WO2023140852A1

WO2023140852A1 - Apparatus and method implementing frequency offset estimation

Info

Publication number: WO2023140852A1
Application number: PCT/US2022/013188
Authority: WO
Inventors: Ping Hou; Jian Gu
Original assignee: Zeku, Inc.
Priority date: 2022-01-20
Filing date: 2022-01-20
Publication date: 2023-07-27

Abstract

A baseband chip and a method implementing frequency offset (TO) estimation at a receiver are provided. In certain embodiments, the baseband chip includes a primary synchronization signal (PSS) processing circuit configured to obtain a PSS symbol through an antenna from a transmitter over a channel. The PSS symbol is associated with a time domain. The PSS symbol is inputted to a probability function. The probability function includes a variable of FO and matrices indicating associated with at least one of the channel, the transmitter, or the receiver. The probability function is maximized according to the variable of FO to obtain a coarse FO estimation for the antenna. The probability function may be configured to optimize a detection rate of a transmitted PSS at the transmitter.

Description

APPARATUS AND METHOD IMPLEMENTING FREQUENCY OFFSET ESTIMATION

BACKGROUND

[0001] Embodiments of the present disclosure relate to apparatus and method for wireless communication.

[0002] Wireless communication systems are widely deployed to provide various telecommunication services such as telephony, video, data, messaging, and broadcasts. In wireless communication systems, such as the 4th-generation (4G) Long Term Evolution (LTE) or the 5 th- generation (5G) New Radio (NR), various mechanisms are defined and introduced for cell search and selection. During the process, a user equipment acquires synchronization with a cell both in a time domain and in a frequency domain. Frequency offset (FO) and phase noise, however, may cause loss of the synchronization performance.

SUMMARY

[0003] Embodiments of apparatus and method implementing frequency offset (FO) estimation are disclosed herein.

[0004] According to one aspect of the present disclosure, a baseband chip implementing frequency offset (FO) estimation at a receiver is provided. The baseband chip may include a primary synchronization signal (PSS) processing circuit configured to obtain a PSS symbol through an antenna from a transmitter over a channel. The PSS symbol may be associated with a time domain and inputted to a probability function. The probability function may include a variable of FO and matrices indicating knowledge associated with at least one of the channel, the transmitter, or the receiver. The probability function may be maximized according to the variable of FO to obtain a coarse FO estimation for the antenna. The probability function may be configured to optimize a detection rate of a transmitted PSS symbol at the transmitter.

[0005] According to another aspect of the present disclosure, a baseband chip implementing frequency offset (FO) estimation at a receiver is provided. The baseband chip may include a primary synchronization signal (PSS) processing circuit configured to process PSS symbols based on a probability function. The PSS symbols may be obtained through a plurality of antennas from a transmitter over a channel and associated with a time domain. The probability function may include a variable of FO and matrices indicating knowledge associated with at least one of the channel, the transmitter, or the receiver. The probability function may be maximized according to the variable of FO to obtain a final coarse FO estimation for the plurality of antennas. The baseband chip may further include a processor operatively coupled to the plurality of antennas and memory storing instructions. Execution of the instructions may cause the processor to process secondary synchronization signal (SSS) symbols and the final coarse FO estimation to obtain a final fine FO estimation based on the probability function. The probability function may be configured to optimize detection rates of a transmitted PSS symbol and a transmitted SSS symbol at the transmitter.

[0006] According to still another aspect of the present disclosure, a method for frequency offset (FO) estimation implemented at a receiver is provided. The method may include obtaining a PSS symbol through an antenna from a transmitter over a channel. The PSS symbol may be associated with a time domain and inputted to a probability function. The probability function may include a variable of FO and matrices indicating knowledge associated with at least one of the channel, the transmitter, or the receiver. The probability function may be maximized according to the variable of FO to obtain a coarse FO estimation for the antenna. The probability function being configured to optimize a detection rate of a transmitted PSS symbol at the transmitter.

BRIEF DESCRIPTION OF THE DRAWINGS

[0007] The accompanying drawings, which are incorporated herein and form a part of the specification, illustrate embodiments of the present disclosure and, together with the description, further serve to explain the principles of the present disclosure and to enable a person skilled in the pertinent art to make and use the present disclosure.

[0008] FIG. 1 illustrates a block diagram of coarse frequency offset estimation, according to some embodiments of the present disclosure.

[0009] FIG. 2 illustrates an exemplary wireless network, according to some embodiments of the present disclosure.

[0010] FIG. 3 illustrates a block diagram of an exemplary node, according to some embodiments of the present disclosure.

[0011] FIG. 4 illustrates a block diagram of an apparatus including a baseband chip, a radio frequency (RF) chip, and a host chip, according to some embodiments of the present disclosure. [0012] FIG. 5 illustrates a block diagram of an expanded view of a portion of the baseband chip of FIG. 4, according to some embodiments of the present disclosure.

[0013] FIG. 6 illustrates a block diagram of a synchronization signal block, according to some embodiments of the present disclosure.

[0014] FIG. 7 illustrates a flow diagram of an exemplary method of frequency offset estimation of a baseband chip, according to some embodiments of the present disclosure.

[0015] FIG. 8 illustrates a flow diagram of another exemplary method of frequency offset estimation of a baseband chip, according to some embodiments of the present disclosure.

[0016] FIG. 9 illustrates simulation results, using matched-filter and MAP approaches over an extended typical urban channel model, in view of mean square errors of residual FOs, according to some embodiments of the present disclosure.

[0017] Embodiments of the present disclosure will be described with reference to the accompanying drawings.

DETAILED DESCRIPTION

[0018] Although specific configurations and arrangements are discussed, it should be understood that this is done for illustrative purposes only. A person skilled in the pertinent art will recognize that other configurations and arrangements can be used without departing from the spirit and scope of the present disclosure. It will be apparent to a person skilled in the pertinent art that the present disclosure can also be employed in a variety of other applications.

[0019] It is noted that references in the specification to “one embodiment,” “an embodiment,” “an example embodiment,” “some embodiments,” “certain embodiments,” “other embodiments,” etc., indicate that the embodiment described may include a particular feature, structure, or characteristic, but every embodiment may not necessarily include the particular feature, structure, or characteristic. Moreover, such phrases do not necessarily refer to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it would be within the knowledge of a person skilled in the pertinent art to effect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described.

[0020] In general, terminology may be understood at least in part from usage in context. For example, the term “one or more” as used herein, depending at least in part upon context, may be used to describe any feature, structure, or characteristic in a singular sense or may be used to describe combinations of features, structures, or characteristics in a plural sense. Similarly, terms, such as “a,” “an,” or “the,” again, may be understood to convey a singular usage or to convey a plural usage, depending at least in part upon context. In addition, the term “based on” may be understood as not necessarily intended to convey an exclusive set of factors and may, instead, allow for existence of additional factors not necessarily expressly described, again, depending at least in part on context.

[0021] Various aspects of wireless communication systems will now be described with reference to various apparatus and methods. These apparatus and methods will be described in the following detailed description and illustrated in the accompanying drawings by various blocks, modules, units, components, circuits, steps, operations, processes, algorithms, etc. (collectively referred to as “elements”). These elements may be implemented using electronic hardware, firmware, computer software, or any combination thereof. Whether such elements are implemented as hardware, firmware, or software depends upon the particular application and design constraints imposed on the overall system.

[0022] The techniques described herein may be used for various wireless communication networks, such as code division multiple access (CDMA) system, time division multiple access (TDMA) system, frequency division multiple access (FDMA) system, orthogonal frequency division multiple access (OFDMA) system, single-carrier frequency division multiple access (SC- FDMA) system, wireless local area network (WLAN) system, and other networks. The terms “network” and “system” are often used interchangeably. A CDMA network may implement a radio access technology (RAT), such as Universal Terrestrial Radio Access (UTRA), evolved UTRA (E-UTRA), CDMA 2000, etc. A TDMA network may implement a RAT, such as the Global System for Mobile Communications (GSM). An OFDMA network may implement a RAT, such as LTE or NR. A WLAN system may implement a RAT, such as Wi-Fi. The techniques described herein may be used for the wireless networks and RATs mentioned above, as well as other wireless networks and RATs.

[0023] Synchronization, either in wired or wireless communication systems, plays a key role in communication networks. Synchronization enables successful communication between nodes on the networks, particularly being vital for wireless communication. Wireless communication systems may establish synchronization, e.g., at an initial cell search, at a discontinuous reception (DRX) mode, at a cell measurement, at a frequency scan, and/or at a cell re-selection.

[0024] The cell search is an initial process by which any available base station is identified by a user equipment (UE) that requests the communication. During the process, the UE acquires synchronization with the base station (a cell) in a time domain as well as in a frequency domain, and a cell identification (ID) of the cell is decoded. After the cell search procedure consisting of serial processes of synchronization, the UE obtains the precise timing of the symbols from the cell, which is crucial to, e.g., subsequent demodulation of downlink signals and transmission of uplink signals.

[0025] In certain scenarios, the UE may switch to a DRX mode when, e.g., the UE does not need to continuously monitor all possible paging or other channels if downlink data packets are only received intermittently. Under these scenarios, in order for power saving, the DRX mode at the UE introduces the mechanisms that allow the UE to turn off the radio frequency (RF) chip during a predefined time period. While the RF chip is switched off, however, the UE may lose the synchronization as established earlier. As a result, the receiver at the UE may need to wake up earlier before the predefined time period, such that the system can perform the synchronization again.

