A multi-mode bluetooth digital transmitter and a control method thereof

Abstract

The present disclosure proposes a multi-mode Bluetooth digital transmitter and its control method. The transmitter comprises a coordinate rotation digital computer module, a multi-mode digital algorithm control module, a hybrid polar-quadrature processing module and an IQ multiplexing digital power amplifier array. The control method is as follows: converting baseband I / Q signals into amplitude and phase through CORDIC; selecting polar modulation or hybrid polar-quadrature mode according to modulation mode; decomposing phase into low-bandwidth component and residual component in polar-IQ mode, and remapping residual component into new I / Q code word through CORDIC; multiplexing I / Q code word into the same DPA array and applying range constraint; splitting into quadrant switching and residual phase when detecting large phase jump. The present disclosure realizes flexible compromise between low power consumption and high linearity through multi-mode switching and IQ multiplexing DPA, and effectively reduces phase path bandwidth requirement and DPA area.

Description

Technical Field

[0001] This disclosure belongs to the fields of Bluetooth communication, radio frequency integrated circuits and digital transmitter technology, and particularly relates to a multi-mode Bluetooth digital transmitter and its control method. Background Technology

[0002] Bluetooth systems, with their low cost, low power consumption, and high integration, have been widely used in headphones, wearable devices, IoT terminals, and short-range wireless connectivity devices. The transmitter is one of the modules in a Bluetooth chip that accounts for a relatively high proportion of power consumption and area; its linearity, efficiency, and digital configurability directly affect indicators such as transmit power consumption, error vector magnitude (EVM), spectral template, and adjacent channel leakage.

[0003] Traditional Bluetooth transmitters typically employ analog quadrature modulation, polar architecture, digital power amplifier (DPA), or hybrid architectures. For constant envelope modulation, such as BDR / GFSK modulation, the polar architecture is highly efficient because the amplitude path is essentially constant, with the modulation information primarily carried by the phase path. However, the polar architecture places high demands on phase path bandwidth, DCO modulation linearity, and the coordination between the two modulation points.

[0004] For non-constant envelope modulations such as EDR, the signal contains both amplitude and phase variations. If a polar architecture is used exclusively, delays, bandwidth, and gain errors between the amplitude and phase paths will worsen linearity. If an IQ architecture is used exclusively, a large IQ multiplexed digital power amplifier array and high dynamic range are required, increasing power consumption and area. Especially when digital PAs use finite bit quantization, multiplexing the I and Q paths into the same transmit array presents a trade-off between amplitude overflow, clip distortion, and array codeword utilization.

[0005] Therefore, a multi-mode Bluetooth digital transmitter is needed, which allows the same set of digital transmission hardware to switch between polar modulation high-efficiency mode, hybrid polarity-orthogonal high linearity mode, orthogonal multiplexing DPA mode, and phase transition assisted mode according to modulation method, linearity requirements, and power consumption targets, thereby achieving a better trade-off between small area, low power consumption, and high linearity.

[0006] It should be noted that the above introduction to the technical background is only for the purpose of providing a clear and complete explanation of the technical solutions of this application and facilitating understanding by those skilled in the art. It should not be assumed that these technical solutions are known to those skilled in the art simply because they have been described in the background section of this application. Summary of the Invention

[0007] To address at least one of the aforementioned problems, as well as one or more other potential problems, this disclosure aims to overcome the shortcomings of the prior art by proposing a multi-mode Bluetooth digital transmitter and its control method. This method uses a digital algorithm control module to uniformly configure the amplitude path, phase path, coordinate rotation digital computation (CORDIC) transformation, DPA codeword mapping, and DCO modulation scheme, enabling the transmitter to select the most suitable transmission path for BDR, EDR, and other operating modes with different linearity / power consumption requirements.

[0008] This disclosure further aims to reduce the area of ​​the IQ multiplexed digital power amplifier array by using IQ multiplexing DPA and codeword constraint algorithms, reduce the phase path bandwidth requirement by using hybrid polar-orthogonal (polar-IQ) decomposition, reduce the phase path jump amplitude by using 90-degree phase switching / quadrant compensation, and compensate for amplitude-phase mismatch by using digital calibration, thereby reducing power consumption and improving linearity.

[0009] In a first aspect of this disclosure, a multi-mode Bluetooth digital transmitter is proposed, comprising a baseband I / Q input port, a coordinate rotation digital computation (CORDIC) module, an amplitude path, a phase path, a digital power amplifier (DPA) array, and a phase-locked loop / digitally controlled oscillator (PLL / DCO) control interface. It further includes: a multi-mode digital algorithm control module, a hybrid polar-orthogonal (polar-IQ) processing module, an orthogonal-polar (IQ-polar) processing module, a selector, and an IQ multiplexed digital power amplifier (DPA) array; the coordinate rotation digital computation (CORDIC) module is used to process the input baseband complex signal s[n] = I[n] + jQ[n] is converted into amplitude signal A[n] and phase signal φ[n]; the multi-mode digital algorithm control module controls the selector to select the polar modulation transmission path and / or the hybrid polar-IQ transmission path according to the Bluetooth working mode and / or performance indicators; the hybrid polar-IQ processing module is used to decompose the phase signal φ[n] into a low-bandwidth phase component φ_p[n] and a residual phase component φ_r[n], where φ[n] = φ_p[n] + φ_r[n], and remap the amplitude signal A[n] and the residual phase component φ_r[n] to the updated orthogonal (I / Q) codeword: I_new[n] = A[n]·cos(φ_r[n]), Q_new[n] = A[n]·sin(φ_r[n]); The phase path processes the low-bandwidth phase component φ_p[n], and the updated orthogonal (I / Q) codeword is sent to the IQ multiplexed digital power amplifier (DPA) array; In the IQ multiplexed digital power amplifier (DPA) array, the I-way codeword and the Q-way codeword share the same physical unit array, and the digital domain applies codeword constraints to the I-way codeword and the Q-way codeword so that the total occupancy of the two codewords on the unit array does not exceed the maximum range of the array (i.e., the maximum codeword of the IQ multiplexed digital power amplifier array).

