3 shows a plurality of directed energy beams emanating from the mobile phone 125, where one beam is larger than the others to indicate the directed energy in a particular orientation. Note that in a typical beamforming procedure, the particular device (e.g., 125) is operated to have a plurality of propagation sectors. Once the optimal sector has been detected or determined, the device orients the antenna (or antenna array) to operate in the optimal sector. Generally, training sequences are used to determine the optimal direction for orienting the antenna. Accordingly, in Fig. 3 the larger lobe represents the alignment of the directional antennas so that a first wireless device can optimally communicate with a second wireless device.

[0062] Transmit beamforming can improve system performance by utilizing multiple antennas at the transmitter to create spatial diversity and consequently reduce the effects of channel fading. In transmit beamforming, the same signal is transmitted from each transmit antenna, but the phase of the respective signals is adjusted so that the phases are progressively added at the receiver. Consequently, the SNR (signal-to-noise ratio) at the receiving end is increased (or maximized). In this respect, improved SNR reduces the number of packet errors. Consequently, improved error rate performance reduces the amount of retransmissions. This, in turn, leads to better bandwidth utilization. In some manifestations, no change in the BT packet format is required to perform transmit beamforming.

[0063] Fig. Figure 4 illustrates an example of a Bluetooth communication system 400 according to one or more implementations of the claimed technology. However, not all of the depicted components may be required, and one or more implementations may include additional components not shown in the figure. Variations in the arrangement and type of components are possible without departing from the spirit or scope of the claims set forth herein. Additional components, different components, or fewer components may be provided.

[0064] The Bluetooth communication system 400 includes a master device 410 and a slave device 420. The Bluetooth communication system 400 may be operable to utilize a frequency division multiple access (FDMA) scheme and a time division multiple access (TDMA) scheme to support voice and / or data communication. In some implementations, the Bluetooth communication system 400 may be capable of utilizing a TDMA-based polling scheme in the communication between the master device 410 and the slave device 420 at the link layer. In this regard, the TDMA-based polling method involves a device (e.g., the master device 250) transmitting a packet at a predetermined time and a corresponding device (e.g., the slave device 260) responding with a packet after a predetermined time.

[0065] As in Fig. 4, the master device 410 includes multiple antennas or an antenna array to create spatial diversity over the communication channel with the slave device 420. For example, the master device 410 may transmit a same signal from each transmit antenna, with the phase of each signal adjusted such that the phases are progressively added at the slave device 420. Within a link event, the master and slave devices alternate sending data packets using the same data channel. In some implementations, the master device 410 initiates the start of each link event and may terminate each link event at any time.

[0066] In some implementations, the master device 410 includes a Maximum Ratio Combining (MRC) module (not shown) in a receiver circuit of the master device 410 to enable receiver diversity in a wireless link. During reception of a packet, the MRC module estimates the relative phase between two antennas. Then, at a transmitter of the master device 410, this relative phase estimate is used to adjust the phase of signals from each antenna so that they add up at the receiver (e.g., at the slave device 420). In this regard, the received signal after beamforming can be expressed as follows: y(k)=h1x(k)+h2ej(θ1−θ2)x(k)+n =(|h1|+|h2|)ejθ1x(k)+n where h1 = |h1|e jθ1 , h2 = |h2|e jθ2 , θ1 - θ2 = relative phase between two antennas.

[0067] Fig. 5A and Fig. 5B illustrate examples of a frame exchange between a master device and a slave device according to one or more implementations of the claimed technology. During a link event, data packets may be sent with interframe spacing, and at least one data packet at the link event may originate from a master, such as master device 410. Master device 410 may send the first data packet at each link event to a designated slave, such as slave device 420. In this regard, slave device 420 may send a response after a predetermined time (e.g., 514), which may then be followed by another master transmit (not shown). Master device 410 may be operable to utilize a TDMA-based polling scheme at each link event to perform a packet transmit poll of the designated slave.The master device 410 may be capable of determining the packet payload size for data packets and the timing for sending packets at each connection event.

[0068] The slave device 420 may be connected to the master device 410 at the data link layer via one or more connections. The slave device 420 may be capable of synchronizing with connection start points, called anchor points from the perspective of a slave device, for data communication with the master device 410. The slave device 420 may consider the establishment of a connection with the master device 410 at the data link layer complete after receiving a connection request packet from the master device 410. The slave device 420 may be operable to transmit data packets on the data channel after receiving a packet from the master device 410 on the associated data link layer connection.

[0069] In Fig. 5A, a first frame exchange 510 includes a master device (e.g., 410) with a single antenna first transmitting to a slave device (e.g., 420) in a first time slot 512. In this regard, the slave device 420, which has multiple antennas, first receives and second transmits. Consequently, the slave device 420 may obtain a phase estimate during reception and then apply beamforming in a subsequent time slot 514. For example, the MRC module of the slave device 420 may calculate the phase estimate (e.g., θ2-θ1) in the time slot 512 and apply the phase estimate θ2-θ1 for transmit beamforming to the master device 410 via the multiple antennas. The frame exchange of Fig. Figure 5A represents an optimal use case, with the minimum delay between estimating and applying beamforming coefficients.

[0070] In Fig. 5B, a second frame exchange 520 includes the multi-antenna master device 410 first performing a transmission to the slave device 420 in a first time slot 522. However, the master device 410 first performs the transmission before the reception, and therefore, a current phase estimate is not available for Tx beamforming. In this regard, the master device 410 must use phase estimates from previous packets. In a second time slot 524, the MRC module of the master device 410 may calculate the phase estimate (e.g., θ2-θ1) after receiving from the slave device 420 in the time slot 524. The frame exchange of Fig. Figure 5B illustrates a challenging use case where previous phase estimates are likely to be outdated due to frequency hops and channel changes.