[0026] In wireless communication systems, e.g., 4G LTE or 5GNR, such synchronization is mainly performed using the synchronization signals that include a primary synchronization signal (PSS) and a secondary synchronization signal (SSS). On the basis of the successful synchronization, subsequent procedures, e.g., the channel estimation and physical broadcast channel (PBCH) detection/decoding, may be accomplished and/or enhanced. Accordingly, the performance of the channel may be improved.

[0027] Frequency offset (FO) and phase noise, however, may cause a loss of synchronization performance. Usually, the FO emerges because the frequencies at a transmitter and a receiver mismatch. The FO adversely influences the effectiveness of the communication systems, results in a reduction of desired signal amplitude, and introduces the inter-carrier interference (ICI). With the orthogonal frequency division multiplexing (OFDM) systems, the orthogonality between the sub-carriers may be lost due to the FO. Therefore, it may be advantageous if the FO can be estimated and compensated before receiving and processing the data. Once the FO of the received signals is better estimated, it becomes more possible to receive and process the data correctly.

[0028] FIG. 1 illustrates a block diagram of coarse FO estimation. For simplicity, FIG. 1 merely shows in part, regarding the PSS processing for the coarse FO estimation, of the scheme. In certain instances, an approach similar to FIG. 1 may be applied to the SSS processing to obtain fine FO estimation. In this approach, the PSS symbol in the OFDM symbols, detected through PSS detector 102, may be segmented into multiple portions. Accordingly, the coarse FO can be estimated based on phase differences, over the frequency domain, between the multiple portions of the PSS symbol. For example, as illustrated in FIG. 1, a first half and a second half of the PSS symbol may be fed into respective matched filters 104 where the demanded reference signals can serve as the bounds of the matched filters, and the coarse FO can be estimated through coarse FO estimator 106 according to comparison on the outputs of the matched filters.

[0029] In this approach, the outputs of the matched filters can be sensitive to and thus influenced by the phase noise. In particular, in certain cases where the signal-to-noise ratio (SNR) of the channel is low, the phase noise may substantially dominate the real data and thus contaminate the samples. As a result, the coarse FO estimation based on the samples may not be reliable. It turns out that this estimation approach may be, to some extent, applicable for the channels within a relatively narrow range of the SNR. Another remedy is that before the received signals are fed to the matched filters, they may be compensated in advance with some FO hypotheses. The introduction of the FO hypotheses to the system can incur extra costs in view of additional multiplication operations, larger cycle time, and more power consumption. Meanwhile, the requirement for smaller residual FO through the matched-filter approach also places more and higher restrictions in the design of the RF chip, which is also translated into more cost. The matched-filter approach may be optimal, specifically, with the additive white Gaussian noise (AWGN) channel configured to maximize the SNR. For some fading channels, the approach may not yield a good result at a low SNR or low signal-to-interference ratio (SINR).

[0030] To solve these and other problems, the present disclosure provides another approach to estimate carrier FO using maximum a posterior (MAP) estimation with the PSS and/or the SSS. In the MAP estimation, a probability density function indicative of a priori knowledge associated with at least one of the channel, the receiver, or the transmitter is being maximized. In view of the information theory, this method can approach a channel capacity. Hence, compared to the matched- filter approach at the same SNR, the MAP estimation can provide a better estimation result, even for the fading channels. In particular, with higher-complexity iterative implementation, this approach can greatly enhance the FO estimation. Consequently, it can improve the performance of the cell detection and the channel estimation significantly, even when the SNR is relatively slow. This scheme can also be applied to cell measurement. The FO estimation and compensation usually correlate to the quality of received signals, and a better cell measurement can lead to a successful cell selection.

[0031] FIG. 2 illustrates an exemplary wireless network 200, in which certain aspects of the present disclosure may be implemented, according to some embodiments of the present disclosure. As shown in FIG. 2, wireless network 200 may include a network of nodes, such as a user equipment (UE) 202, an access node 204, and a core network element 206. UE 202 may be any terminal device, such as a mobile phone, a desktop computer, a laptop computer, a tablet, a vehicle computer, a gaming console, a printer, a positioning device, a wearable electronic device, a smart sensor, or any other device capable of receiving, processing, and transmitting information, such as any member of a vehicle to everything (V2X) network, a cluster network, a smart grid node, or an Intemet-of-Things (loT) node. It is understood that UE 202 is illustrated as a mobile phone simply by way of illustration and not by way of limitation.

[0032] Access node 204 may be a device that communicates with UE 202, such as a wireless access point, a base station (BS), a Node B, an enhanced Node B (eNodeB or eNB), a next-generation NodeB (gNodeB or gNB), a cluster master node, or the like. Access node 204 may have a wired connection to UE 202, a wireless connection to UE 202, or any combination thereof. Access node 204 may be connected to UE 202 by multiple connections, and UE 202 may be connected to other access nodes in addition to access node 204. Access node 204 may also be connected to other user equipments. It is understood that access node 204 is illustrated by a radio tower by way of illustration and not by way of limitation.

[0033] Core network element 206 may serve access node 204 and UE 202 to provide core network services. Examples of core network element 206 may include a home subscriber server (HSS), a mobility management entity (MME), a serving gateway (SGW), or a packet data network gateway (PGW). These are examples of core network elements of an evolved packet core (EPC) system, which is a core network for the LTE system. Other core network elements may be used in LTE and in other communication systems. In some embodiments, core network element 206 includes an access and mobility management function (AMF) device, a session management function (SMF) device, or a user plane function (UPF) device, of a core network for the NR. system. It is understood that core network element 206 is shown as a set of rack-mounted servers by way of illustration and not by way of limitation.

[0034] Core network element 206 may connect with a large network, such as the Internet 208, or another Internet Protocol (IP) network, to communicate packet data over any distance. In this way, data from UE 202 may be communicated to other user equipments connected to other access points, including, for example, a computer 210 connected to Internet 208, for example, using a wired connection or a wireless connection, or to a tablet 212 wirelessly connected to Internet 208 via a router 214. Thus, computer 210 and tablet 212 provide additional examples of possible user equipments, and router 214 provides an example of another possible access node. [0035] A generic example of a rack-mounted server is provided as an illustration of core network element 206. However, there may be multiple elements in the core network including database servers, such as a database 216, and security and authentication servers, such as an authentication server 218. Database 216 may, for example, manage data related to user subscription to network services. A home location register (HLR) is an example of a standardized database of subscriber information for a cellular network. Likewise, authentication server 218 may handle authentication of users, sessions, and so on. In the NR system, an authentication server function (AUSF) device may be the specific entity to perform user equipment authentication. In some embodiments, a single server rack may handle multiple such functions, such that the connections between core network element 206, authentication server 218, and database 216, may be local connections within a single rack.

[0036] Each element in FIG. 2 may be considered a node of wireless network 200. More detail regarding the possible implementation of a node is provided by way of example in the description of a node 300 in FIG. 3. Node 300 may be configured as UE 202, access node 204, or core network element 206 in FIG. 2. Similarly, node 300 may also be configured as computer 210, router 214, tablet 212, database 216, or authentication server 218 in FIG. 2. As shown in FIG. 3, node 300 may include a processor 302, a memory 304, and a transceiver 306. These components are shown as connected to one another by a bus, but other connection types are also permitted. When node 300 is UE 202, additional components may also be included, such as a user interface (UI), sensors, and the like. Similarly, node 300 may be implemented as a blade in a server system when node 300 is configured as core network element 206. Other implementations are also possible.

[0037] Transceiver 306 may include any suitable device for sending and/or receiving data. Node 300 may include one or more transceivers, although only one transceiver 306 is shown for simplicity of illustration. An antenna 308 is shown as a possible communication mechanism for node 300. Multiple antennas and/or arrays of antennas may be utilized for receiving multiple spatially multiplex data streams. Additionally, examples of node 300 may communicate using wired techniques rather than (or in addition to) wireless techniques. For example, access node 204 may communicate wirelessly to UE 202 and may communicate by a wired connection (for example, by optical or coaxial cable) to core network element 206. Other communication hardware, such as a network interface card (NIC), may be included as well.

[0038] As shown in FIG. 3, node 300 may include processor 302. Although only one processor is shown, it is understood that multiple processors can be included. Processor 302 may include microprocessors, microcontroller units (MCUs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), programmable logic devices (PLDs), state machines, gated logic, discrete hardware circuits, and other suitable hardware configured to perform the various functions described throughout the present disclosure. Processor 302 may be a hardware device having one or more processing cores. Processor 302 may execute software. Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executables, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise. Software can include computer instructions written in an interpreted language, a compiled language, or machine code. Other techniques for instructing hardware are also permitted under the broad category of software. [0039] As shown in FIG. 3, node 300 may also include memory 304. Although only one memory is shown, it is understood that multiple memories can be included. Memory 304 can broadly include both memory and storage. For example, memory 304 may include random-access memory (RAM), read-only memory (ROM), static RAM (SRAM), dynamic RAM (DRAM), ferroelectric RAM (FRAM), electrically erasable programmable ROM (EEPROM), CD-ROM or other optical disk storage, hard disk drive (HDD), such as magnetic disk storage or other magnetic storage devices, Flash drive, solid-state drive (SSD), or any other medium that can be used to carry or store desired program code in the form of instructions that can be accessed and executed by processor 302. Broadly, memory 304 may be embodied by any computer-readable medium, such as a non-transitory computer-readable medium.