[0010] Furthermore, in some embodiments, the codeword constraint is: |C_I[n]| + |C_Q[n]| ≤ C_max; Where C_I[n] and C_Q[n] are the I-channel and Q-channel codewords respectively, and C_max is the maximum codeword of the IQ multiplexed digital power amplifier (DPA) array.

[0011] Furthermore, in some embodiments, when the codeword exceeds the maximum codeword C_max, proportional normalization is employed: α[n] = min(1, C_max / M[n]); C_I'[n] = α[n]·C_I[n]; C_Q'[n] = α[n]·C_Q[n]; Where M[n] is the Manhattan metric corresponding to the selected codeword constraint.

[0012] Furthermore, in some embodiments, in the IQ multiplexed digital power amplifier (DPA) array, the I-way codeword and the Q-way codeword are mapped to the same physical cell array from different directions through their respective cell-fill decoders; the digital domain scales, limits, or provides error feedback on the I / Q codewords before entering the cell enable decoding to ensure that the total occupancy of the two channels does not exceed the total number of array cells.

[0013] Furthermore, in some embodiments, the multi-mode Bluetooth digital transmitter further includes a limiting distortion suppression module, which uses a soft limiting function C'[n] = C_max·tanh(C[n] / C_max) or a piecewise linear compression function to replace hard limiting; or, the limiting distortion suppression module feeds back the limiting error e[n] = C[n]- C'[n] to subsequent sampling points after high-pass or noise shaping filtering.

[0014] Furthermore, in some embodiments, the multi-mode digital algorithm control module is also used to: when the phase increment |dφ[n]| of the phase path is detected to be greater than a preset threshold, enable the phase jump auxiliary mode, and split the phase increment into dφ[n] = k·π / 2 + dφ_p[n], where k is an integer quadrant compensation amount, and |dφ_p[n]| ≤ π / 2; the integer quadrant compensation amount j^k is implemented through I / Q swapping and sign flipping, and the phase path only processes the residual phase dφ_p[n].

[0015] In a second aspect of this disclosure, a control method for a multi-mode Bluetooth digital transmitter is also proposed, comprising the following steps: Step S1: Receive the baseband quadrature (I / Q) signal and convert the baseband quadrature (I / Q) signal into an amplitude signal A[n] and a phase signal φ[n] through coordinate rotation digital calculation (CORDIC); Step S2: Select the transmission mode based on the current Bluetooth modulation mode: If it is constant envelope modulation mode, then polar modulation mode is selected. The amplitude path outputs a constant or nearly constant amplitude codeword, and the phase path carries the phase modulation information. If it is a non-constant envelope modulation mode, then select the hybrid polar-orthogonal (polar-IQ) mode; Step S3: In the hybrid polar-IQ mode, the phase signal φ[n] is subjected to low-pass filtering and / or limiting to obtain a low-bandwidth phase component φ_p[n] and a residual phase component φ_r[n], where φ[n] = φ_p[n] + φ_r[n]. The low-bandwidth phase component φ_p[n] is then fed into the phase / digital controlled oscillator (DCO) path. Step S4: Remap the amplitude signal A[n] and the residual phase component φ_r[n] into updated orthogonal (I / Q) codewords I_new[n] and Q_new[n] using coordinate rotation digital computation (CORDIC) or lookup table; Step S5: Multiplex the updated orthogonal (I / Q) codewords to the same digital power amplifier (DPA) unit array, and apply range constraints to the I-way codewords and Q-way codewords so that the total occupancy of the unit array by the two codewords does not exceed the maximum range of the array (i.e., the maximum codeword C_max of the IQ multiplexed digital power amplifier array). Step S6: Monitor the phase increment |dφ[n]| in real time. When |dφ[n]| is greater than a preset threshold, perform a phase jump auxiliary operation: split the phase increment into integer multiple quadrant switching and residual phase. The integer multiple quadrant switching is achieved through I / Q swapping and sign flipping. The phase path only processes the residual phase.

[0016] Further, in some embodiments, in step S5, the range constraint is: |C_I[n]| + |C_Q[n]| ≤ C_max; when the codeword exceeds the maximum codeword C_max, proportional normalization α[n] = min(1, C_max / M[n]) is used to proportionally compress the I / Q codeword, where M[n] takes the Manhattan metric according to the selected constraint.

[0017] Furthermore, in some embodiments, step S5 further includes a step of suppressing clipping distortion: replacing hard clipping with a soft clipping function C'[n] = C_max·tanh(C[n] / C_max) or piecewise linear compression; and / or feeding back the clipping error e[n] = C[n]- C'[n] to subsequent sampling points after high-pass or noise shaping filtering, thereby pushing the quantization / clipping noise out of the effective modulation bandwidth.

[0018] Furthermore, in some embodiments, in step S2, the constant envelope modulation mode includes BDR / GFSK modulation; the non-constant envelope modulation mode includes EDR modulation; and in step S3, the bandwidth and / or limiting threshold of the low-pass filter are configurable parameters, which are dynamically adjusted by the multi-mode digital algorithm control module according to the phase path bandwidth capability, error vector magnitude (EVM) index, and DPA range, to achieve a trade-off between the phase path bandwidth requirement and the dynamic range of the I / Q path.