[0071] Fig. Figure 6A illustrates a graphical representation of an exemplary histogram 600 depicting the time for a signal to return to a same frequency, according to one or more implementations of the claimed technology. The channel coherence bandwidth is a statistical measure of a range of frequencies at which the signals are highly correlated. Adjacent frequencies have a similar phase due to the channel coherence bandwidth. Fig. 6A, the channel coherence time (T c ) can be expressed as follows: Tc=9 / (16*π*fd), where the Doppler frequency (f d) is 3 Hz, which can result in a channel coherence time of 60 ms. If the time to return to the same frequency is greater than the channel coherence time, the last phase estimate of a current frequency hopping frequency is likely to be outdated. This can cause a significant difference in performance compared to a case scenario with knowledge of an ideal phase value. As in Fig. As illustrated in Figure 6A, the mean return time is about 98.75 ms, while the channel coherence time is about 60 ms, which presumably suggests that the current frequency hopping frequency is outdated in this case given that the mean return time is greater than the channel coherence time.

[0072] Fig. 6B illustrates a graphical representation of example curves representing packet error rates of various transmit signals using a single and multiple antennas. A first curve 612 represents a transmit signal using a single antenna, a second curve 614 represents a transmit signal using transmit beamforming based on a final phase estimate, and a third curve 616 represents a transmit signal using transmit beamforming based on the ideal phase estimate, which is known. The second curve 614 provides improved SNR measurements compared to the first curve 612, however, the second curve 614 still lags behind the third curve 616. The SNR gap between the second curve 614 and the third curve 616 becomes wider with increasing PER (for example, 2.9 dB at PER 10 -1 , 7 dB at PER 10 -2 and 11.7 dB at PER 10 -3 ).

[0073] Fig. 7A schematically illustrates an example global binning structure 700 according to one or more implementations of the claimed technology. However, not all of the depicted components may be required, and one or more implementations may include additional components not shown in the figure. Variations in the arrangement and type of components are possible without departing from the spirit or scope of the claims set forth herein. Additional components, different components, or fewer components may be provided.

[0074] A Bluetooth communication system, such as the wireless communication part 200 of Fig. 2, may be operable to utilize a Frequency Division Multiple Access (FDMA) scheme and a Time Division Multiple Access (TDMA) scheme to support voice and / or data communications. The Bluetooth communication system may be configured to operate on 80 non-overlapping channels spaced 1 MHz apart according to the FDMA scheme. As described in Fig. 7A, the global binning structure 700 includes an allocation of Bluetooth channels 710 (e.g., numbered 0 to 79). The allocation of Bluetooth channels 710 is divided into groupings referred to as "channel groups" to form a channel group allocation 720. The channel group allocation 720 includes a first channel group 722 (referred to as "KG 1"), a second channel group 724 (referred to as "KG 2"), a fifteenth channel group 726 (referred to as "KG 15"), and a sixteenth channel group 728 (referred to as "KG 16"). The total number of channel groups in the global binning structure 700 may be 16 for a channel group width of 5 MHz.

[0075] In the global binning structure 700, consecutive channels are grouped into channel groups, and thus, channel groups do not have any shared channels. For example, the first channel group 722 includes channels 0 through 4, while the second channel group includes channels 5 through 9. The consecutive channels are grouped into channel groups with spacing determined by the channel group width. For example, each of the channel groups includes 5 channels based on a given channel group width of 5 MHz, with each channel having a spacing of 1 MHz. As a result, with a channel group width of 5 MHz, five (5) consecutive channels can be placed in the same channel group (i.e., channels 0, 1, 2, 3, 4 are in the first channel group 722, channels 5, 6, 7, 8, 9 are in the second channel group 724, and so on).In this regard, channels belonging to the same channel group use the same phase estimate for Tx beamforming. In some implementations, the phase estimate in a channel group is updated each time a new phase estimate is received from any of the channels in the same channel group.

[0076] However, the channels located at the edges of the channel groups (for example, channels 0 and 4 in the first channel group 722, channels 5 and 9 in the second channel group 724, etc.) do not fully utilize the phase estimates of their neighboring channels because the upper and lower halves of the neighboring channels belong to the adjacent channel groups (i.e., to a different channel group). For example, if the current frequency hopping frequency corresponds to channel 4, the latest phase estimate comes from channel 0. However, the phase estimate of the second channel group 724, where the last phase estimate comes from channel 5, is more recent than the phase estimate of the first channel group 722.

[0077] Fig. 7B illustrates a graphical representation of an exemplary histogram 730 depicting the time for a signal to return to a same frequency based on global binning according to one or more implementations of the claimed technology. As in Fig. As illustrated in Figure 7B, the mean return time is about 20.0 ms, while the channel coherence time is about 60 ms, which presumably suggests that the current frequency hopping frequency is current in this case. Compared to the Fig. 6B is the graphical representation shown in Fig. 7B is considerably shorter than the mean return time shown in Fig. 6B, illustrating the performance benefit of using global binning for transmit beamforming.

[0078] Fig. 7C illustrates a graphical representation 740 of example curves representing packet error rates of various transmit signals using single and multiple antennas based on global binning, according to one or more implementations of the claimed technology. The first curve 612 represents a transmit signal using a single antenna, the second curve 614 represents a transmit signal using transmit beamforming based on global binning with a channel group width of 1 MHz, the third curve 616 represents a transmit signal using transmit beamforming based on the ideal, known phase estimate, and a fourth curve 742 represents a transmit signal using transmit beamforming based on global binning with a channel group width of 5 MHz.The second curve 614 offers improved SNR measurements compared to the first curve 612, but the second curve 614 still lags behind the third curve 616. The SNR gap between the second curve 614 and the fourth curve 742 becomes wider with increasing PER (for example, 2.1 dB at PER 10). -1 , 4.3 dB' at PER 10 -2 and 5.1 dB at PER 10 -3 ). In this respect, the fourth curve 742 provides improved SNR measurements compared to the second curve 614 by closing the gap to the third curve 616.

[0079] As in Fig. 7B and Fig. As shown in Figure 7C, grouping nearby frequency hopping frequencies into channel groups and using the most recent phase estimate in the channel group reduces the age of the phase to be used by Tx beamforming.