[0040] Processor 302, memory 304, and transceiver 306 may be implemented in various forms in node 300 for performing wireless communication functions. In some embodiments, processor 302, memory 304, and transceiver 306 of node 300 are implemented (e.g., integrated) on one or more system-on-chips (SoCs). In one example, processor 302 and memory 304 may be integrated on an application processor (AP) SoC (sometimes known as a “host,” referred to herein as a “host chip”) that handles application processing in an operating system (OS) environment, including generating raw data to be transmitted. In another example, processor 302 and memory 304 may be integrated on a baseband processor (BP) SoC (sometimes known as a “modem,” referred to herein as a “baseband chip”) that converts the raw data, e.g., from the host chip, to signals that can be used to modulate the carrier frequency for transmission, and vice versa, which can run a real-time operating system (RTOS). In still another example, processor 302 and transceiver 306 (and memory 304 in some cases) may be integrated on an RF SoC (sometimes known as a “transceiver,” referred to herein as an “RF chip”) that transmits and receives RF signals with antenna 308. It is understood that in some examples, some or all of the host chip, baseband chip, and RF chip may be integrated as a single SoC. For example, a baseband chip and an RF chip may be integrated into a single SoC that manages all the radio functions for cellular communication.

[0041] Referring back to FIG. 2, in some embodiments, any suitable node of wireless network 200 that receives signals from another node (e.g., from access node 204 to UE 202 or vice versa) may implement the FO estimation in a baseband chip using the proposed MAP estimation. Compared with other approaches, the scheme can work at a relatively low SNR while achieving better synchronization performance. Even under a poor environment, with the precise estimation of the FO, the channel estimation can be accomplished, and the performance of the data channel can be improved.

[0042] FIG. 4 illustrates a block diagram of an apparatus 400 including a baseband chip 402, an RF chip 404, and a host chip 406, according to some embodiments of the present disclosure. Apparatus 400 may be an example of any suitable node of wireless network 200 in FIG. 2, such as UE 202 or access node 204. As shown in FIG. 4, apparatus 400 may include a baseband chip 402, RF chip 404, a host chip 406, and one or more antennas 410. In some embodiments, baseband chip 402 is implemented by processor 302 and memory 304, and RF chip 404 is implemented by processor 302, memory 304, and transceiver 306, as described above with respect to FIG. 3. Besides the on-chip memory (also known as “internal memory” or “local memory,” e.g., registers, buffers, or caches) on each chip 402, 404, or 406, apparatus 400 may further include an external memory 408 (e.g., the system memory or main memory) that can be shared by each chip 402, 404, or 406 through the system/main bus. Although baseband chip 402 is illustrated as a standalone SoC in FIG. 4, it is understood that in one example, baseband chip 402 and RF chip 404 may be integrated as one SoC; in another example, baseband chip 402 and host chip 406 may be integrated as one SoC; in still another example, baseband chip 402, RF chip 404, and host chip 406 may be integrated as one SoC, as described above.

[0043] In the uplink, host chip 406 may generate raw data and send it to baseband chip 402 for encoding, modulation, and mapping. Baseband chip 402 may also access the raw data generated by host chip 406 and stored in external memory 408, for example, using the direct memory access (DMA). Baseband chip 402 may first encode (e.g., by source coding and/or channel coding) the raw data and modulate the coded data using any suitable modulation techniques, such as multi-phase shift keying (MPSK) modulation or quadrature amplitude modulation (QAM). Baseband chip 402 may perform any other functions, such as symbol or layer mapping, to convert the raw data into a signal that can be used to modulate the carrier frequency for transmission. In the uplink, baseband chip 402 may send the modulated signal to RF chip 404. RF chip 404 may convert the modulated signal in the digital form into analog signals, i.e., RF signals, and perform any suitable front-end RF functions, such as filtering, digital pre-distortion, up-conversion, or sample-rate conversion. Antenna 410 (e.g., an antenna array) may transmit the RF signals provided by a transmitter TX of RF chip 404.

[0044] In downlink, antenna 410 may receive RF signals and pass the RF signals to the receiver RX of RF chip 404. RF chip 404 may perform any suitable front-end RF functions, such as filtering, DC offset compensation, IQ imbalance compensation, down-conversion, or samplerate conversion, and convert the RF signals into low-frequency digital signals (baseband signals) that can be processed by baseband chip 402. In downlink, baseband chip 402 may demodulate and decode the baseband signals to extract raw data that can be processed by host chip 406. Baseband chip 402 may perform additional functions, such as error checking, de-mapping, channel estimation, descrambling, etc. The raw data provided by baseband chip 402 may be sent to host chip 406 directly or stored in external memory 408.

[0045] In certain implementations, in downlink, baseband chip 402 in FIG. 4 may implement the MAP estimation approach using the synchronization signals, including the PSS and the SSS, to obtain the carrier FO estimation. This approach considers the FO as a variable to be estimated, rather than a constant, and a probability density function indicative of a priori knowledge associated with at least one of the channel, the receiver, or the transmitter is being maximized. Therefore, the FO estimation offered by the MAP approach may be more robust and is more suitable at a low SNR.

[0046] FIG. 5 illustrates a block diagram of an expanded view of a portion of the baseband chip of FIG. 4, according to some embodiments of the present disclosure. Baseband chip 402 may include a plurality of functional blocks configured to perform the FO estimation that is compatible with the initial cell search and selection for a wireless communication system, such as 4G LTE or 5G NR. In some embodiments, the functional blocks may include, e.g., a PSS processing component and a SSS processing component. These components may be implemented using electronic hardware, firmware, computer software, or any combination thereof. Whether such elements are implemented as hardware, firmware, or software depends upon the particular application and design constraints imposed on the overall system. For simplicity, FIG. 5 only shows and expends a portion of baseband chip 402 in regard to the FO estimation using the PSS and SSS. It can be understood that baseband chip 402 may include one or more other functional blocks configured for other baseband operations, such as PBCH processing, filtering, up- conversion, down-conversion, and/or sample-rate conversion.

[0047] In some embodiments, PSS processing component in baseband chip 402 may be implemented as an integrated circuit (IC) dedicated to performing its functions disclosed herein, such as an ASIC. Accordingly, in the present disclosure, PSS processing component and PSS processing circuit 502 may be used interchangeably.

[0048] In some embodiments, SSS processing component 504 in baseband chip 402 may be implemented as software modules executed by a baseband processor to perform the respective functions described below in regard to FIGs. 5, 7, and 8. At least a portion of SSS processing component 504 may be implemented by processor 302 and memory 304, as shown in FIG. 3. Processor 302 may be a hardware device having one or more processing cores, or the processor may execute software stored in memory on the hardware device to control the hardware device, depending upon the particular application and design constraints imposed on the overall system. For example, baseband chip 402 may include a baseband processor (BP) executing instructions stored in memory. The baseband processor may be a generic processor, such as a central processing unit or a DSP, not dedicated to the FO estimation. That is, the baseband processor may also be responsible for any other functions of baseband chip 402 and can be interrupted during the FO estimation being performed due to another process with higher priority. Some or all of the modules in SSS processing component 504 may be implemented as firmware modules executed by the baseband processor to perform the respective functions described below in regard to FIGs. 5, 7, and 8.

[0049] In some embodiments, a synchronization signal block (SSB) in 5GNR may be used by baseband chip 402 of, e.g., the UE 202 to perform the FO estimation. FIG. 6 illustrates a block diagram of the SSB in 5G NR. The SSB in 5G NR is functionally equivalent to LTE reference signals. Therefore, in other embodiments, the reference signals may also be applied to estimate the FO in the 4G LTE systems.

[0050] As illustrated in FIG. 6, SSB 600 in downlink may include one or more PSS symbols 602, one or more SSS symbols 604, and one or more PBCH symbols 606. In both the time and frequency domains, the SSS symbols may be surrounded by the PBCH symbols. In the time domain, four OFDM symbols are included in one SSB, where the PSS is transmitted in the first OFDM symbol, the SSS in the third OFDM symbol, and the PBCH in the second and fourth symbols. In the time domain, multiple SSBs are transmitted periodically (at 20ms intervals) in different beams each as a synchronization signal (SS) burst, and the PSS and SSS within each SSB are transmitted periodically to provide cell-specific timing information. Each of the PSS consists of three orthogonal sequences, either in the time domain or the frequency domain. The orthogonality of the PSS can help the SSB find the symbol timing and perform the coarse FO. When the SSB position is not known, a 5GNR synchronization raster indicates the frequency positions of the SSB that can be used by the UE. Although the contexts take the 5G NR SSB as an example to explain its contained PSS and SSS, the LTE communication systems, without introducing the SSB, may still apply the PSS and SSS in the reference signals to perform the FO estimation provided by the present disclosure.

[0051] FIG. 7 illustrates a flow diagram of an exemplary method of the FO estimation of baseband chip 402. FIG. 8 illustrates a flow diagram of another exemplary method of the FO estimation of baseband chip 402. With reference to FIGs. 5, 7, and 8, some embodiments of the present disclosure are described below.

[0052] Baseband chip 402 in FIG. 5 may include PSS processing circuit 502 and SSS processing component 504. At 702 in FIG. 7, PSS processing circuit 502 may process the PSS symbol to obtain a coarse FO estimation for an antenna by maximizing a probability density function of detecting the desired PSS symbol(s) correctly. The probability density function may be configured to optimize and correctly detect a transmitted PSS symbol and a transmitted SSS symbol at the transmitter. In other words, the probability density function may be maximized for detecting the desired symbol(s) correctly. The mismatch of carrier frequency between the transmitted symbol at the transmitter and the received symbol at the receiver results in the FO. Through maximizing the probability density function of detecting the target symbol(s) (e.g., the PSS symbol and/or the SSS symbol) correctly, the FO may be estimated.