[0019] Furthermore, in some embodiments, the preset threshold in step S6 is π / 2; the phase jump auxiliary operation is not limited to 90-degree quadrant switching, but can be extended to 180-degree, 270-degree or other arbitrary angle quadrant switching.

[0020] Furthermore, in some embodiments, when the polar modulation mode is selected in step S2 and the phase path bandwidth is insufficient, a two-point modulation method is adopted: the low-frequency phase information is sent to the PLL (Phase-Locked Loop) / DCO (Digital Controlled Oscillator) control terminal, and the high-frequency phase information is sent to the modulation compensation terminal, so that the modulation bandwidth of the original phase signal is covered by the synthesis of the low-frequency phase information and the high-frequency phase information, that is, H_low(z) + H_high(z)≈1.

[0021] This disclosure has the following advantages over the prior art: In some embodiments, the same Bluetooth digital transmitter can switch between multiple modes according to modulation scheme and performance requirements, achieving a flexible trade-off between low power consumption and high linearity. Additionally, in some embodiments, the number of transmitter units is reduced by using an I / Q multiplexed digital power amplifier (DPA) array, significantly reducing chip area and static / dynamic power consumption. In some embodiments, codeword normalization, amplitude limiting, or amplitude limiting mapping algorithms ensure that the sum of the absolute values ​​of the I and Q paths (i.e., |CI|+|CQ|) does not exceed the maximum range of the DPA, preventing array overflow. In some embodiments, for constant envelope modulation such as BDR, a polarity modulation mode can be used to keep the amplitude path essentially constant, with modulation primarily achieved through the phase path, improving transmission efficiency. In some embodiments, for non-constant envelope modulation such as EDR, a hybrid polarity-orthogonal mode can be used to transfer high-frequency or residual phase information to the amplitude / IQ path, reducing phase path bandwidth and DCO modulation pressure. In some embodiments, 90-degree phase jump detection and I / Q switching compensation break down large phase jumps into quadrant switching and smaller residual phase modulation, reducing the transient amplitude of the phase path. Attached Figure Description

[0022] The above and other features, advantages and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description, wherein: Figure 1 A schematic diagram of the main frame of a multi-mode Bluetooth digital transmitter according to an embodiment of the present disclosure is shown; Figure 2 A schematic diagram of the algorithm signal chain of a multi-mode Bluetooth digital transmitter according to an embodiment of the present disclosure is shown; Figure 3 A schematic diagram of the range management and coding structure of an I / Q multiplexing DPA array according to an embodiment of the present disclosure is shown; Figure 4 A schematic diagram of a 90-degree phase jump auxiliary mode according to an embodiment of the present disclosure is shown; Figure 5 A flowchart illustrating a control method for a multi-mode Bluetooth digital transmitter according to an embodiment of the present disclosure is shown. And in each of the accompanying figures, the same or corresponding reference numerals indicate the same or corresponding parts. Detailed Implementation

[0023] Embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the accompanying drawings and embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of protection of the present disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0024] In the description of embodiments of this disclosure, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment" or "this embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc., may refer to different or the same objects. Other explicit and implicit definitions may also be included below. It should also be understood that the term "and / or" as used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0025] The following detailed explanation is provided in conjunction with the accompanying drawings. It should be understood that in this disclosure, the term "codeword" specifically refers to the digital control signal input to the digital power amplifier (DPA) array, which can be simply understood as a "digital amplitude control instruction." In other words, the codeword is the digital switching quantity instruction (0~1023) sent to the DPA. In this disclosure, it needs to undergo "multiplexing constraint" and "amplification limiting" algorithms to ensure that the sum of the I-channel and Q-channel codewords does not exceed the carrying capacity of the physical hardware's total of 1023 units.

[0026] Regarding the multi-mode Bluetooth digital transmitter architecture. It should be understood that, in some embodiments, reference is made to... Figure 1 and combined Figure 2 The multi-mode Bluetooth digital transmitter disclosed herein includes: a baseband I / Q input port, a first coordinate rotation digital calculation (CORDIC) module, an amplitude path, a phase path, a multi-mode digital algorithm control module, a hybrid polar-orthogonal (polar-IQ) processing module, an orthogonal-polar (IQ-polar) processing module, a selector, a polar digital front-end (polar DFE), a Bluetooth digital-to-analog converter (DAC) interface, a DCO control interface, and an IQ multiplexed DPA array. The baseband complex signal s[n] = I[n] + jQ[n] is transformed by the first coordinate rotation digital calculation (CORDIC) module to obtain the amplitude signal A[n] and the phase signal φ[n], which are then sent to the amplitude path and the phase path, respectively. The multi-mode digital algorithm control module controls the selector to select the appropriate transmission path based on the current Bluetooth operating mode (BDR / EDR, etc.), power consumption requirements, and linearity indicators. The implementation of the IQ multiplexed DPA array is discussed below.

[0027] Regarding the polar modulation (polar) operating mode: In some embodiments, when Bluetooth operates in BDR mode, the multi-mode digital algorithm control module selects the polar modulation operating mode. For approximate constant envelope modulation such as BDR / GFSK, the amplitude signal A[n] is approximately a constant A0, the amplitude path outputs a stable codeword, and the phase path controls the DCO according to the phase increment dφ[n]. The transmitted signal is: s_BDR[n] ≈ A0exp(jφ[n]); If the phase path bandwidth is insufficient to carry all phase modulation information, a two-point modulation method can be adopted: the low-frequency phase information is sent through the PLL / DCO control terminal, and the high-frequency phase information is sent through the modulation compensation terminal, so that H_low(z) + H_high(z) ≈ 1. Since the original phase signal is an all-pass response, its transfer function is 1. Therefore, the goal of two-point modulation is to make the sum of the transfer functions of the two paths approximately 1, so as to cover the complete modulation bandwidth of the original phase signal by synthesizing the low-frequency and high-frequency information, thereby achieving high-quality modulation under limited phase path bandwidth.