[0080] Fig. 8A schematically illustrates an example of a local binning structure 800 according to one or more implementations of the claimed technology. However, not all of the depicted components may be required, and one or more implementations may include additional components not shown in the figure. Variations in the arrangement and type of components are possible without departing from the spirit or scope of the claims set forth herein. Additional components, different components, or fewer components may be provided.

[0081] Unlike the global binning structure 700, the channel groups in the local binning structure 800 have overlapping channels. As in Fig. 8A, the local binning structure 800 includes an allocation of Bluetooth channels 810 (e.g., numbered 0 to 79). The allocation of Bluetooth channels 810 is divided into groupings referred to as "channel groups" to form multiple channel group allocations (e.g., 820, 830, 840, 850). The channel group allocation 820 includes a first channel group 822 (referred to as "KG 1"), the channel group allocation 830 includes a second channel group 832 (referred to as "KG 2"), the channel group allocation 840 includes a third channel group 842 (referred to as "KG 3"), and the channel group allocation 850 includes a seventy-sixth channel group 852 (referred to as "KG 76").

[0082] In the local binning structure 800, channel groups are formed locally around a center specified by the respective channel. For example, if a channel group width is 5 MHz, the first channel group 722 includes channel 2 as the channel group center, as well as its neighboring channels that are ±1 MHz and ±2 MHz away from channel 2 (i.e., channels 1, 3, and channels 0, 4, respectively). Similarly, the second channel group 724 includes channel 3 as the channel group center, along with its adjacent channels 1, 2, 4, and 5. The total number of channel groups in the local binning is 76 for a channel group width of 5 MHz. In this regard, channel 2 uses the phase estimate in the first channel group 722, where it represents a center frequency of this channel group. Similarly, channel 3 uses the phase estimate in the second channel group 724, where it represents a center frequency of this channel group.

[0083] Unlike global binning, in local binning, the channel can belong to multiple channel groups. For example, the first channel group 722 and the second channel group 724 have four channels in common, i.e., channels 1, 2, 3, and 4. Based on the channel group formation structure, in which channel groups are formed locally around the relevant channel, the local binning structure 800 can utilize the frequency correlation of the beamforming vectors of the neighboring channels more efficiently than the global binning structure 700.

[0084] Fig. 8B schematically illustrates an example of a phase estimate update 860 for the local binning structure 800 according to one or more implementations of the claimed technology. In some implementations, the phase estimate in a channel group is updated each time a new phase estimate is received from any of the channels in the same channel group. As in Fig. 8B, assuming that a new phase estimate of channel 6 is available, phase estimates are updated in a channel group centered on channel 6 (e.g., among channel group centers 862) and in adjacent channel groups containing channel 6 (e.g., 864).

[0085] In some implementations, a channel group update rule is specified as follows: 1) The (using B c labeled) channel group centered around the channel that has a new phase estimate. 2) The phase estimation in the channel groups {Bc,Bc±1,…,Bc±⌊KG−width2⌋} update with the new phase information.

[0086] Fig. 8C illustrates a graph 870 of example curves representing packet error rates of various transmit signals using single and multiple antennas based on local binning, according to one or more implementations of the claimed technology. The graph 870 includes a first graph 872 representing a transmit signal using a single antenna, a second graph 874 representing a transmit signal using transmit beamforming based on the ideal, known phase estimate, a third graph 876 representing a transmit signal using transmit beamforming based on global binning with a channel group width of 5 MHz, and a fourth graph 878 representing a transmit signal using transmit beamforming based on local binning with a channel group width of 5 MHz.The third curve 876 offers improved PER measurements compared to the first curve 872, but the third curve 876 still lags behind the fourth curve 878. As shown in . Fig. As shown in Figure 8C, local binning results in a greater SNR improvement (i.e., up to 1.27 dB) than global binning.

[0087] Fig. 9A and Fig. 9B illustrate graphical representations depicting examples of packet error rates for different channel group widths according to one or more implementations of the claimed technology. In Fig. 9A, a graph 910 includes a curve 918 illustrating the packet error rate at a given SNR for a given channel group width, where the Doppler frequency is 3 Hz and the spacing is set to 3.75 ms. In the graph 910, an optimal channel group width exists within a region 912 of the curve 918 where the packet error rate is minimal. In some manifestations, the optimal channel group width for a case where the SNR is approximately 25 dB may be in a range of 8 MHz to 11 MHz. As the channel group width is decreased from the optimal channel group width range (e.g., 914), the packet error rate performance degrades because the mean time to return to a channel group increases. Consequently, the phase of the channel group is more likely to become stale due to a time-varying channel.

[0088] In Fig. 9B, a graph 920 includes a curve 928 representing the SNR at a given packet error rate for a given channel group width, where the Doppler frequency is 3 Hz and the spacing is set to 3.75 ms. In the graph 920, an optimal channel group width exists within a region 922 of the curve 928 where the SNR is minimal. In some manifestations, the optimal channel group width may be for a case where the packet error rate is approximately 0.01 (or 10 -2 ) range from 8 MHz to 11 MHz. When the channel group width is increased from the optimal channel group width range (for example, 926), packet error rate performance deteriorates because, with increasing channel group width, a channel group contains additional frequencies and its mean frequency correlation decreases due to the channel coherence bandwidth.

[0089] Fig. 10 illustrates a graphical representation 1000 of example curves representing various packet error rates of transmit signals with different channel group widths according to one or more implementations of the claimed technology. The graphical representation 1000 includes a first curve 10002 representing a transmit signal using a single antenna, a second curve 1004 representing a transmit signal using transmit beamforming based on the ideal, known phase estimate, and a bundle of curves 1006 each representing various transmit signals with varying, predetermined channel group widths. In some implementations, a thorough search may be performed over the bundle of channel group widths such that the channel group width that minimizes the packet error rate is selected and used as the predetermined channel group width during TX beamforming. As shown in Fig. 10, the channel group width 1008 (referred to as “9 MHz KG width”) is selected during the thorough search and applied to a binning structure, such as global binning (see Fig. 7A to Fig. 7C) or local binning (see Fig. 8A to Fig. 8C), applied.