[0053] Certain operation details in regard to FIG. 7 may be illustrated with reference to FIG. 8. The method may proceed to 802 in FIG. 8, where based on received samples 506, PSS processing circuit 502 may perform time-domain correlation to acquire symbol timing for each of one or more antennas 410. The samples 506 may be captured by and transmitted through RF chip 404 in FIG. 4 to baseband chip 402.

[0054] Each of the PSS symbols consists of three orthogonal sequences either in the time domain or in the frequency domain, and the orthogonality of the PSS can help to find the symbol timing, and subsequentially the coarse FO may be calculated based on the symbol timing. The symbol timing and the coarse FO estimation may be fed into SSS processing component 504 for further processing to acquire a final fine FO estimation. In the present disclosure, the PSS symbol may refer to the PSS sample sequences in the time domain, and thus the PSS symbol, the PSS, and the PSS sample sequences may be used interchangeably.

[0055] As shown in FIG. 5, PSS processing circuit 502 may include a plurality of functional modules, e.g., a PSS detection module 5022, a coarse FO estimation module 5024, and an antenna combining module 5026. For ease of the description, PSS processing circuit 502 in baseband chip 402 may be divided into the modules according to the functions of the modules. The present disclosure, however, does not place limitations thereto. In other embodiments, other division manners of the modules contained in PSS processing circuit 502, depending on the application and limitation of the system, may also be provided. In some embodiments, each module in PSS processing circuit 502 may be implemented as an integrated circuit (IC) dedicated to performing its functions disclosed herein, such as an ASIC.

[0056] At 802, PSS detection module 5022 may receive samples 506 over the time domain, and perform the time-domain correlation, under which, e.g., the symbol timing and a PSS identity (ID) may be detected. Samples 506 may include time-domain sample sequences associated with the PSS and SSS. In some embodiments, PSS detection module 5022 may be configured to obtain samples 506 from each of the one or more antennas 410 to determine symbol boundaries and accordingly extract the time-domain PSS sample sequences before the correlation stage, such as, through one or more filtering operations. The PSS sample sequences may be processed to locate a peak value associated with timing that may be determined as the PSS symbol timing and outputted to SSS processing component 504 to subsequentially determine the fine FO estimation. The SSS detection is performed and accomplished only after successful identification of the PSS sample sequences, which makes the PSS detection becomes a crucial stage.

[0057] At 802, PSS detection module 5022, though receiving sample 506 over the time domain, may perform the equivalent correlation in the frequency domain for reducing the computation complexity. This is the benefits of Fast Fourier Transform/ Inverse Fast Fourier Transform (FFT/IFFT) as their complexity is proportional to Nlog2(N) while N is the size of FFT/IFFT. By contrast, the time-domain correlation of N samples with a filter length N has complexity of N*N, much higher than that by doing correlation in the frequency domain through FFT/IFFT for larger N (The size ofPSS/SSS symbol is N =128).

[0058] In the matched-filter estimation approach, the detected PSS symbol timing at the PSS correlation is fed to the matched filters for the coarse FO estimation. As pointed out, when the SNR is relatively low on the wireless channels, however, the detected PSS symbol timing may be inaccurate in the time domain and/or may associate with a symbol timing offset that degrades the FO estimation result. Therefore, some embodiments of the present disclosure provide the maximum a posterior (MAP) estimation to replace the matched-filter estimation, and algorithms of the MAP estimation may be implemented to coarse FO estimation module 5024. Through treating the FO as a variable for the estimation and higher-complexity implementation, it can be expected that the MAP estimation can work better, even at a much lower SNR.

[0059] The method proceeds to 804 in FIG. 8. Through the MAP estimation, the carrier FO may be obtained according to the PSS symbol timing. The description and illustration of the present disclosure mainly cover some embodiments based on individual FO estimation per antenna where the FO for each antenna is first estimated, and the estimated FOs for the multiple antennas are combined to obtain an overall FO estimation. It can be understood that, however, other embodiments may also be employed. For example, a coarse FO for multiple antennas may be jointly estimated based on the received samples over the multiple antennas.

[0060] To begin with the MAP estimation, a channel with channel response h having memory of L samples may be considered, i.e., the channel response h is Z-dimensional. A data sequence x as transmitted may have a length of N, e.g., equivalent to the length of an OFDM symbol where the cyclic prefix (CP) is removed. That is, all x, y (the N-length received samples corresponding to x), and the additive white Gaussian noise (AWGN) are complex valued . Further, knowledge of the receiver may be denoted as /, and the FO at the receiver may be 0. Therefore, the knowledge about the FO 0 to the receiver may be modeled by a priori probability density function, i.e,/#).

[0061] Further, based on the knowledge of the receiver I and the received sequence y, finding and maximizing p(0 \ y, I) become the goal.

[0062] By Bayes’ rule,

[0063] From equation (1), in order to find and maximize the a posterior probability density distribution /?(19|y, I), an estimation method about p(y\0, 1) may need to be examined.

[0064] By the definition of conditional probability,

where h represents the channel response for the fading channel.

[0065] If x is known to the receiver, which reflects the current scenario because x is the detected sequence (either the detected PSS or SSS), then x in equation (2) is a single mass. Accordingly, equation (2) can be simplified as: p(y|0, /) = J h p(y|0, h, I)p h, I) dh (3), where p(h, I) can be expressed as p(h, I)= p(h\I)p(I) .

[0066] Because / is the knowledge about the receiver, consequently, the next goal may become finding the probability density function p(h\I) that can maximize p(y\0, 1) in equation (3). Based on Jaynes’ maximum entropy principle, the probability distribution that best represents the current state of knowledge is the one having the largest entropy. In the context of precisely stated prior data, to meet this goal, p(h\I) per se must be a distribution of h that can maximize the entropy of (h\ I).

[0067] As A is a multi-path channel response, if the receiver has the knowledge that h has L paths and is of average variance of unit, to maximize entropy of (h\I)

where H is the Hermitian transpose of a matrix.

[0068] Equation (4) indicates that the distribution of p(h\I) is Gaussian independent and identically distributed. Moreover, if the covariance matrix of h, Q = E/hh" ], is a priori known, a more general Gaussian distribution of p(h\I) may be expressed as:

where \Q\ is the determinant of matrix Q.

[0069] Equation (5) can be used to maximize the entropy. With a given p(h\I), based on the MAP estimation, the FO 0 can be found.

[0070] Due to the FO 0, the time-domain sample Xk will be rotated by 2nk 0/N where each OFDM symbol has N subcarriers. The initial phase of the samples may be ignored because all samples Xk share the same value of the initial phase. As obtaining the FO estimation is the goal, the common initial phase for all the samples can be ignored from the consideration.

[0071] Diagonal matrix De with vector de = (1, e^j27ie/N,

r represents a rotation transformation impacted on the samples due to the FO 0, where * is a multiplication operator. Matrix X represents the sample and has N rows with the first row of sample vector (xo, XN-I, ... , XN-L-I) , and the second row being a circular rotation shifted right from the first row by one sample, (xi, xo, ..., XN-L-2) , associated with channel impulse response of h with one sample delay, until to the X^th row of X, which is (XN-I, XN-2, ... , XN-L). This rotation operation is based on implicit assumption with channel response of L samples less than CP length. Based on De and X, a relationship between the input and the output with the FO 0 at the receiver can be modeled in a matrix format as: y = D_eXh + w (6), where w represents the complex-valued additive white Gaussian noise.

[0072] With the assumptions that /. and Q are a priori known, through equations (3), (5), and (6), equation (7) can be obtained as:

where a² is an average power of Gaussian noise per sample, and M = 1/ a² X^!/X Q~ , and where

[0073] A posteriori optimal estimate 0_map can be obtained by maximizing:

[0074] In practice, it may be challenging to know exactly the distribution p(0\I) in equation

(9). However, in some embodiments regarding the NR applications for cell search and measurement, the boundary of 0 may be known. Therefore, it may be reasonable to assume that the distribution p(0\I) is uniform within a fixed lower bound LB and a fixed upper bound UB . With the assumptions, equation (9) can be simplified as:

0_map = min C(0) for LB < 0 < 0UB (10).

[0075] Considering that C(0) is real-valued and smooth for 0 within the pre-determined ranges, therefore, _map may be achieved by equation (10) when UB - 0LB < 1 subcarrier spacing applies based on the assumptions.

[0076] Although it is assumed that p(0\I) is uniform as discussed above, in other embodiments where a potential FO due to the temperature change of the RF chip occurs, p(0\I) may not present as a uniform distribution and equation (10) may apply for the scenarios. For this reason, obtaining the a priori knowledge about 0 may be helpful to generate a better MAP estimation of the FO in applying equation (9).

[0077] Returning to FIGs. 5 and 7, in some embodiments, coarse FO estimation module 5024 may implement the algorithms of the MAP estimation as described above to realize the coarse FO estimation process at 702.

[0078] It is known that the PS S is defined by the orthogonality in the frequency domain and over a relatively narrow bandwidth (also referred to as “narrowband frequency”). The PSS reference also (almost) achieves the orthogonality when being shifted in the time domain with the integer sampling period defined by the Nyquist rate. Consequently, the PSS correlation can be performed in either the frequency domain or the time domain depending on the target implementation latency, complexity, power consumption, shared hardware, etc. of the system.