[0028] Regarding the hybrid polar-orthogonal (polar-IQ) operating mode. In some embodiments, when Bluetooth operates in EDR mode, the multi-mode digital algorithm control module selects the hybrid polar-orthogonal operating mode. The phase signal φ[n] is first processed by a low-pass filter (LPF) or limiter to obtain the low-bandwidth phase φ_p[n] and the residual phase φ_r[n]: φ[n] = φ_p[n] + φ_r[n]; φ_p[n] = LPF{φ[n]} or φ_p[n] = Limiter{φ[n]}; φ_r[n] = φ[n] - φ_p[n]; The phase path only processes φ_p[n], while the residual phase φ_r[n] and amplitude A[n] are remapped together by the second coordinate rotation digital computation (CORDIC) module into a new I / Q codeword: I_new[n] = A[n]cos(φ_r[n]); Q_new[n] = A[n]sin(φ_r[n]); The phase path bandwidth is a configurable parameter. The control module dynamically adjusts the low-pass filter bandwidth or limiting threshold based on the phase path capability, EVM index, and DPA range, achieving an optimal trade-off between phase path bandwidth requirements and I / Q path dynamic range.

[0029] It should be understood that Figure 2 This illustration shows a digital transmitter signal processing link based on a polar coordinate architecture (commonly used in Bluetooth or low-power wireless communications) according to an embodiment of this disclosure. Its core utilizes the CORDIC algorithm for coordinate transformation and phase processing. The link primarily implements the process of converting the baseband I / Q signal into polar coordinate form, performing phase correction preprocessing, and then splitting it into an amplitude path and a frequency / phase path, which are then sent to a power amplifier and an oscillator, respectively.

[0030] Step 1: I / Q to Polar Coordinates. The input signal IQ enters the first coordinate rotation digital computation module (CORDIC), which converts it from the Cartesian coordinate system (I / Q) to the polar coordinate system, outputting the amplitude AM and phase change dphase. The amplitude and phase signals are then split and processed.

[0031] Step 2: Phase Correction and Multiplexing. The original phase dphase branch enters two coordinate transformation modules: the hybrid polar-orthogonal processing module (polar-IQ) and the quadrature-polarity processing module (IQ-polar). These two modules remap the phase, outputting two signals which are then selected by a multiplexer (MUX). The multiplexer is controlled by the configuration / mode selection bit (Zcp=00 or 01), selecting one signal as dphase_ift (the processed phase) according to the configuration. Next, this processed phase dphase_ift is summed with the original phase dphase at the subtraction node (circled with a + sign). (It should be understood that the right-hand minus sign ("-") in the diagram represents the input data on the right being inverted and incremented by one before entering the adder (to obtain two's complement), or the sign bit of the input is directly set to negative, resulting in a new phase error / correction value. It should also be understood that this is equivalent to subtracting dphase_ift from dphase, thus subtracting the low-pass portion, and then multiplying the remaining signal into the IQ path.

[0032] Step 3: Update the I / Q signals. The original amplitude AM, along with the new phase error dphase_ift output from the previous step, enters the second coordinate rotation digital calculation module (CORDIC) to perform inverse coordinate rotation and resynthesize the updated in-phase quadrature signal iq_new.

[0033] Step 4: Polar Digital Front-End (Polar DFE) Processing. The updated signal enters the Polar Digital Front-End (Polar DFE) module for final digital front-end processing, such as waveform shaping, oversampling, and predistortion.

[0034] Step 5: Final Output Separation. After processing by the Polar Digital Front-End Module (Polar DFE), the signal is split into two paths: Amplitude Path: Output to the Bluetooth Digital-to-Analog Converter (Bt_dac_intf) interface, where the signal is ultimately fed into a Digital Power Amplifier (DPA) for amplification and transmission of the wireless RF signal. Frequency / Phase Path: Output the corresponding Digital Controlled Oscillator (DCO) Code, which is fed into a Digital Controlled Oscillator (DCO) to control the frequency or phase synthesis of the RF carrier.

[0035] According to the above embodiments, a typical polar coordinate transmitter with error correction / predistortion loop utilizes a coordinate rotation digital computing module (CORDIC) to perform efficient conversion between the I / Q and Polar domains, thereby achieving better power consumption and performance, which is crucial for power-sensitive wireless communications such as Bluetooth.

[0036] Regarding IQ multiplexing DPA arrays, see [link / reference]. Figure 3As shown, this is an ultra-low power, highly digitized direct radio frequency (RF) transmission architecture based on a DPA (digital power amplifier) ​​array, which directly converts I / Q baseband data into high-power RF signals through pure digital logic.

[0037] Specifically, this system addresses the problem of how to handle dynamically changing I / Q amplitude signals with limited physical hardware resources (1023 units). Input constraints include: the physical array provides C_max = 1023 DPA units. The instantaneous total demand is calculated as: M[n] = |C_I[n]| + |C_Q[n]|, which is the sum of the absolute values ​​of the I and Q amplitudes at the current moment. The scaling factor is calculated as: α[n] = min(1, C_max / M[n]). If the current total I and Q amplitudes are less than or equal to 1023 (not exceeding the hardware limit), the scaling factor α[n] is 1, with no adjustment. If the current total amplitude exceeds 1023, the system automatically calculates a scaling factor α[n] less than 1 for amplitude limiting (normalization). The final output signals to the hardware are: C_I'[n] = α[n]C_I[n], C_Q'[n] = α[n]C_Q[n]. It should be understood that in the illustrated embodiment, it is guaranteed that at any given time, the sum of the absolute values ​​of I and Q fed into the hardware array below will not exceed 1023. This "absolute value summation limiting" instead of "root mean square summation limiting" greatly reduces computational complexity and is very suitable for rapid implementation in pure digital hardware.