[0090] In some implementations, the selection of the predetermined channel group width is performed manually and offline through a thorough search, making it inapplicable to real-time operation. In some implementations, the offline process of determining the channel group width through the thorough search may not be adaptive to changing channel conditions and / or packet spacing variations.

[0091] Fig. 11A schematically illustrates an example of an adaptive binning structure 1110 according to one or more implementations of the claimed technology. However, not all of the depicted components may be required, and one or more implementations may include additional components not shown in the figure. Variations in the arrangement and type of components are possible without departing from the spirit or scope of the claims set forth herein. Additional components, different components, or fewer components may be provided.

[0092] It is important to determine the optimal value of the channel group width so that both the time and frequency correlation among the channels in a channel group are fully utilized and, consequently, the best performance is obtained. If the channel group width is chosen too large, the frequency correlation of the channels in a channel group will be lower. On the other hand, if small values ​​are chosen for the channel group width, the total number of channel groups increases, and it takes longer to return to the same channel group due to the increasing number of channel groups. This results in beamforming weights to be used by the transmit beamformer becoming more easily outdated.

[0093] The optimal value of the channel group width depends on channel conditions, since the channel coherence time and the channel coherence bandwidth directly affect the correlation among the channels grouped in the same channel group. In a first approach to determining the optimal channel group width, given a given channel condition, offline simulations are run for a set of channel group width values, and the channel group width at which the packet error rates are most minimized would be selected. In a second approach to determining the optimal channel group width, as in Fig. Figure 11A illustrates how values ​​for the channel group width are adaptively found in light of changing channel conditions. Specifically, an optimal channel group width is found by maximizing a penalty-reward-based objective function, which is adaptively computed when a new beamforming vector estimate is available.

[0094] In Fig. 11A, the adaptive binning structure 1110 comprises a first channel group 1112, a second channel group 1114, a third channel group 1116 and a fourth channel group 1118. The first channel group 1112 corresponds to channel group i with the phase (i,t) . The second channel group 1114 corresponds to channel group n, and the third channel group 1116 corresponds to channel group p. The fourth channel group 1118 corresponds to channel group i with phase (i,t+T) . As in Fig. 11A, the first channel group 1112 is the same channel group as the fourth channel group 1118 (i.e., returned after a predetermined time for return).

[0095] In some implementations, it provides an improvement in transmit beamforming performance when highly correlated phase values (i,t) and θ (i,t+T) in a channel group for the channels {ch k , ch k+1 ,..., ch k+BW-1} are present when a return to the same channel group occurs after Tms. In this regard, determining the optimal channel group width is desirable for best use of the time and frequency correlation of phase values ​​in a channel group.

[0096] The adaptive binning structure 1110 may utilize adaptive global binning to select a channel group width that maximizes the correlation of phase values ​​averaged over a number of channels in both the time and frequency dimensions. This optimization of the channel group width can be formulated as follows: BWopt=arg maxBWi∈{1,2,3,…}given tp,NchRavg(BWi,tp) =arg maxBWigiven tp,Nch1Nch∑i=1NKGRavg,i(BWi,tp), where the term R avg,i (.) is the mean correlation of phase values ​​for a given channel group width (for example, i-th channel group), the term Δt corresponds to the packet spacing, the term N Kancorresponds to the total number of channels in adaptive frequency hopping (AFH), the term N KG corresponds to the total number of channel groups, the term BW i corresponds to the channel group width of the i-th channel group, the term BW opt corresponds to the optimal channel group width and the term t p corresponds to the Bluetooth packet spacing.

[0097] For some manifestations, the mean correlation per channel group R avg,i (BW i , t p ) can be calculated as follows: Ravg,i(BWi,tp)=BWi×H(BWi,tp,Δf=0)+2∑Δf′=1BWi−1((BWi−Δf′)H(BWi,tp,Δf=Δf′))

[0098] In equation (4) the term H (BW b,i , Δf, Δt) can be expressed as follows: H(BWi,tp,Δf)=∑k=1Nh(1−BWip)k−1p|Rf(Δf)||Rt(Δt=ktp)| where p=1NKan and N h corresponds to the total number of frequency jumps.

[0099] For some manifestations, the frequency correlation function can be calculated as follows: Rf(Δf)=E{CN(f,t)CN*(f−Δf,t−tp)}

[0100] For some manifestations, the time correlation function can be calculated as follows: Rt(Δt)=E{CN(f,t)CN*(f,t−Δt)} where C n (f,t) denotes the normalized cross spectrum between two channels, which is expressed as follows: CN(f,t)=H1(f,t)H2*(f,t)|H1(f,t)H2*(f,t)| where H1 (f,) is a complex channel frequency response of channel 1 and H2 (f,t) is a complex channel frequency response of channel 2.

[0101] Fig. 11B schematically illustrates an example of an adaptive global binning structure 1120 having a first predetermined channel group width according to one or more implementations of the claimed technology. However, not all of the depicted components may be required, and one or more implementations may include additional components not shown in the figure. Variations in the arrangement and type of components are possible without departing from the spirit or scope of the claims set forth herein. Additional components, different components, or fewer components may be provided.

[0102] In Fig. 11B, the adaptive global binning structure 1120 includes a first channel group 1122, a second channel group 1124-1, and a third channel group 1124-2. In some embodiments, the first channel group 1122 and the second channel group 1124-1 are separated by a predetermined time difference 1126, denoted as Δt. As shown in Fig. 11A, the second channel group 1124-1 is the same channel group as the third channel group 1124-2 (i.e., returned after a predetermined time for return). In the illustrated example of Fig. 11B, it is assumed that the first predetermined channel group width is equal to 1 MHz, with each channel group containing one channel. In this regard, the term R avg,i (BW i = 1, t p ) can be calculated as follows: Ravg,i(BWi=1,tp,Δf)= {|Rf,t(Δf=0,Δt=tp)|≈|Rf(Δf=0)||Rt(Δt=tp)||Rf(Δf=0)||Rt(Δt=2tp)||Rf(Δf=0)||Rt(Δt=3tp)| ⋮|Rf(Δf=0)||Rt(Δt=Nhtp)|with probability pwith prob.(1−p)pwith prob.(1−p)2pwith prob.(1−p)Nh−1p. where p = 1 / N ch . In equation (9), the correlation of the phase values ​​for each channel upon returning to its channel group is calculated. In this calculation, both the time and frequency domains are assumed to be independent.