Equation (8) above is reproduced herein as:

[0079] Due to the orthogonality of the PS S reference in the time domain, the square matrix consisting of the PSS (i.e., X) and its cyclic shift, includes relatively large components in the diagonal of the square matrix and almost zero otherwise. This property can ensure that square matrix X^!IX has an inverse matrix (X¹X)~¹. In other words, using the PSS, component X(X^HX)~¹X^H in equation (8) exists, without any need to estimate the power of noise o² and/or the information of matrix Q. In certain embodiments, it may be assumed that the distribution p(0\I) is uniform within a fixed lower bound 0LB and a fixed upper bound UB to simplify the estimation processing. As a result, p(y\0, I) can be estimated to maximize the posterior distribution

I). Accordingly, at 804 in FIG. 8, the coarse FO estimation for each antenna 410 can be estimated at coarse FO estimation module 5024.

[0080] In some embodiments, the receiver may include a plurality of antennas configured to receive data transmitted from the transmitter over the channel. At 704 in FIG. 7, PSS processing circuit 502 may combine coarse FO estimations for the plurality of antennas to obtain a final coarse FO estimation. For this purpose, PSS processing circuit 502 in FIG. 5 may further include antenna combining module 5026. As, at coarse FO estimation module 5024, the coarse FO may be estimated for each antenna 410 using the MAP estimation, antenna combining module 5026 may further combine the estimated coarse FOs over the plurality of antennas to obtain the final coarse FO estimation associated with the receiver.

[0081] The final coarse FO estimation is an overall coarse FO estimation over the plurality of antennas. In the present disclosure, the “final” FO estimation and the “overall” FO estimation may be used interchangeably. Meanwhile, the final “coarse” FO estimation may indicate that the obtained FO is rough in texture and may have more estimation errors, as compared to a final “fine” FO estimation. The FO estimation(s) “for an antenna,” “for a plurality of antennas,” “over an antenna,” or “over a plurality of antennas” may denote the raw samples inputted for processing to obtain the FO estimation(s) are captured through “one antenna” or “a plurality of antennas.” [0082] The method proceeds to 806 in FIG. 8 to determine whether the estimated coarse FOs for all the antennas are obtained. If all the estimated coarse FOs are acquired, the method may proceed to 808 for the combination. Otherwise, the method may repeat estimating the coarse FO for another antenna at 804. At 808, antenna combining module 5026 may combine the estimated coarse FOs for all the antennas to acquire the final coarse FO associated with the receiver.

[0083] In the scenarios where the receiver includes multiple antennas, based on equations (8)-(l 0) listed above, various combination methods can be applied. For example, considering the weight wn, for antenna index z, M for the total number of the antennas (i.e., i=l, M), and is the received sample data for antenna index z, the FO can be estimated based on a joint detection over the multiple antennas. The expression of “antenna index z” is employed to refer to the antenna with index z. In a similar manner, a symbol/parameter having index z may denote that the symbol/parameter is associated with the antenna with index z. Based on equation (8), a combined cost function based on respective C(0, y^l) can be expressed as:

where * represents a multiplication operator.

[0084] In equation (8) defining C(0, y^l), is the same reference for all the antennas, but noise variance Q in C(0, y^l) may be different for respective antennas. Equation (11) calculates the combined cost function based on the weighted sum of C(0, y^l) for antenna index i= 1, M. Accordingly, the answer to equation (11) (i.e., the final FO 0_map) is the solution that minimizes Ji(0).

[0085] In other embodiments, the FO estimation may be achieved at each antenna, and the final FO estimation may be calculated based on the antenna combination of the individual FO estimations. That is, 0'_map may be first obtained for respective y', where y' represents the sample data corresponding to antenna index z and z is a positive number. Accordingly, a cost function combining respective optimal 0'_map can be expressed as:

where * represents a multiplication operator.

[0086] For simplicity, the description and illustration of the present disclosure mainly encompass embodiments based on equation (12) to obtain an overall FO estimation where the FO per antenna is first estimated and the estimated FOs for the multiple antennas are combined to obtain the overall FO estimation.

[0087] In some instances, the weight M’Z/_; in equations (11) and (12) can be selected as an equal gain for all the antennas. In other instances, the weight M’Z/_; can be an optimized value which may depend on respective signal quality or reliability indicators, such as a correlation score. The present disclosure does not limit thereto. Therefore, different manners for determining the weight M’z/_; can be considered, such as equal gain combining (EGC) scheme and maximum ratio combining (MRC) scheme. The former EGC scheme applies an equal weight w to the estimated FOs per antenna before calculating the final estimation. This scheme can be more applicable when no (sufficient) a priori information about the signal quality of each antenna. By contrast, the MRC scheme may assign different weights to the estimated FOs from each antenna according to some knowledge about, e.g., the quality and/or reliability of each antenna. Although the MRC scheme can show better performance to the EGC scheme, the EGC scheme may be superior to the MRC scheme in its simple implementation. The present disclosure does not limit thereto.

[0088] In the combination of the respective FO estimation using the MRC scheme, the weight wni per antenna in equation (12) may need to be determined in advance before the antenna combination. As described above, 0'map for each antenna index z can be obtained through the MAP estimation. In some instances, considering that the respective cost function for antenna index z, C(ff_map, ), is positive and the goal is to minimize the combined cost function O) in equation (12), therefore, a larger weight may be assigned to a smaller individual cost function C(ff_map, ) so as to arrive at the goal to minimize the combined cost. This concept motivates the choices of the weight wn, for finding the final FO 0_map from equation (12) as:

[0089] Equation (13) indicates that the weight wni may be determined according to the concept that the weight wn, may be negatively proportional to the respective cost function C(ff_map, ) for each antenna. Equation (13) offers an exemplary application of the MRC scheme, but the present disclosure does not place limitations thereto. For example, in other instances, the determination of the weight wn, may incorporate the PSS or SSS correlation results that can provide valuable information about the channel. Generally speaking, if the receiver has certain a priori knowledge about the antennas, the antenna combination can obtain the benefit. For example, in some scenarios where a specific antenna is impacted by RF jammer or interference severely, or a specific antenna experiences a relatively long delay profile with a large variation or low receive signal power (resulting in low SNR or low SINR), the signals from these specific antennas may be less reliable, and thus the FO estimation from these antennas are less reliable. In the MRC scheme, the specific antennas may be ignored by specifying zero(es) to the weights wn, corresponding to the antennas or assigned with a smaller weight in the antenna combination so as to produce a good estimation result.

[0090] Returning back to FIG. 5, in some embodiments, baseband chip 402 may include SSS processing component 504. SSS processing component 504 may be configured to receive the symbol timing and the final coarse FO estimation from PSS processing circuit 502. Based on at least these inputs, SSS processing component 504 may be configured to perform the cell detection for getting cell ID, and MAP estimation using SSS symbol and detected cell ID (thereby providing SSS reference signal, based on 3 GPP specification), perform channel estimation and UE measurements (such as reference signal received power, RSRP, and, reference signal received quality, RSRQ), and find a final fine FO estimation.

[0091] SSS processing component 504 may include a plurality of functional modules, e.g., an SSS compensation module 5042, an SSS detection module 5044, a channel estimation module 5046, a noise variance estimation module 5048, a fine FO estimation module, and an antenna combining module 5052. For easy description, SSS processing component 504 may be divided into the modules according to the functions of the modules. In other embodiments, other division manners of the modules, depending on the application and limitation of the system, may also be offered.

[0092] With reference to FIG. 7, the method proceeds to 706, where the SSS symbol may be compensated at least based on the final coarse FO estimation to obtain compensated SSS symbol. SSS compensation module 5042 may be configured to compensate the SSS symbol over all the antennas. FIG. 8 illustrates details of the processes. With reference to FIG. 8, the method proceeds to 810, 812, and 814. At 810, SSS compensation module 5042 may be configured to receive the samples transmitted through RF chip 404 and the symbol timing transmitted through PSS processing circuit 502 so that raw SSS symbol over all the antennas can be found. At 812, SSS compensation module 5042 may be further configured to take the final coarse FO estimation from PSS processing circuit 502 and perform frequency compensation on the raw SSS symbol to obtain the compensated SSS symbol over all the antennas.

[0093] The processes can be expressed as:

SSScffc] = S[fc] * conjugate (do [k]) (14), where S[k] represents input samples, and in this case, will be Msample time-domain SSS sequence S[0], S[l], S[N-1J and its cycle-shift version, diagonal matrix De with vector de = (1, e^j2"" e j2ne/N* 2

represents rotation transformation impacted on the samples due to the final coarse FO 0, and conjugate represents a conjugate matrix.

[0094] The SSS sequence is cell-associated, so it can be determined at the cell detection stage. Based on equation (14), the compensated SSS, i.e., SSSc[k], can be determined according to the final coarse FO 0 and the raw S[k],

[0095] In some embodiments, the final coarse FO estimation may be transformed from the time domain to the frequency domain using Fast Fourier Transform (FFT) such that the raw SSS symbol can be compensated over the frequency domain.

[0096] SSS compensation module 5042 may include or be connected with, e.g., a local memory, registers, a buffer, a cache, a DRAM, or an SRAM. At 814, the compensated SSS symbol may be buffered in, e.g., the local memory, and the compensated SSS symbol per antenna may be sequentially fed into channel estimation module 5046, noise variance estimation module 5048, and fine FO estimation module 5050 later for respective processing.