[0038] Further, according to the data flow. Input end (left): Input the I-way codeword (C_I[n]) and Q-way codeword (C_Q[n]) of the digital baseband. Simultaneously input the mode control (BDR / EDR). This indicates that the architecture is applied to the Bluetooth standard. BDR (basic rate, 1Mbps) and EDR (enhanced data rate, 2Mbps / 3Mbps) have different requirements for I / Q mapping under GFSK (Gaussian frequency shift keying, the specific modulation technique for implementing BDR mode) and DPSK (differential quadrature phase shift keying) modulation. The range management module will adjust its internal strategy according to the mode. Further, regarding the digital range management module: it handles the above-mentioned "normalization / limiting" mathematical operations. The figure marks "error feedback", which means that after the limiting causes signal distortion, the system may feed back the truncation error to the next sampling period for compensation to improve the EVM (error vector amplitude) quality of the transmitted signal. Furthermore, regarding the fill decoder module: the scaled C_I' and C_Q' respectively enter the "I / Q to cell-fill fill decoder". It should be understood that the decoder's function is to convert the amplitude value (e.g., calculating that 200 units are needed) into control switching instructions for the physical array. Further, regarding the physical execution layer: shared 1023-cell physical DPA array (middle): This is a dynamically allocated DPA unit array. The diagram visually illustrates that the I-channel signal "fills" to the right, activating a portion of the units, while the Q-channel signal "fills" to the left, activating another portion. Physically, these 1023 DPA units are shared by I and Q, dynamically allocated according to the strength of the two signals at the current moment. Further, regarding the RF output (right side): RF synthesis and matching network: synthesizes the output current / voltage of the 1023 tiny DPA units in the power domain and connects it to the antenna matching network. Finally, the RF signal is output and directly transmitted.

[0039] It should be understood that the illustrated embodiment enables: 1) Extreme digitization: It eliminates the traditional transmitter's DAC and up-conversion mixer, completing everything in the digital domain and RF switch array, making it highly compatible with advanced CMOS processes, resulting in a small chip area and low cost. 2) High energy efficiency: The DPA array acts like a giant one-bit digital-to-analog converter, only powering the units that need to be turned on, resulting in extremely high efficiency, making it ideal for battery-powered Bluetooth wearable devices. 3) Intelligent clipping ensures linearity: Through the simple 1023 full-scale normalization constraint, it prevents the signal from exceeding the peak output capability of the hardware array, thereby avoiding severe clipping distortion.

[0040] Furthermore, in some alternative embodiments, the IQ multiplexed DPA array comprises 1023 cells (corresponding to 10-bit codewords). The I and Q paths share the same physical array, rather than being configured with separate DPA arrays. Accordingly, combined with Figure 3Full-scale constraints in the digital domain: Cmax = 1023 cells; M[n] = |C_I[n]| + |C_Q[n]|, α[n] = min(1, Cmax / M[n]); C_I'[n] = α[n]C_I[n], C_Q'[n] = α[n]C_Q[n], |C_I'[n]| + |C_Q'[n]| ≤ 1023.

[0041] It should be understood that the maximum codeword for DPA is C_max = 1023. The control algorithm employs one of the following constraints: |C_I[n]| + |C_Q[n]| ≤ 1023.

[0042] Furthermore, for inputs exceeding the range, proportional normalization is employed: α[n] = min(1, 1023 / M[n]); C_I'[n] = α[n]·C_I[n]; C_Q'[n] = α[n]·C_Q[n]; Where M[n] takes L1 (Manhattan metric) according to the selected constraints.

[0043] It should be understood that the I-path codeword and Q-path codeword are mapped to the physical DPA array through their respective cell-filling decoders. The I-path allocates cells into the array from one direction, and the Q-path allocates cells into the array from the other direction. The digital algorithm ensures that the total cell occupancy of the two paths does not exceed 1023, avoiding cell conflicts or output distortion.

[0044] It should also be understood that, in order to reduce the in-band distortion caused by clipping, error feedback can be introduced: e[n] = C[n] - C'[n] is fed back to subsequent sampling points after passing through a high-pass or noise-shaping filter, so that the main quantization / clipping error is far away from the effective modulation bandwidth.

[0045] Regarding the 90-degree phase jump assist mode. In some embodiments, see [link to documentation]. Figure 4 As shown, when the phase increment |dφ[n]| is detected to be greater than the preset threshold π / 2, the control module splits the phase into: dφ[n] = k·π / 2 + dφ_p[n]; Where k is an integer, |dφ_p[n]| ≤ π / 2. Using exp(j·dφ[n]) = j^k·exp(j·dφ_p[n]), j^k is achieved through I / Q swapping and sign flipping (multiplying by j is equivalent to I' = -Q, Q' = I). The phase path only needs to handle the small residual phase dφ_p[n], while the larger 90-degree jump is achieved by I / Q switching, thereby reducing the transient amplitude requirements of the phase path.

[0046] Specifically, based on embodiments of this disclosure, see [link to relevant documentation]. Figure 4 As shown, a phase rotation optimization architecture for digital signal processing is proposed, specifically called the "90-degree phase jump auxiliary mode," the illustration of which is equivalent to... Figure 2 The phase processing section. Its core purpose is to reduce the complexity of hardware implementation, especially to reduce the operating range and accuracy requirements of phase modulation modules (such as DCO, digitally controlled oscillator).