[0103] In some implementations, the term R avg,i (BW i = 1, t p ) by calculating the expected value of R i (BW i = 1, t p , Δf) is formed by a change in frequency (Δf) and a change in time (Δt). For example, the calculation of the empirical value can be expressed as follows: Ravg,i(BWi=1,tp)=EΔf,Δt{Ri(BWi=1,tp,Δf)}=∑k=1Ntot(1−p)k−1p|Rf(Δf=0)||Rt(Δt=ktp)|

[0104] Fig. 11C schematically illustrates an example of an adaptive binning structure 1130 with a second predetermined channel group width according to one or more implementations of the claimed technology. However, not all of the depicted components may be required, and one or more implementations may include additional components not shown in the figure. Variations in the arrangement and type of components are possible without departing from the spirit or scope of the claims set forth herein. Additional components, different components, or fewer components may be provided.

[0105] In Fig. 11C, the adaptive global binning structure 1130 includes a first channel group 1132 (denoted as "KG 1"), a second channel group 1134-1 (denoted as "KG 4"), and a third channel group 1134-2 (denoted as "KG 4"). In some embodiments, the first channel group 1132 and the second channel group 1134-1 are separated by a predetermined time difference 1136, denoted as Δt. As shown in Fig. 11C, the second channel group 1134-1 is the same channel group as the third channel group 1134-2 (i.e., returned after a predetermined time for return). In the illustrated example of Fig. 11C, it is assumed that the first predetermined channel group width is equal to 2 MHz, with each channel group containing two channels. In this respect, the term R avg,i (BW i = 2, t p , Δf) can be calculated as follows: Ravg,i(BWi=2,tp,Δf)={|Rf(Δf=0)Rt(Δt=tp)|+|Rf(Δf=1)||Rt(Δt=tp)||R f(Δf=0)||Rt(Δt=2tp)|+|Rf(Δf=1)||Rt(Δt=2tp)||Rf(Δf=0)||Rt(t=3tp)| + |Rf(Δf=1)||Rt(Δt=3tp) | ⋮|Rf(Δf=0)||Rt(Δt=Nhtp)| + |Rf(Δf=1)||Rt(T=Nhtp)|with prob. pwith prob. (1−2p)pwith prob. (1−2p)2pwith prob. (1−2p)Nh−1p

[0106] In some manifestations, the time and frequency correlation expressions at Δf=0 represent the penalty for adding an adjacent channel to the computation. For example, in a channel group, the correlation of a channel with itself (i.e., Δf=0) occurs less frequently by adding adjacent channels to its channel group. In some manifestations, the time and frequency correlation expressions at Δf=1 represent the reward for adding an adjacent channel to the computation. For example, additional correlation can be determined by adding a neighboring node to a channel group.

[0107] In some implementations, the term R avg,i (BW i = 2, t p ) by calculating the expected value of R i (BW i = 2, t p , Δf) is formed by a change in frequency (Δf) and a change in time (Δt). For example, the calculation of the empirical value can be expressed as follows: Ravg,i(BWi=2,tp)=EΔf,Δt{Ri(BWi=2,tp,Δf)}=H(BWi=2,tp,Δf=0)+H(BWi=2,tp,Δf=1)

[0108] The term H(BW i , t p , Δf) can be expressed as follows: ∑k=1Ntot(1−BWip)k−1p|Rf(Δf)||Rt(Δt=ktp)

[0109] In some manifestations, the channel statistics (i.e., time and frequency correlation functions) are known a priori based on simulated channel models. In other manifestations, the time and frequency correlation functions can be calculated based on a measured channel response.

[0110] In some implementations, when new channel estimates are available, an update rule for a frequency correlation function can be expressed as follows: Rf,new(Δf=Δf′)=αRf,existing(Δf=Δf)+(1−α)CN(fi,t)CN*(fi−Δf′,t−tp)

[0111] Similarly, an update rule for a time correlation function can be expressed as follows: Rt, new(Δt=Δt′)=αRt, existing(Δt=Δt′)+(1−α)Cn(fi,t)CN*(fi,t−Δt′) where α is a forgetting factor.

[0112] When a new MRC angle is available, an adaptive, iterative global binning can be performed instead of recalculating the entire term R each time. avg For a frequency update, the calculation can be formulated as follows: Δf′=|f−f′| where the term f corresponds to the current frequency hopping frequency at a current time k and the term f' corresponds to the previous frequency hopping frequency at the time k - t p corresponds.

[0113] The complex, as H neu (BW i , t p , Δf = Δf') for the frequency update can be expressed as follows: Hnew(BWi,tp,Δf=Δf′)=|Rf,new(Δf=Δf′)||Rf,prev.(Δf=Δf′)|Hprev.(BWi,tp,Δf=Δf′)

[0114] The frequency update based on the current frequency hopping frequency and the previous frequency hopping frequency can be formulated as follows: Δu,f=Hnew(BWi,tp,Δf=Δf′)−Hprev.(BWi,tp,Δf=Δf′)

[0115] The term R avg,i (BW i , t p ) for the new MRC angle can be expressed as follows: Ravg,i(BWi,tp)= {Ravg,i(BWi,tp)+Δu,f×BWiif Δf′=0Ravg,i(BWi,tp)+2×(BWi−Δf′)×Δu,f×BWiif Δf′≤BWi−1 and Δf′≠0