[0097] In FIG. 7, at 708, SSS-related information may be obtained based on the compensated SSS symbol. The SSS-related information may refer to at least the cell ID and SSS reference. With reference to FIG. 8, the method process to 816, 818, 820, and 822. In SSS processing component 504, SSS detection module 5044 may be configured to process SSS detection over all the antennas to determine the cell ID at 816 and generate SSS reference based on the cell ID at 818. For example, after the PSS synchronization is completed, the exact position of the PSS can be determined and sent to SSS processing component 504 as the symbol timing. At PSS processing circuit 502, once the PSS is decoded, one out of three identities (i.e., NID⁽²⁾, 0 to 2, an identity of the group) can be determined. After the compensated SSS is obtained and decoded at 816, a physical cell identity (PCI) group number (i.e., NID⁽¹⁾, 0 to 335 for NR) can also be determined. By means of the obtained NID⁽¹⁾ and NID⁽²⁾, the cell ID can be calculated as cell ID = 3*NID⁽¹⁾ + NID⁽²⁾. Based on the cell ID, at 818, the SSS reference in the time domain can be generated with the help of inverse Fast Fourier Transform (IFFT). Meanwhile, the UE measurements, such RSRP and/or RSRQ, can be determined as well.

[0098] By means of the detected cell ID at the SSS detection, the fine FO estimation may be obtained to improve the accuracy of the FO estimation because the SSS is more immune to the interference from other nearby users. In other words, the application of the MAP estimation scheme with the SSS reference associated with the channel response can be expected to provide more accurate results and thus enhance the outcome of PBCH processing.

[0099] In some embodiments, the algorithms of the MAP estimation can be implemented to SSS processing component 504. The equations derived in regard to the MAP estimation (e.g., equation (8)) may be employed. For example, the average power of Gaussian noise per antenna o² in equation (8) may be calculated through a comparison of the SSS reference and the compensated SSS symbol. At least based on the noise power per antenna o² and the compensated SSS symbol, by using the individual FO estimation per antenna, a fine FO estimation for each antenna can be calculated by equation (8). For the purposes, in some embodiments, SSS processing component 504 may include channel estimation module 5046 configured to perform channel estimation in addition to noise variance estimation module 5048 configured to estimate the noise power.

[0100] Wireless communication systems are adversely affected by multi-path interference. In order to provide reliable data to the receiver, a system requires an accurate estimate of the timevarying channel. In some embodiments, channel estimation module 5046 may take the cell-specific SSS reference and a corresponding compensated SSS symbol as fetched to determine the channel estimation for each antenna. As described above, the compensated SSS symbol per antenna may be calculated and stored or buffered at, e.g., a local or internal memory. At this stage, the compensated SSS symbol per antenna may be sequentially fetched out and provided to channel estimation module 5046. Based on the cell-specific SSS reference, a channel response for each antenna may be measured so as to determine channel information for estimating the behavior of the time-varying channel. The channel information may provide the a priori knowledge to better estimate the fine FO. Further at 822 in FIG. 8, based on the SSS reference, the channel estimation per antenna, and the compensated SSS symbol per antenna, the noise power per antenna can be estimated through noise variance estimation module 5048.

[0101] Referring back to FIG. 7, at 710, a fine FO estimation may be obtained for each antenna also by maximizing the probability density function through the MAP estimation based on the SSS symbol, according to some embodiments of the present disclosure.

[0102] As described above, the compensated SSS symbol per antenna may be calculated and stored or buffered at, e.g., a local or internal memory. At this stage, the compensated SSS symbol per antenna may be sequentially fetched out and provided to fine FO estimation module 5050. That is, through channel estimation module 5046, the channel information can be obtained, and through noise variance estimation module 5048, the noise power per antenna can be estimated. At 824 in FIG. 8, by inputting at least the noise power per antenna, the compensated SSS symbol per antenna, the channel estimation per antenna, the raw SSS symbol per antenna, and the cellspecific SSS reference into equation (8), fine FO estimation module 5050 can estimate the fine FO for each antenna. In some embodiments, the channel estimation per antenna may include a channel response in the time domain for each antenna.

[0103] The method proceeds to 712 at FIG. 7, where fine FO estimations for the plurality of antennas may be combined to obtain the final fine FO estimation. SSS processing component 504 in FIG. 5 may include antenna combining module 5052. As at fine FO estimation module 5050, the individual FO may be estimated for each antenna, antenna combining module 5052 may be configured to combine the estimated fine FOs over all the plurality to obtain the final fine FO. The final fine FO described herein and an overall FO estimation over the plurality of antennas may be used interchangeably. In some embodiments, at 826 in FIG. 8, antenna combining module 5052 may determine whether the estimated fine FOs for all the antennas are obtained. If all the estimated fine FOs are acquired, the method may proceed to 828 for the antenna combination. Otherwise, the method may repeat performing the procedures, including the channel estimation, the estimation of the noise variance, and the estimation of the fine FO for another antenna till all the fine FOs are estimated. At 828, antenna combining module 5052 may be configured to combine the estimated fine FOs for all the antennas to obtain the final fine FO.

[0104] Similar to the manners as described above in regard to PSS processing circuit 502, antenna combining module 5052 in SSS processing component 504 may employ a weight based on either the EGC scheme or MRC scheme toward the antenna combination. In other embodiments of the present disclosure, however, SSS processing component 504 may estimate an overall FO estimation over multiple antennas using the MAP estimation but by a joint detection technique, rather than estimating individual FO for each antenna for the combination.

[0105] In certain scenarios, a range of the estimated FO may be narrowed down through the MAP estimation so as to reduce calculation cycle time. For example, at an initial cell search, if the frequency at resolution of ’A subcarrier spacing (SCS) is applied with the 3 PSS references, the residual FO using the MAP estimation will be within a range of [-’A, %] SCS. Generally speaking, if the PSS references at a resolution of p SCS are scanned where p is a fractional number between 0 and 1, the residual FO will be within a range of [-p/2, p/2] SCS. In some embodiments, therefore, if the frequency at relatively larger resolution can be scanned and the MAP estimation can be applied to narrow down the FO, the overall implementation cost will be reduced substantially. This strategy may be helpful, in particular, when no timing information is given and search for symbol boundary with large FO is quite expensive.

[0106] In some scenarios that the cell link between the UE and the base station is already existent, the FO can become smaller in comparison with the initial cell search mode. These scenarios include, but not limited to, the DRX mode for re-synchronization, cell measurement to seek potential cells to switch, and search within multiple cells for a given target cell. Running the MAP scheme will provide more than a better result in these scenarios because the MAP estimation can yield the FO estimation with smaller variance and is free of bias.

[0107] With the decoded PBCH symbols, corresponding equalized time-domain reference can be iteratively used to estimate the FO with the MAP estimation approach. As there are multiple PBCH symbols, the FO estimation can be much accurate and can produce better performance for the data channel. With the compensated PBCH samples by the latest FO estimation, the PBCH detection and decoding can be more robust against the low SNR, strong interference, and severe fading. As a result, with the much smaller residual FO, the timing offset can be compensated to provide communication with higher quality for PDSCH processing.

[0108] As described above, the MAP estimation of frequency offset is optimal for a typical fading channel compared to the matched-filter estimation. Although the matched-filter estimation may underperform the MAP estimation, in certain scenarios (e.g., at high SNR and/or with short channel response) the matched-filter estimation can still be acceptable. Moreover, considering the implementation cost, in some embodiments, the MAP estimation may be optionally implemented based on knowledge level about the channel, the receiver, and the transmitter. When the UE considers implementing either the MAP estimation and/or the matched-filter based estimation, the UE may take account of, e.g., received signal strength indicator (RSSI) and PSS correlation score (e.g., for frequency scan mode and initial cell search mode), or reference signal received power (RSRP)/reference signal received quality (RSRQ) and PSS/SSS correlation scores (e.g., for DRX mode and cell measurement mode), the like, or a combination thereof, but not limited thereto. In some embodiments, the matched-filter approach may be implemented to replace the MAP approach when certain conditions are satisfied. For example, in response to the SNR of the channel being greater than or equal to a threshold, the matched-filter approach may be applied to estimate the FO in consideration of the cost.

[0109] In the implementation of the MAP estimation, the length of channel response L samples is assumed known in the derivation of 0_map. The selection of L may be implemented in different manners, including but not limited to a predefined value through the simulations or multipath channel profile analysis. The length of channel response L may be selected values in a range LI to Z2, where L2 is equal to or less than the size of cyclic prefix (CP). Consequently, with the known range, equation (3) may turn to be in a discrete form as:

[0110] In applying equation (15), for the multi-path channel, the root mean square of power delay spread can be applied to determine the length of channel response L for the MAP estimation. As an example, a typical channel for 3GPP Specification (Extended Vehicular A model, or EVA channel, at 15KSCS) has the maximal power delay spread at roughly 4.875 samples at the Nyquist sampling rate. In comparison, the CP length is 9 samples for the PSS/SSS reference. The root mean square value of the power delay spread may indicate that L = 3 can be chosen for the design. That is, most of the power of channel response may be concentrated within roughly the first several samples. As a result, a selection of unnecessarily larger L may introduce more computation demand for a matrix and its inverse but achieve a slight performance gain. Generally speaking, even with a selection of channel response L less than the actual length, the MAP estimation provided by the present disclosure can still offer a satisfactory result close to an optimal solution by acceptable performance loss.

[oni] In regard to the term Q'¹ in equation (8), the simulation results indicate that even matrix Q is not accurately known, the MAP scheme may still perform well. The results motivate the simplified implementation of choosing Q'¹. In one instance, a choice of matrix Q = I^L/L (assuming that h is Gaussian independent and identically distributed, and Q is L by Z, i.e., a dimensional identity matrix normalized by Z) can be considered. Since the inverse matrix in equation (8) is a dimension of Z by Z, when Z may be not substantially large, the implementation of Hermitian matrix inverse in equation (8) can be realized by various schemes efficiently, such as lower-diagonal -lower (LDL) decomposition and Cholesky approach, but not limited thereto. In another instance, the (^-associated term in equation (8) may be totally omitted, resulting in the only little sacrifice of performance. In the meantime, X^Xin equation (8) can still have an inverse matrix under this simplified implementation, and the MAP algorithms can still work well because X consists of the orthogonal PSS references, and X^HX is the dominant term inside X'X r>²0~'', thereby making the removal of the (^-associated term reasonable.