[0047] Specifically, regarding the phase decomposition, the system decomposes the input "phase increment" Δφ[n] into two parts: 1) the large angle part (integer multiples of 90 degrees): k·π / 2; 2) the residual small angle part: Δφ_p[n]; where Δφ[n]=k·π / 2+Δφ_p[n], where k takes values ​​of {-2, -1, 0, 1, 2}, and the residual phase |Δφ_p[n]|≤π / 2. The corresponding complex rotation formula is: exp(jΔφ[n])=j^k·exp(jΔφ_p[n]).

[0048] Regarding the dual-path processing mechanism shown in the diagram (i.e.) Figure 4 (The two branches, upper and lower).

[0049] Regarding the upper branch (handling large angle integer multiples): When the phase rotation is 90 degrees, 180 degrees, 270 degrees, or −90 degrees, complex multipliers or phase accumulators are not needed. It is achieved through "I / Q swapping and sign compensation." This can be accomplished in hardware with simple wire swapping and sign inversion. Specifically, as shown in the example in the figure: multiplying by j (i.e., rotating 90 degrees) is equivalent to the real part I′ = −Q and the imaginary part Q′ = I.

[0050] Regarding the lower branch (handling residual small angles): Only small angles with |Δφ_p[n]| ≤ π / 2 are handled by the "phase / DCO modulation path" for fine processing. Since this angle range is strictly limited to ±90 degrees, this greatly reduces the dynamic range requirements of the DCO or phase modulator, thereby simplifying circuit design and reducing power consumption.

[0051] In summary, the final output is: the result j^k generated by the upper path and the result exp(jΔφ_p) generated by the lower path are finally combined to complete the "equivalent phase rotation output signal".

[0052] It should be understood that in traditional digital communication or radar systems, directly performing large-scale phase rotations typically requires high-precision lookup tables (LUTs) or complex complex multipliers (CORDIC algorithms). However, by adopting this "90-degree phase-jump assisted mode": The circuit is simplified: large-angle rotations are transformed into basic I / Q path switching and sign flipping. Accuracy requirements are reduced: the phase modulation path only needs to process signals within a ±90-degree range, significantly lowering the hardware implementation threshold and cost. Energy efficiency is improved: because a large number of complex multiplication operations are avoided, this system is very suitable for low-power wireless communication chips (such as Bluetooth, Wi-Fi, and IoT devices).

[0053] It should also be understood that this auxiliary mode is not limited to 90 degrees, but can also be extended to 180 degrees, 270 degrees or other quadrant / multi-phase switching.

[0054] In some alternative embodiments, a schematic diagram of a digital RF transmitter control architecture supporting multiple modes (polar coordinate / IQ hybrid) is also proposed for application in the transmission link of Bluetooth (Bluetooth BR / EDR / LE) wireless communication chips. This will be elaborated upon in detail below.

[0055] Regarding the input: Baseband IQ: Baseband digital signal, the input is a data stream in phase (I) and quadrature (Q).

[0056] Regarding signal distribution and control: Mode control (BDR / EDR / test): This module generates select and switch signals based on the communication protocol requirements (such as Bluetooth Basic Rate (BDR), Enhanced Rate (EDR), or Test mode). This determines the processing path for subsequent signals. CORDIC A phi: This module converts the input I / Q signals into polar coordinates and outputs amplitude A and phase phi.

[0057] This involves three parallel signal processing paths. Based on mode control, the system sends the signal to one of three different processing paths: Polar mode (A + phase): Employs a fully polar coordinate architecture. It receives the amplitude (A) and phase (phi) output from the CORDIC sensor and uses them directly for subsequent control. Polar-IQ mode (LPF + IQ residual): This mode uses a low-pass filter (LPF) and processes the IQ residual signal. It is a hybrid processing method that balances energy consumption and performance. IQ 90-degree switch mode: Modulates the IQ signal by switching the phase by 90 degrees (commonly used in EDR modulation methods such as π / 4-DQPSK or 8DPSK).

[0058] Regarding the output and RF driver stage: IQ multiplex 10-bit DPA (IQ Multiplexed 10-bit Digital Power Amplifier): Receives signals from the aforementioned paths. The 10-bit designation indicates high-precision amplitude / phase control; it's a digital power amplifier module integrating DAC functionality. DCO / phase path (Digital Controlled Oscillator / Phase Path): Receives signals from the polar coordinate path or IQ switching path, used to directly modulate the carrier phase and generate the RF frequency. RF output: The amplitude / power signal output from the DPA and the phase / frequency signal output from the DCO are combined here to ultimately generate the RF signal for antenna transmission.

[0059] In summary, compared to traditional IQ modulation, the polar coordinate architecture of the above embodiments directly modulates the phase using the DCO and controls the amplitude using the DPA. This architecture boasts extremely low power consumption, making it highly suitable for Bluetooth Low Energy (BLE) applications. Furthermore, the multi-path switching diagram illustrates the chip's high compatibility, enabling it to efficiently run in polar coordinate mode while also being compatible with older IQ modulation standards (such as the traditional Bluetooth BR / EDR).

[0060] Regarding the multi-mode control procedure, in some embodiments, the multi-mode digital algorithm control module configures the transmission path according to the following procedure: Detect the current Bluetooth operating mode (BDR / EDR, etc.); If it is BDR or other constant envelope mode, polarity modulation mode is preferred; For EDR or other non-constant envelope modes with high linearity requirements, select polar modulation mode with two-point modulation and real-time digital calibration according to the phase path bandwidth capability and power consumption requirements, or select hybrid polar-orthogonal mode. Real-time monitoring of phase transitions; when |dφ[n]| exceeds the threshold, 90-degree I / Q switching compensation is enabled. Real-time monitoring of DPA codeword utilization; dynamic adjustment of phase path bandwidth, IQ normalization coefficient, or limiting strategy when approaching the range. Real-time digital calibration compensates for DPA amplitude-amplitude (AM-AM) / amplitude-phase (AM-PM) distortion, DCO gain, I / Q mismatch, path delay, and amplitude and phase errors between polarity / IQ paths.