[0116] For a time update, the calculation can be formulated as follows: k=Δt′Δt where the term Δt' corresponds to the time difference between the current time and the previous measurement taken for the current frequency hopping channel. If for some manifestations k ≤ N h , then an update is performed for all channel group widths (BW i ) at a permissible rate. Such a calculation can be formulated as follows: Δu,t=(|Rt, new(Δt=Δt′)|−|Rt, existing(Δt=Δt′)|)×(1−BWix p)k−1x px (BW ×|Rf(Δf=0)|+2∑Δf′=1BWi−1(BWi−Δf′)|Rf(Δf=Δf′)|)

[0117] For some manifestations, the mean correlation with the time update can be composed using the following formulation: Ravg,i(BWi,tp)=Ravg,i(BWi,tp)+Δu,t

[0118] In some implementations, adaptive local binning can be performed to select the channel group width to maximize the correlation of phase values ​​averaged over a number of channels in both the time and frequency dimensions. The optimization of the channel group width can be formulated as follows: BWopt=arg maxBW ∈{1,3,5,7,…}given tp,NKanRavg(BW,tp) =arg maxBW ∈{1,3,5,7,…}given tp,NKan1NKan(NKG×Ravg,b(BW,tp)+Ravg,e(BW,tp))

[0119] The optimization problem for local binning is similar to that for global binning, but the objective function R avg (BW, p) for the two binning methods is different. In some manifestations, the allowable channel width set for global binning includes consecutive channel group values ​​{1, 2, 3, 4, ..}, while the allowable channel width set for local binning includes odd channel group values ​​{1, 3, 5, 7, ..}.

[0120] In some implementations, the mean correlation per channel group (except for channel groups containing channels at the top and bottom of a band) can be R avg,b (BW i , t p ) can be formulated as follows: Ravg,b(BW,tp)=H(BW,tp,Δf=0)+2∑Δf′=1(BW−1) / 2H(BW,tp,Δf=Δf′)

[0121] The term H(BW, t p , Δf) is the same as for global binning (see, for example, equation (5)). In some manifestations, the mean correlation of channel groups containing channels at the upper and lower edges in a band can be R avg,e (BW i , tp ), as follows: Ravg,e(BW,tp)=[2+(BW−3)〚(BW>3)]H(BW,tp,Δf=0)+2∑Δf′=1BW−1H(BW,tp,Δ f=Δf′)+∑n=1(BW−3) / 3(2∑k=1BW−1−nH(BW,tp,Δf=k)+2∑l=1nH(BW,tp,Δf=l))

[0122] If the condition within the argument of (.) is met, then the value of the function is 1. Otherwise, it is 0. This condition can be expressed as follows: 〛(BW>3)={1if BW>30if BW≤3

[0123] If a new MRC angle is available, an adaptive, iterative local binning can be performed instead of R avg,b (BW, t p ) and R avg,e (BW, t p ). For a frequency update, the calculation can be formulated as follows: Δf′=|f−f′| where the term f corresponds to the current frequency hopping frequency at a current time k and the term f' corresponds to the previous frequency hopping frequency at the time k - t pcorresponds.

[0124] The complex, as H neu (BW, t p , Δf = Δf') for the frequency update can be expressed as follows: Hnew(BW,tp,Δf=Δf′)=|Rf,new(Δf=Δf′)||Rf,prev.(Δf=Δf′)|Hprev.(BW,tp,Δf=Δf′)

[0125] The frequency update based on the current frequency hopping frequency and the previous frequency hopping frequency can be formulated as follows: Δu,f=Hnew(BW,tp,Δf=Δf′)−Hprev.(BW,tp,Δf=Δf′)

[0126] The term R avg,b (BW, t p ) for the new MRC angle can be expressed as follows: Ravg,b(BW,tp)=Ravg,b(BW,tp)+Δf if Δf′≤BW−12

[0127] The term R avg,e (BW, t p ) for the new MRC angle can be expressed as follows: ravg,e(BW,tp)={Ravg,e(BW,tp)+[2+(BW−3)〚(BW>3)]×Δu,f if Δf′=0Ravg,e(BW,tp)+[2+2∑k=1BW−32〚(Δf′≤k)+2∑n=1BW−32〚(Δf′≤(BW−1−n))]×Δu,fif Δf′≤BW−1 and Δf′≠0

[0128] If for some manifestations k ≤ N h , then an update is performed for all channel group widths (BW i ) at a permissible rate. Such a calculation can be formulated as follows: Δu,t=(|Rt,new(Δt=Δt′)|−|Rt,previous(Δt=Δt′)|)×(1−BWi×p)k−1×p×(|Rf(Δf=0)|+2∑Δf′=1(BW−1) / 2|Rf(Δf=Δf′)| where the term Δt' corresponds to the time difference between the current time and the previous measurement taken for the current frequency hopping channel.

[0129] In some manifestations, the mean correlation per channel group (except for channel groups containing channels at the top and bottom of a band) exhibiting the time update can be composed using the following formulation: Ravg,b(BW,tp)=Ravg,b(BW,tp)+Δu,t Δue,t=(|Rt,new(Δt=Δt′)|−|Rt,previous(Δt=Δt′)|)x(1−BWix p)k−1x px 2 x([1+(BW−3)2〚(BW>3)]|Rf(Δf=0)|+∑Δf′=1BW−1|Rf(Δf=Δf′)|+∑k=1(BW−1−n)|Rf(Δf=k)|+∑l=1n|Rf(Δf=l)|))

[0130] In some manifestations, the mean correlation of channel groups containing channels at the top and bottom edges in a band, which exhibits the time update, can be composed via the following formulation: Ravg,e(BW,tp)=Ravg,e(BW,tp)+Δue,t

[0131] Fig. 11D illustrates a graphical representation 1140 of an exemplary curve 1142 representing the average correlation of phase values ​​for different channel group widths according to one or more implementations of the claimed technology. In Fig. 11D, the Doppler frequency is 3 Hz, and the interval is set to 1.25 ms. As in Fig. 11D, the mean correlation value initially increases with increasing values ​​for the channel group width and then begins to decrease due to penalty terms in the calculation of the mean correlation. Consequently, there is an optimal channel group width that results in the highest mean correlation. In the plot 1140, an optimal channel group width exists within a region 1144 of the curve 1142 where the mean correlation value is highest. For example, the optimal channel group width is 3 MHz. In some manifestations, the optimal channel group width for a case where the AFH includes 20 channels may be in a range of 2 MHz to 4 MHz, but may vary depending on packet spacing, the number of frequency hopping channels, and / or wireless channel conditions.