[0112] In regard to the added white Gaussian noise power cr², its value can be estimated, even for the low SNR. Mismatch between the estimated and the true value of o² can provide a sub- optimal MAP FO estimated.

[0113] FIG. 9 illustrates simulation results, using matched-filter and MAP approaches over an extended typical urban (ETU) channel model, in view of the mean square errors (MSE) of the residual FOs, according to some embodiments of the present disclosure. Curves 902 and 902 show the MSE of the residual FOs over different numbers of antennas using the matched-filter approach, while curves 906, 908, and 910 show the MSE of the residual FOs over different numbers of antennas using the MRC scheme in the MAP approach. According to FIG. 9, the MAP estimation, in regard to curves 906, 908, and 910, demonstrates the performance enhancement when the SNR increases, regardless of the number of antennas. Generally speaking, the MAP approach can offer better performance for the ETU channel with large FO. By contrast, curves 902 and 904 using the matched-filter approach substantially remain flat, implying that the estimation error is not improved at the low SNR range with increased signal power. Moreover, in FIG. 9, the curves using the MAP approach demonstrate the truncated Gaussian distribution with decreasing variance as the SNR increases, while the matched-filter approach fails to show a similar feature, which implies the MAP approach can work well at the fading channels.

[0114] The MAP approach considers the FO as a variable to be estimated, rather than a constant, for which it can provide better accuracy, even for the fading channels at the low SNR. In application, the MAP approach can reduce the frequency resolution for the raster scanning with the PSS by (at least) two times. It can also provide more accurate fine FO estimation with the SSS reference, thus bringing a better chance for successful PBCH decoding. Generally, it can save the frequency scanning associated time and power and enable the UE to build the cell link at the initial cell search sooner and successfully, especially at a low SNR.

[0115] At the DRX mode, as the length of channel response L is known from the channel estimation, the information about the length of channel response L can be used to perform the MAP estimation, which leads to better accuracy of the FO with either the PSS or SSS when the UE wakes up for the re-synchronization. In other words, the UE may sleep longer time for the power saving without losing the synchronization in the frequency domain.

[0116] When the cell measurements are involved, the signal quality is generally a variable, and the SNR/SINR can be within a broad range. Besides the matched-filter estimation approach, the MAP estimation approach can provide another tool for the UE to choose. The FO estimation using the MAP approach is more accurate, so the signal quality indicator RSRP and/or RSRQ will be accordingly more precise and reliable, and eventually, it is also beneficial for the cell reselection.

[0117] According to one aspect of the present disclosure, a baseband chip implementing frequency offset (FO) estimation at a receiver is provided. The baseband chip may include a primary synchronization signal (PSS) processing circuit configured to obtain a PSS symbol through an antenna from a transmitter over a channel. The PSS symbol may be associated with a time domain. The PSS symbol may be inputted to a probability function. The probability function may include a variable of FO and matrices indicating knowledge associated with at least one of the channel, the transmitter, or the receiver. The probability function may be maximized according to the variable of FO and the PSS symbol to obtain a coarse FO estimation for the antenna. The probability function may be configured to optimize a detection rate of a transmitted PSS symbol at the transmitter. [0118] In some embodiments, the matrices may include at least one of a channel response, a length of channel response, a potential FO, bounds of the potential FO, a predicted or measured signal -to-noise ratio (SNR), a correlation score, or a potential cell comprising a cell identification (ID).

[0119] In some embodiments, the receiver may include a plurality of antennas. The PSS processing circuit may be further configured to obtain coarse FO estimations for the plurality of antennas of the receiver and combine coarse FO estimations for the plurality of antennas to obtain a final coarse FO estimation.

[0120] In some embodiments, the PSS processing circuit may be further configured to add weights to the coarse FO estimations to obtain weighted coarse FO estimations for the plurality of antennas. The weighted FO estimations for the plurality of antennas may be combined to obtain the final coarse FO estimation.

[0121] In some embodiments, the weights may be determined according to information in part about the plurality of antennas.

[0122] In some embodiments, in response to a signal-to-noise ratio of the channel being greater than or equal to a threshold, the PSS processing circuit may be configured to divide PSS symbols from the plurality of antennas into a first portion and a second portion. The final coarse FO estimation may be obtained based on the first portion of the PSS symbol and the second portion of the PSS symbol.

[0123] In some embodiments, the PSS processing circuit may be further configured to process PSS symbols for the plurality of antennas to obtain symbol timing of the PSS symbols. The symbol timing may include a PSS symbol boundary corresponding to each of the plurality of antennas.

[0124] In some embodiments, the baseband chip may further include a processor operatively coupled to the plurality of antennas and memory storing instructions. Execution of the instructions may cause the processor to process an SSS symbol for a respective antenna based on a corresponding symbol timing and the final coarse FO estimation to obtain respective compensated SSS symbol. SSS symbols and compensated SSS symbols, for the plurality of antennas, may be processed to obtain SSS-related information, channel estimation for each antenna, and estimation of power of noise for each antenna, the SSS-related information comprising a cell ID and an SSS reference. The SSS symbol corresponding to the respective antenna, the compensated SSS symbol corresponding to the respective antenna, the channel estimation corresponding to the respective antenna, the estimation of power of noise corresponding to the respective antenna, and the SSS-related information may be inputted to the probability function to maximize the probability function according to the variable of FO for obtaining a fine FO estimation for each of the plurality of antennas. The probability function may be configured to optimize a detection rate of a transmitted SSS symbol at the transmitter.

[0125] In some embodiments, execution of the instructions may further cause the processor to combine fine FO estimations for the plurality of antennas to obtain a final fine FO estimation.

[0126] In some embodiments, execution of the instructions may further cause the processor to add weights to the fine FO estimations to obtain weighted fine FO estimations for the plurality of antennas. The weighted FO estimations may be combined to obtain the final fine FO estimation. [0127] In some embodiments, execution of the instructions may further cause the processor to buffer the compensated SSS symbols corresponding to the plurality of antennas in a storage device. Each of the compensated SSS symbols corresponding to one of the plurality of antennas may be retrieved for processing to obtain the fine FO estimation for each of the plurality of antennas.

[0128] In some embodiments, the PSS processing circuit may be further configured to process the PSS symbol to obtain an identity within a group of a cell. Execution of the instructions may further cause the processor to process the compensated SSS symbols to obtain a physical cell identity (PCI) group number and obtain the cell ID based on the PCI group number and the identity within the group.

[0129] In some embodiments, execution of the instructions further causes the processor to compensate SSS symbols for the plurality of antennas based on the final coarse FO estimation over a frequency domain to obtain the compensated SSS symbols. The compensated SSS symbols and the cell ID may be processed to obtain the SSS reference.

[0130] According to another aspect of the present disclosure, a baseband chip implementing frequency offset (FO) estimation at a receiver is provided. The baseband chip may include a primary synchronization signal (PSS) processing circuit configured to process PSS symbols based on a probability function. The PSS symbols may be obtained through a plurality of antennas from a transmitter over a channel and associated with a time domain. The probability function may include a variable of FO and matrices indicating knowledge associated with at least one of the channel, the transmitter, or the receiver. The probability function may be maximized according to the variable of FO to obtain a final coarse FO estimation for the plurality of antennas. The baseband chip may further include a processor operatively coupled to the plurality of antennas and memory storing instructions. Execution of the instructions may cause the processor to process SSS symbols and the final coarse FO estimation to obtain a final fine FO estimation based on the probability function. The probability function may be configured to optimize a detection rate of a transmitted PSS symbol and a transmitted SSS symbol at the transmitter

[0131] In some embodiments, the matrices may include at least one of a channel response, a length of channel response, a potential FO, bounds of the potential FO, a predicted or measured signal -to-noise ratio (SNR), a correlation score, or a potential cell comprising a cell identification (ID).

[0132] According to still another aspect of the present disclosure, a method for frequency offset (FO) estimation implemented at a receiver is provided. The method may include obtaining a PSS symbol through an antenna from a transmitter over a channel. The PSS symbol may be associated with a time domain and inputted to a probability function. The probability function may include a variable of FO and matrices indicating knowledge associated with at least one of the channel, the transmitter, or the receiver. The probability function may be maximized according to the variable of FO to obtain a coarse FO estimation for the antenna. The probability function may be configured to optimize a detection rate of a transmitted PSS symbol at the transmitter.

[0133] In some embodiments, the matrices may include at least one of a channel response, a length of channel response, a potential FO, bounds of the potential FO, a predicted or measured signal -to-noise ratio (SNR), a correlation score, or a potential cell comprising a cell identification (ID).

[0134] In some embodiments, the receiver may include a plurality of antennas. The method further may include obtaining coarse FO estimations for the plurality of antennas of the receiver. Coarse FO estimations for the plurality of antennas may be combined to obtain a final coarse FO estimation.