[0061] Further, refer to Figure 5 As shown, this disclosure also proposes a control method for a multi-mode Bluetooth digital transmitter, which includes the following steps: Step S1: Receive a baseband quadrature (I / Q) signal, and convert the baseband quadrature (I / Q) signal into an amplitude signal A[n] and a phase signal φ[n] through coordinate rotation digital calculation (CORDIC); Step S2: Select the transmission mode according to the current Bluetooth modulation mode: if it is a constant envelope modulation mode, select the polar modulation mode, the amplitude path outputs a constant or approximately constant amplitude codeword, and the phase path carries the phase modulation information; if it is a non-constant envelope modulation mode, select the hybrid polar-IQ mode; Step S3: In the hybrid polar-IQ mode, perform low-pass filtering and / or amplitude limiting processing on the phase signal φ[n] to obtain a low-bandwidth phase component φ_p[n] and a residual phase component φ_r[n], where φ[n] = φ_p[n] + Step S4: The amplitude signal A[n] and the residual phase component φ_r[n] are remapped into the updated orthogonal (I / Q) codewords I_new[n] and Q_new[n] through coordinate rotation digital calculation (CORDIC) or lookup table; Step S5: The updated orthogonal (I / Q) codewords are multiplexed into the same digital power amplifier (DPA) unit array, and range constraints are applied to the I-channel codewords and Q-channel codewords so that the total occupancy of the two codewords on the unit array does not exceed the maximum range of the array. It should be understood that "maximum range of the array" is the physical expression of "maximum codeword C_max of IQ multiplexed DPA array"; C_max is its name in the digital domain. Both values ​​are the same, and together they define the core constraint that the sum of the two codewords in the I / Q shared array must not exceed 1023; Step S6: Real-time monitoring of phase increment |dφ[n]|, when |dφ[n]| is greater than the preset threshold, perform phase jump auxiliary operation: split the phase increment into integer multiple quadrant switching and residual phase, and realize the integer multiple quadrant switching through I / Q swapping and sign flipping, and the phase path only processes the residual phase.

[0062] It should also be understood that in some alternative embodiments, the IQ multiplexed DPA array can be 10 bits or extended to other bit numbers; codeword constraints can employ absolute values ​​and constraints. Alternatively, phase processing in the hybrid polarity-orthogonal mode can employ low-pass filtering, bandwidth limiting, slope limiting, amplitude limiting, lookup table mapping, or adaptive filtering. Alternatively, residual phase compensation can be performed by coordinate rotation digital computation (CORDIC), or by lookup tables, approximate trigonometric functions, polynomials, or rotation matrices. Alternatively, the 90-degree phase jump auxiliary mode is not limited to 90 degrees and can be extended to 180 degrees, 270 degrees, or other quadrant / multi-phase switching. Alternatively, real-time digital calibration can calibrate DPA amplitude-amplitude / amplitude-phase distortion, DCO gain, I / Q mismatch, path delay, and amplitude-phase error between polarity / IQ paths.

[0063] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or technical improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

[0064] The above description is merely an optional embodiment of this disclosure and is not intended to limit this disclosure. Various modifications and variations can be made to this disclosure by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A multi-mode Bluetooth digital transmitter, comprising a baseband I / Q input port, a coordinate rotation digital calculation module, an amplitude path, a phase path, a digital power amplifier array, and a phase-locked loop / digital controlled oscillator control interface, characterized in that, Also includes: Multi-mode digital algorithm control module, hybrid polarity-orthogonal processing module, orthogonal-polarity processing module, selector, and IQ multiplexed digital power amplifier array; The coordinate rotation digital calculation module is used to convert the input baseband complex signal s[n] = I[n] + jQ[n] into an amplitude signal A[n] and a phase signal φ[n]. The multi-mode digital algorithm control module controls the selector to select the polarity modulation transmission path and / or the mixed polarity-orthogonal transmission path according to the Bluetooth working mode and / or performance indicators. The hybrid polarity-orthogonal processing module is used to decompose the phase signal φ[n] into a low-bandwidth phase component φ_p[n] and a residual phase component φ_r[n], where φ[n] = φ_p[n] + φ_r[n], and remap the amplitude signal A[n] and the residual phase component φ_r[n] into updated I / Q codewords: I_new[n] = A[n]·cos(φ_r[n]), Q_new[n] = A[n]·sin(φ_r[n]); The phase path processes the low-bandwidth phase component φ_p[n], and the updated I / Q codeword is fed into the IQ multiplexed digital power amplifier array. In the IQ multiplexed digital power amplifier array, the I-way codeword and the Q-way codeword share the same physical unit array. The digital domain applies codeword constraints to the I-way codeword and the Q-way codeword so that the total occupancy of the unit array by the two codewords does not exceed the maximum codeword of the IQ multiplexed digital power amplifier array.

2. The multi-mode Bluetooth digital transmitter according to claim 1, characterized in that, The codeword constraint is as follows: |C_I[n]| + |C_Q[n]| ≤ C_max; Where C_I[n] and C_Q[n] are the I-channel codeword and Q-channel codeword, respectively, and C_max is the maximum codeword of the IQ multiplexed digital power amplifier array.