[0132] Fig. Figure 12 illustrates a block diagram of a process 1200 for binning-based transmit beamforming according to one or more implementations of the claimed technology. For purposes of explanation, the process 1200 will be described herein primarily with reference to the wireless communication portion 200 of Fig. 2 in accordance with the binning structures of Fig. 7A to Fig. 7C, Fig. 8A to Fig. 8C and Fig. 11A to Fig. 11C. However, the process 1200 is not limited to the wireless communication part 200 of Fig. 2, and one or more blocks (or operations) of process 1200 may be performed by one or more other components or circuits of wireless communication portion 200, such as transmitter 201. Further, for purposes of explanation, the blocks of process 1200 are described herein as executing serially or linearly. However, multiple blocks of process 1200 may occur in parallel. Furthermore, the blocks of process 1200 need not be executed in the order shown, and / or one or more of the blocks of process 1200 need not be executed and / or may be replaced by other operations.

[0133] The process 1200 begins at step 1202, where a time correlation and a frequency correlation of a beamforming vector estimated from a previous transmission on a current frequency-hopping channel and one or more adjacent channels are determined. Next, at step 1204, one or more beamforming weights are determined for the current frequency-hopping channel. Subsequently, at step 1206, a signal is provided for transmission using transmit beamforming based on one or more beamforming weights.

[0134] Fig. 13 schematically illustrates an electronic system 1300 with which one or more implementations of the claimed technology may be implemented. For example, the electronic system 1300 may be a network device, a media converter, a desktop computer, a laptop computer, a tablet computer, a server, a switch, a router, a base station, a receiver, a telephone, or generally any electronic device that transmits signals over a network. Such an electronic system 1300 includes various types of computer-readable media and interfaces for various other types of computer-readable media. In one or more implementations, the electronic system 1300 is, or is a component of, one or more of the wireless communication devices 130-127 and the BT devices 150-159.The electronic system 1300 includes a bus 1308, one or more processing units 1312, a system memory 1304, a read-only memory (ROM) 1310, a permanent storage device 1302, an input device interface 1314, an output device interface 1306, and a network interface 1316, or subsets and variations thereof.

[0135] Bus 1308 collectively embodies all system buses, peripheral buses, and chipset buses that communicatively connect the numerous internal devices of electronic system 1300. In one or more implementations, bus 1308 communicatively connects one or more processing units 1312 to ROM 1310, system memory 1304, and persistent storage device 1302. From these various storage units, one or more processing units 1312 retrieve instructions to be executed and data to be processed in order to perform the processes of the claimed disclosure. The one or more processing units 1312 may be a single processor or a multi-core processor in various implementations.

[0136] ROM 1310 stores static data and instructions required by the one or more processing units 1312 and other modules of the electronic system. Permanent storage device 1302, on the other hand, is a read-write memory device. Permanent storage device 1302 is a non-volatile memory unit in which instructions and data are stored even when electronic system 1300 is powered off. In one or more implementations of the claimed disclosure, a mass storage device (such as a magnetic disk or an optical disk and the corresponding disk drive) may be used as permanent storage device 1302.

[0137] In other implementations, a removable storage device (such as a floppy disk, a flash drive, and the corresponding disk drive) is used as the permanent storage device 1302. Like the permanent storage device 1302, the system memory 1304 is a read-write memory device. However, unlike the permanent storage device 1302, the system memory 1304 is a volatile read-write memory, such as random access memory. The system memory 1304 stores any of the instructions and data required by the one or more processing units 1312 at runtime. In one or more implementations, the processes of the claimed disclosure are stored in the system memory 1304, the permanent storage device 1302, and / or the ROM 1310.From these various memory units, the one or more processing units 1312 retrieve instructions to be executed and data to be processed to perform the processes of one or more implementations.

[0138] Bus 1308 also connects to input device interface 1314 and output device interface 1306. Input device interface 1314 allows a user to communicate information and select commands to the electronic system. Input devices used with input device interface 1314 include, for example, alphanumeric keyboards and pointing devices (also referred to as "cursor control devices"). Output device interface 1306 enables, for example, the display of images generated by electronic system 1300.Output devices used with the output device interface 1306 include, for example, printers and display devices such as a liquid crystal display (LCD), a light-emitting diode (LED), an organic light-emitting diode (OLED), a flexible display, a flat panel display, a solid-state display, a projector, or any other device for outputting information. One or more implementations include devices that function as both an input and output device, such as a touchscreen.In these implementations, the feedback provided to the user may be any form of sensory feedback, such as visual feedback, auditory feedback, or tactile feedback; and input from the user may be received in any form, including auditory input, speech input, or tactile input.

[0139] Finally, as in Fig.13, bus 1308 also connects electronic system 1300 to one or more networks (not shown) via one or more network interfaces 1316. In this manner, the computer may be part of one or more networks of computers (such as a local area network ("LAN"), a wide area network ("WAN"), an intranet, or a network of networks, such as the Internet). Any or all components of electronic system 1300 may be used in connection with the claimed disclosure.

[0140] Implementations within the scope of the present disclosure may be realized in part or in whole using a tangible computer-readable storage medium (or multiple tangible computer-readable storage media of one or more types) encoding one or more instructions. The tangible computer-readable storage medium may also be persistent in nature.

[0141] The computer-readable storage medium may be any storage medium that can be read, written to, or otherwise accessed by a general-purpose or special-purpose computing device, including any processing electronics and / or processing circuitry capable of executing instructions. For example, the computer-readable medium may include any volatile semiconductor memory, such as, but not limited to, RAM, DRAM, SRAM, T-RAM, Z-RAM, and TTRAM. The computer-readable medium may also include any non-volatile semiconductor memory, such as ROM, PROM, EPROM, EEPROM, NVRAM, Flash memory, nvSRAM, FeRAM, FeTRAM, MRAM, PRAM, CBRAM, SONOS, RRAM, NRAM, Racetrack memory, FJG, and Millipede memory.