[0135] In some embodiments, an SSS symbol for a respective antenna may be processed based on a corresponding symbol timing and the final coarse FO estimation to obtain respective compensated SSS symbol. SSS symbols and compensated SSS symbols, for the plurality of antennas, may be processed to obtain SSS-related information, channel estimation for each antenna, and estimation of power of noise for each of antenna, the SSS-related information comprising a cell ID and an SSS reference. The SSS symbol corresponding to the respective antenna, the compensated SSS symbol corresponding to the respective antenna, the channel estimation corresponding to the respective antenna, and the estimation of power of noise corresponding to the respective antenna, and the SSS-related information may be inputted to the probability function to maximize the probability function according to the variable of FO for obtaining a fine FO estimation for each of the plurality of antennas. The probability function may be configured to optimize a detection rate of a transmitted SSS at the transmitter.

[0136] In some embodiments, fine FO estimations for the plurality of antennas may be combined to obtain a final fine FO estimation.

[0137] The foregoing description of the specific embodiments will so reveal the general nature of the present disclosure that others can, by applying knowledge within the skill of the art, readily modify and/or adapt for various applications such specific embodiments, without undue experimentation, without departing from the general concept of the present disclosure. Therefore, such adaptations and modifications are intended to be within the meaning and range of equivalents of the disclosed embodiments, based on the teaching and guidance presented herein. It is to be understood that the phraseology or terminology herein is for the purpose of description and not of limitation, such that the terminology or phraseology of the present specification is to be interpreted by the skilled artisan in light of the teachings and guidance.

[0138] Embodiments of the present disclosure have been described above with the aid of functional building blocks illustrating the implementation of specified functions and relationships thereof. The boundaries of these functional building blocks have been arbitrarily defined herein for the convenience of the description. Alternate boundaries can be defined so long as the specified functions and relationships thereof are appropriately performed.

[0139] The Summary and Abstract sections may set forth one or more but not all exemplary embodiments of the present disclosure as contemplated by the inventor(s), and thus, are not intended to limit the present disclosure and the appended claims in any way.

[0140] Various functional blocks, modules, and steps are disclosed above. The particular arrangements provided are illustrative and without limitation. Accordingly, the functional blocks, modules, and steps may be re-ordered or combined in different ways than in the examples provided above. Likewise, certain embodiments include only a subset of the functional blocks, modules, and steps, and any such subset is permitted.

[0141] The breadth and scope of the present disclosure should not be limited by any of the above-described exemplary embodiments but should be defined only in accordance with the following claims and their equivalents.

Claims

WHAT IS CLAIMED IS:

1. A baseband chip implementing frequency offset (FO) estimation at a receiver, comprising: a primary synchronization signal (PSS) processing circuit configured to: obtain a PSS symbol through an antenna from a transmitter over a channel, the PSS symbol being associated with a time domain; input the PSS symbol to a probability function, the probability function comprising a variable of FO and matrices indicating knowledge associated with at least one of the channel, the transmitter, or the receiver; and maximize the probability function according to the variable of FO to obtain a coarse FO estimation for the antenna, the probability function being configured to optimize a detection rate of a transmitted PSS symbol at the transmitter.

2. The baseband chip of claim 1, wherein the matrices comprise at least one of a channel response, a length of channel response, a potential FO, bounds of the potential FO, a predicted or measured signal-to-noise ratio (SNR), a correlation score, or a potential cell comprising a cell identification (ID).

3. The baseband chip of claim 1, wherein: the receiver comprises a plurality of antennas; and the PSS processing circuit is further configured to: obtain coarse FO estimations for the plurality of antennas of the receiver; and combine the coarse FO estimations for the plurality of antennas to obtain a final coarse FO estimation.

4. The baseband chip of claim 3, wherein the PSS processing circuit is further configured to: add weights to the coarse FO estimations, to obtain weighted coarse FO estimations for the plurality of antennas; and combine the weighted FO estimations for the plurality of antennas to obtain the final coarse

FO estimation.

5. The baseband chip of claim 4, wherein the weights are determined according to information in part about the plurality of antennas.

6. The baseband chip of claim 3, wherein in response to a signal-to-noise ratio of the channel being greater than or equal to a threshold, the PSS processing circuit is configured to: divide PSS symbols from the plurality of antennas into a first portion and a second portion; and obtain the final coarse FO estimation based on the first portion of the PSS symbol and the second portion of the PSS symbol.

7. The baseband chip of claim 3, wherein the PSS processing circuit is further configured to: process PSS symbols for the plurality of antennas to obtain symbol timing of the PSS symbols, the symbol timing comprising a PSS symbol boundary corresponding to each of the plurality of antennas.

8. The baseband chip of claim 7, further comprising: a processor operatively coupled to the plurality of antennas; and memory storing instructions that, when executed by the processor, cause the processor to: process a secondary synchronization symbol (SSS) symbol for a respective antenna based on a corresponding symbol timing and the final coarse FO estimation to obtain respective compensated SSS symbol; process SSS symbols and compensated SSS symbols, for the plurality of antennas, to obtain SSS-related information, channel estimation for each antenna, and estimation of power of noise for each antenna, the SSS-related information comprising a cell ID and an SSS reference; and input the SSS symbol corresponding to the respective antenna, the compensated SSS symbol corresponding to the respective antenna, the channel estimation corresponding to the respective antenna, the estimation of power of noise corresponding to the respective antenna, and the SSS-related information to the probability function to maximize the probability function according to the variable of FO for obtaining a fine FO estimation for each of the plurality of antennas, the probability function being configured to optimize a detection rate of a transmitted SSS symbol at the transmitter.

9. The baseband chip of claim 8, wherein execution of the instructions further causes the processor to combine fine FO estimations for the plurality of antennas to obtain a final fine FO estimation.

10. The baseband chip of claim 9, wherein execution of the instructions further causes the processor to: add weights to the fine FO estimations to obtain weighted fine FO estimations for the plurality of antennas; and combine the weighted FO estimations to obtain the final fine FO estimation.

11. The baseband chip of claim 8, wherein execution of the instructions further causes the processor to: buffer the compensated SSS symbols corresponding to the plurality of antennas in a storage device; and retrieve each of the compensated SSS symbols corresponding to one of the plurality of antennas for processing to obtain the fine FO estimation for each of the plurality of antennas.

12. The baseband chip of claim 8, wherein: the PSS processing circuit is further configured to process the PSS symbol to obtain an identity within a group of a cell; and execution of the instructions further causes the processor to: process the compensated SSS symbols to obtain a physical cell identity (PCI) group number; and obtain the cell ID based on the PCI group number and the identity within the group.

13. The baseband chip of claim 12, wherein execution of the instructions further causes the processor to: compensate SSS symbols for the plurality of antennas based on the final coarse FO estimation over a frequency domain to obtain the compensated SSS symbols; and process the compensated SSS symbols and the cell ID to obtain the SSS reference.

14. A baseband chip implementing frequency offset (FO) estimation at a receiver, comprising: a primary synchronization signal (PSS) processing circuit configured to: process PSS symbols based on a probability function, the PSS symbols being obtained through a plurality of antennas from a transmitter over a channel and associated with a time domain, and the probability function comprising a variable of FO and matrices indicating knowledge associated with at least one of the channel, the transmitter, or the receiver; and maximize the probability function according to the variable of FO to obtain a final coarse FO estimation for the plurality of antennas; a processor operatively coupled to the plurality of antennas; and memory storing instructions that, when executed by the processor, cause the processor to: process secondary synchronization signal (SSS) symbols and the final coarse FO estimation to obtain a final fine FO estimation based on the probability function, the probability function being configured to optimize detection rates of a transmitted PSS symbol and a transmitted SSS symbol at the transmitter.

15. The baseband chip of claim 14, wherein the matrices comprise at least one of a channel response, a length of channel response, a potential FO, bounds of the potential FO, a predicted or measured signal-to-noise ratio (SNR), a correlation score, or a potential cell comprising a cell identification (ID).

16. A method for frequency offset (FO) estimation implemented at a receiver, comprising: obtaining a PSS symbol through an antenna from a transmitter over a channel, the PSS symbol being associated with a time domain; inputting the PSS symbol to a probability function, the probability function comprising a variable of FO and matrices indicating knowledge associated with at least one of the channel, the transmitter, or the receiver; and maximizing the probability function according to the variable of FO to obtain a coarse FO estimation for the antenna, the probability function being configured to optimize a detection rate of a transmitted PSS symbol at the transmitter.

17. The method of claim 16, wherein the matrices comprise at least one of a channel response, a length of channel response, a potential FO, bounds of the potential FO, a predicted or measured signal-to-noise ratio (SNR), a correlation score, or a potential cell comprising a cell identification (ID).

18. The method of claim 16, wherein: the receiver comprises a plurality of antennas; and the method further comprises: obtaining coarse FO estimations for the plurality of antennas of the receiver; and combining the coarse FO estimations for the plurality of antennas to obtain a final coarse FO estimation.

19. The method of claim 18, further comprising: processing a secondary synchronization signal (SSS) symbol for a respective antenna based on a corresponding symbol timing and the final coarse FO estimation to obtain respective compensated SSS symbol; processing SSS symbols and compensated SSS symbols, for the plurality of antennas, to obtain SSS-related information, channel estimation for each antenna, and estimation of power of noise for each antenna, the SSS-related information comprising a cell identity (ID) and SSS reference; and inputting the SSS symbol corresponding to the respective antenna, the compensated SSS symbol corresponding to the respective antenna, the channel estimation corresponding to the respective antenna, the estimation of power of noise corresponding to the respective antenna, and the SSS-related information to the probability function to maximize the probability function according to the variable of FO for obtaining a fine FO estimation for each of the plurality of antennas, the probability function being configured to optimize a detection rate of a transmitted SSS symbol at the transmitter.

20. The method of claim 19, further comprising: combining fine FO estimations for the plurality of antennas to obtain a final fine FO estimation.