3. The multi-mode Bluetooth digital transmitter according to claim 2, characterized in that, When the codeword exceeds the maximum codeword C_max, proportional normalization is used: α[n] = min(1, C_max / M[n]); C_I'[n] = α[n]·C_I[n]; C_Q'[n] = α[n]·C_Q[n]; Where M[n] is the Manhattan metric corresponding to the selected codeword constraint.

4. The multi-mode Bluetooth digital transmitter according to claim 1, characterized in that, In the IQ multiplexed digital power amplifier array, the I-way codeword and the Q-way codeword are mapped to the same physical cell array from different directions through their respective cell-filling decoders. Before entering the cell to enable decoding, the digital field scales, limits, or provides error feedback on the I / Q codewords to ensure that the total occupancy of the two paths does not exceed the total number of array cells.

5. The multi-mode Bluetooth digital transmitter according to claim 1, characterized in that, It also includes a limiting distortion suppression module, which uses a soft limiting function C'[n] = C_max·tanh(C[n] / C_max) or a piecewise linear compression function to replace hard limiting; or, the limiting distortion suppression module feeds back the limiting error e[n] = C[n] - C'[n] to subsequent sampling points after high-pass or noise shaping filtering.

6. The multi-mode Bluetooth digital transmitter according to claim 1, characterized in that, The multi-mode digital algorithm control module is also used for: When the phase increment |dφ[n]| of the phase path is detected to be greater than the preset threshold, the phase jump auxiliary mode is activated, and the phase increment is split into dφ[n] = k·π / 2 + dφ_p[n], where k is the integer quadrant compensation amount, and |dφ_p[n]|≤ π / 2; The integer quadrant compensation quantity j^k is implemented through I / Q swapping and sign flipping, and the phase path only processes the residual phase dφ_p[n].

7. A control method for a multi-mode Bluetooth digital transmitter as described in any one of claims 1 to 6, characterized in that, Includes the following steps: Step S1: Receive the baseband I / Q signal, and convert the baseband I / Q signal into an amplitude signal A[n] and a phase signal φ[n] through coordinate rotation digital calculation; Step S2: Select the transmission mode based on the current Bluetooth modulation mode: If it is constant envelope modulation mode, then polarity modulation mode is selected, the amplitude path outputs constant or nearly constant amplitude codeword, and the phase path carries phase modulation information; If it is a non-constant envelope modulation mode, then select the hybrid polarity-orthogonal mode; Step S3: In the hybrid polarity-orthogonal mode, the phase signal φ[n] is subjected to low-pass filtering and / or limiting to obtain a low-bandwidth phase component φ_p[n] and a residual phase component φ_r[n], where φ[n] = φ_p[n] + φ_r[n]. The low-bandwidth phase component φ_p[n] is then fed into the phase / digital control oscillator path. Step S4: Remap the amplitude signal A[n] and the residual phase component φ_r[n] to the updated I / Q codewords I_new[n] and Q_new[n] through coordinate rotation digital calculation or lookup table; Step S5: Multiplex the updated I / Q codewords to the same digital power amplifier unit array, and apply range constraints to the I-channel codewords and Q-channel codewords so that the total occupancy of the unit array by the two codewords does not exceed the maximum codeword C_max of the IQ multiplexed digital power amplifier array; Step S6: Monitor the phase increment |dφ[n]| in real time. When |dφ[n]| is greater than a preset threshold, perform a phase jump auxiliary operation: split the phase increment into integer multiple quadrant switching and residual phase. The integer multiple quadrant switching is achieved through I / Q swapping and sign flipping. The phase path only processes the residual phase.

8. The control method for a multi-mode Bluetooth digital transmitter according to claim 7, characterized in that, In step S5, the range constraint is: |C_I[n]| + |C_Q[n]| ≤ C_max; when the codeword exceeds the maximum codeword C_max, proportional normalization α[n] = min(1, C_max / M[n]) is used to compress the I / Q codeword proportionally, where M[n] takes the Manhattan metric according to the selected constraint.

9. The control method for a multi-mode Bluetooth digital transmitter according to claim 7, characterized in that, Step S5 further includes a step to suppress limiting distortion: replacing hard limiting with a soft limiting function C'[n] = C_max·tanh(C[n] / C_max) or piecewise linear compression; and / or feeding back the limiting error e[n] = C[n] - C'[n] to subsequent sampling points after high-pass or noise shaping filtering, thereby pushing the quantization / limiting noise out of the effective modulation bandwidth.

10. The control method for a multi-mode Bluetooth digital transmitter according to claim 7, characterized in that, In step S2, the constant envelope modulation mode includes BDR / GFSK modulation; the non-constant envelope modulation mode includes EDR modulation; in step S3, the bandwidth and / or limiting threshold of the low-pass filter are configurable parameters, which are dynamically adjusted by the multi-mode digital algorithm control module according to the phase path bandwidth capability, error vector amplitude index and DPA range, to achieve a trade-off between the phase path bandwidth requirement and the dynamic range of the I / Q path.

11. The control method for a multi-mode Bluetooth digital transmitter according to claim 7, characterized in that, The preset threshold in step S6 is π / 2; the phase jump auxiliary operation is not limited to 90-degree quadrant switching, but can be extended to 180-degree, 270-degree or other arbitrary angle quadrant switching.

12. The control method for the multi-mode Bluetooth digital transmitter according to claim 7, characterized in that, When the polarity modulation mode is selected in step S2 and the phase path bandwidth is insufficient, a two-point modulation method is adopted: the low-frequency phase information is sent to the phase-locked loop control terminal or the digitally controlled oscillator control terminal, and the high-frequency phase information is sent to the modulation compensation terminal, so as to cover the modulation bandwidth of the original phase signal by the synthesis of the low-frequency phase information and the high-frequency phase information.

Images