[0142] Furthermore, the computer-readable storage medium may comprise any non-semiconductor storage, such as optical disk storage, magnetic disk storage, magnetic tape, other magnetic storage devices, or any other medium capable of storing one or more instructions. In some implementations, the tangible computer-readable storage medium may be directly coupled to a computing device, while in other implementations, the tangible computer-readable storage medium may be indirectly coupled to a computing device, for example, via one or more wired connections, one or more wireless connections, or any combination thereof.

[0143] Instructions may be directly executable or may be used to develop executable instructions. For example, instructions may be embodied as executable or non-executable machine code, or as high-level language instructions that can be compiled to produce executable or non-executable machine code. Furthermore, instructions may also be embodied as data or may include data. Computer-executable instructions may also be organized in any format, including routines, subroutines, programs, data structures, objects, modules, applications, applets, functions, and so on. As those skilled in the art will recognize, details, including, but not limited to, the number, structure, order, and organization of instructions, may vary considerably without variation in the underlying logic, function, processing, and output.

[0144] While the above discussion primarily concerns microprocessors or multi-core processors executing software, one or more implementations are carried out using one or more integrated circuits, such as ASICs (Application Specific Integrated Circuits) or FPGAs (Field Programmable Gate Arrays). In one or more implementations, such integrated circuits execute instructions stored within the circuit itself.

[0145] It should be understood by those skilled in the art that the various illustrative blocks, modules, elements, components, methods, and algorithms described in this document may be implemented in electronic hardware, computer software, or a combination of both. To illustrate this interchangeability of hardware and software, various illustrative blocks, modules, elements, components, methods, and algorithms have been described above generally in terms of their functionality. Whether such functionality is implemented in hardware or software depends on the particular application and design constraints to which the overall system is subject. Those skilled in the art may implement the described functionality in different ways for each specific application.Various components and blocks may be arranged in different ways (for example, arranged in a different order or divided in a different way) without all this departing from the scope of the claimed technology.

[0146] It should be understood that any specific order or hierarchy of blocks in the disclosed processes is an illustration of example approaches. It should be understood that the specific order or hierarchy of blocks in the processes may be rearranged based on design preferences, or that all of the illustrated blocks may be executed. Any of the blocks may be executed concurrently. In one or more implementations, multitasking and parallel processing may be advantageous. Furthermore, the separation of various system components in the embodiments described above should not be understood to require such separation in all embodiments, and it should be understood that the described program components and systems may generally be integrated together into a single software product or packaged into multiple software products.

[0147] As used in this specification and any claims of this patent application, the terms "base station," "receiver," "computer," "server," "processor," and "memory" all refer to electronic or other technological devices. These terms do not refer to human beings or groups of human beings. For purposes of this specification, the terms "display" and "display" mean displaying on an electronic device.

[0148] The predicative phrases "configured to," "operable to," and "programmed to" do not imply any particular tangible or intangible change to an item, but rather are intended to be used interchangeably. In one or more implementations, stating that a processor is configured to monitor and control an operation or a component may also mean that the processor is programmed to monitor and control the operation, or that the processor is operable to monitor and control the operation. Similarly, stating that a processor is configured to execute code may be construed to mean that a processor is programmed to execute code, or that it is operable to execute code.

Claims

[1] A method for transmit beamforming, the method comprising: Determining a time correlation and a frequency correlation of a beamforming vector estimated from a previous transmission on a current frequency hopping channel and one or more adjacent channels; Determining one or more beamforming weights for the current frequency hopping channel; and Providing a signal using transmit beamforming based on one or more beamforming weights for a transmit operation. [2] The method of claim 1, wherein the one or more adjacent channels have a beamforming vector that corresponds to that of the current frequency hopping channel based on a channel coherence bandwidth. [3] The method of claim 1, further comprising: Combining the current frequency hopping channel and one or more neighboring channels into the same channel group (bin). [4] The method of claim 1, further comprising: Obtaining a latest estimate of the beamforming vector in a channel group, wherein the time correlation and the frequency correlation of the beamforming vector are determined from the latest estimate. [5] The method of claim 1, further comprising: Forming a channel group comprising a number of channels, wherein the number of channels combined in a channel group is determined by means of a channel group width parameter. [6] The method of claim 1, further comprising: Determining whether a new estimate of the beamforming vector is obtained from at least one of a plurality of channels grouped in a same channel group; and Updating a beamforming vector in each channel group of a plurality of channel groups when the new estimate of the beamforming vector is obtained. [7] The method of claim 1, further comprising: Combining consecutive channels into a plurality of channel groups for a first type of binning architecture, each of the plurality of channel groups excluding common channels. [8] The method of claim 1, further comprising: Combining consecutive channels into a plurality of channel groups for a second type of binning architecture, wherein each channel group of the plurality of channel groups is formed locally around a center indicated by a respective one of the consecutive channels. [9] A computer program product comprising instructions stored in a tangible, computer-readable storage medium, the instructions comprising: Instructions for determining a time correlation and a frequency correlation of a beamforming vector estimated from a previous transmission on a current frequency hopping channel and one or more adjacent channels; Instructions for combining the current frequency hopping channel and the one or more adjacent channels into a same channel group; Instructions for determining one or more beamforming weights in the same channel group for the current frequency hopping channel; and Instructions for providing a transmit signal using the one or more beamforming weights for a transmit operation. [10] Apparatus comprising: processing circuitry configured to: Determining a time correlation and a frequency correlation of a beamforming vector estimated from a previous transmission on a current frequency hopping channel and one or more adjacent channels; Combining the current frequency hopping channel and the one or more adjacent channels into a channel group; Determining one or more beamforming weights for the current frequency hopping channel; and Providing a transmit signal for a transmit operation using the one or more beamforming weights.

Images